The Regulatory Potential of Marine

Cyanobacteria: Transcriptional Factors and

Small RNAs
studied in a comparative genomics approach

DISSERTATION
zur Erlangung des akademischen Grades
doctor rerum naturalium
(Dr. rer. nat.)
im Fach Biologie

eingereicht an der
Mathematisch-Naturwissenschaftlichen Fakultät I
Humboldt-Universität zu Berlin

von
Frau Diplom-Ingenieurin Ilka Maria Axmann
geboren am 21.12.1976 in Räckelwitz

Präsident der Humboldt-Universität zu Berlin:

Prof. Dr. Christoph Markschies
Dekan der Mathematisch-Naturwissenschaftlichen Fakultät I:
Prof. Thomas Buckhout, PhD

Gutachter:
1. Prof. Dr. Thomas Börner
2. Prof. Dr. Karl Forchhammer
3. Dr. David J. Scanlan

Tag der mündlichen Prüfung: 17. Juli 2006

 Abstract

Life on Earth is driven by the power of oxygenic photosynthesis transforming solar into chemical
energy. Cyanobacteria such as Prochlorococcus and Synechococcus belong to the most important
primary producers within the oceans and increasingly serve as models for photosynthetic organisms.
To better understand the regulatory mechanisms in these picocyanobacteria, here the information
from four genomes of closely related and even so ecologically divergent marine strains was used in
a combined computational and experimental approach. Sequence signals and RNA-coding genes
as novel elements in the regulation of gene expression were identified and their distribution along
the phylogenetic gradient compared. Phylogenetic footprinting revealed a minimal conserved set
of putative transcription factors, their binding sites and regulons. Sites for NtcA, LexA and ArsR-
like regulators were found as well as new cis elements. RACE experiments verified several of these
predicted sites belonging to the promoter region. The cyanobacterial core promoter shows similar
features as known from E. coli. In addition, sequence elements at position -60 were frequently
found associated with putative bidirectional promoters. For cyanophage P-SSP7, promoter stud-
ies revealed a conserved site, which might be recognised by the phage RNA polymerase or/and
the host’s RNA processing machinery. A search, focussing on conserved secondary structures,
detected several non-coding RNAs named Yfr for cYanobacterial Functional RNA. One of these
ncRNAs, Yfr1, appears dispensable for growth at greater depths. A highly conserved secondary
structure and an unpaired CA dinucleotide repeat might represent essential functional elements of
Yfr1. A comparative analysis of Yfr7 structures, transcript types and accumulation throughout
the cyanobacterial radiation indicated this RNA as the likely homologue of the E. coli 6S RNA.
Two distinct Yfr7 transcripts with a circadian but time-shifted expression pattern suggested a
coupling of their expression to the circadian rhythm or light intensity. Experiments in Synechocys-
tis discovered a novel antisense RNA-mediated regulatory mechanism that controls isiA mRNA
abundance and assembly of IsiA-photosystem I supercomplexes. Functional assignments of these
new elements in the future will contribute to a deeper understanding of the regulatory network of
marine cyanobacteria and promote new studies on bacterial ncRNAs.

Keywords:
cyanobacteria, comparative genomics, non-coding RNAs, promoter
 Zusammenfassung

Das Leben auf der Erde wird maßgeblich durch die Kraft der oxygenen Photosythese bestimmt, die
Sonnen- in chemische Energie umwandelt. Cyanobakterien wie Prochloro- und Synechococcus zäh-
len zu den wichtigsten primären Produzenten der Ozeane und werden zunehmend als Modelle für
photosynthetische Organismen genutzt. Um die Regulationsmechanismen dieser Picocyanobakteri-
en besser zu verstehen, wurde hier die Information von vier Genomen hochgradig verwandter aber
dennoch ökologisch unterschiedlich angepasster mariner Stämme genutzt in einer Kombination aus
computer-gestützten und experimentellen Untersuchungen. Sequenzsignale und RNA-kodierende
Gene wurden als neuartige Regulationselemente identifiziert und entlang des phylogenetischen
Gradienten verglichen. Mittels ’phylogenetic footprinting’ konnte ein minimales, konserviertes Set
möglicher Transkriptionsfaktoren, deren Bindestellen und Regulons aufgedeckt werden. NtcA-,
LexA- und ArsR-ähnliche Motive wurden ebenso gefunden wie neue regulatorische Elemente. Mit
Hilfe von RACE Experimenten wurden einige der vorhergesagten Bindestellen Promotorregionen
zugeordnet. Der cyanobakterielle Promotor besitzt E. coli ähnliche Merkmale. Konservierte Se-
quenzelemente an Position -60 stellen vermutlich ein Merkmal bestimmter bidirektionaler Pro-
motoren dar. Für den Cyanophagen P-SSP7 ergaben Promotorstudien eine konservierte Sequenz,
die entweder durch die Phagen-RNA-Polymerase oder/und die Wirt’s RNA Prozessierungsma-
schinerie erkannt wird. Eine Suche nach konservierten Sekundärstrukturen detektierte mehrere
nicht-kodierende RNAs, benannt Yfr für cYanobacterial Functional RNA. Eine dieser ncRNAs,
Yfr1, scheint für das Wachstum in großen Tiefen verzichtbar. Eine konservierte Sekundärstruktur
und eine ungepaarte CA-Dinukleotid-Wiederholung könnten funktionale Elemente von Yfr1 dar-
stellen. Eine vergleichende Analyse der Strukturen, Transkriptvarianten und Akkumulation von
Yfr7 innerhalb der cyanobakteriellen Linie ergab, dass diese RNA wahrscheinlich ein Homolog der
E. coli 6S RNA ist. Zwei verschiedene Yfr7 Transkripte mit einem zirkadianen aber zeitversetzten
Akkumulationsmuster lassen eine Verknüpfung ihrer Expression mit dem zirkadianen Rhythmus
oder der Lichtintensität vermuten. Experimente in Synechocystis deckten einen neuartigen Regu-
lationsmechanismus durch eine antisense RNA auf, welche die Menge der isiA mRNA kontrolliert
und die Assemblierung von IsiA-Superkomplexen beeinflusst. Die funktionelle Zuordnung dieser
neuen Elemente wird zu einem besseren Verständnis regulatorischer Netzwerke in marinen Cyano-
bakterien und darüber hinaus führen.

Schlagwörter:
Cyanobakterien, Genomvergleich, nicht-kodierende RNAs, Promotor
Phantasie ist wichtiger als Wissen, denn Wissen ist begrenzt

Albert Einstein
Contents

1 Introduction 1
1.1 Shedding light on marine cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Cyanophages are mixing up the oceanic gene pool . . . . . . . . . . . . . . . . . . 5
1.2.1 T7-like transcription system - a characteristic strategy of infection . . . . . 6
1.3 Architecture and regulation of a transcription unit . . . . . . . . . . . . . . . . . . 9
1.4 Phylogenetic footprinting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Transcriptional regulatory networks . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.6 The circadian clock of cyanobacteria is unique . . . . . . . . . . . . . . . . . . . . . 13
1.6.1 Roles of group 2 σ factors in circadian regulation . . . . . . . . . . . . . . . 15
1.7 The new era of small RNA molecules - tiny but mighty regulators . . . . . . . . . 16
1.7.1 How to find them? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.7.2 Are the newly identified ncRNAs functional? . . . . . . . . . . . . . . . . . 19
1.7.3 Riboswitches: mRNA-enslaved ncRNAs . . . . . . . . . . . . . . . . . . . . 19
1.7.4 Are there regulatory RNAs in cyanobacteria? . . . . . . . . . . . . . . . . . 20
1.8 Aims of this work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2 Materials and Methods 23

2.1 Experimental Part . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.1.1 Cultures, cultivation, plasmids . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.1.2 DNA and RNA oligos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.1.3 Genetic manipulation of marine Synechococcus . . . . . . . . . . . . . . . . 26
2.1.4 Isolation of nucleic acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1.5 DNA gel electrophoresis, Restriction, Ligation . . . . . . . . . . . . . . . . . 29
2.1.6 RNA gel electrophoresis, Northern blotting and Hybridisation . . . . . . . . 29
2.1.7 Polymerase chain reaction (PCR) . . . . . . . . . . . . . . . . . . . . . . . . 32
2.1.8 Sequencing of DNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1.9 Rapid amplification of cDNA ends (RACE) . . . . . . . . . . . . . . . . . . 33
2.1.10 Quantitative RT-PCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.2 Bioinformatics part . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.2.1 Genome data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.2.2 Software tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.2.3 Promoter prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.2.4 Phylogenetic footprinting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.2.5 Prediction of ncRNAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3 Results 42
3.1 Cyanobacterial promoters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.1.1 psbD promoter region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.1.2 csoS1-rbcLS promoter region . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.3 atpB promoter region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.1.4 sfsA downstream region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
 3.2 Phylogenetic footprinting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2.1 Orthologous gene sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.2.2 Predicted TF binding sites and regulons . . . . . . . . . . . . . . . . . . . . 48
3.2.3 Experimental verification of transcription initiation next to putative binding
sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.3 Circadian rhythm and kai genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.4 Transcriptional regulation of the cyanophage P-SSP7 . . . . . . . . . . . . . . . . . 57
3.4.1 Mapping of 5’ ends by RACE experiments . . . . . . . . . . . . . . . . . . . 59
3.4.2 Putative motif upstream of genes 1, 3 and 29 . . . . . . . . . . . . . . . . . 60
3.4.3 Transcription of phage-specific RNA polymerase . . . . . . . . . . . . . . . 62
3.4.4 Missing promoter site upstream gene 30 . . . . . . . . . . . . . . . . . . . . 63
3.5 Small RNAs in marine and other cyanobacteria . . . . . . . . . . . . . . . . . . . . 63
3.5.1 Known and abundant RNAs of marine strains . . . . . . . . . . . . . . . . . 63
3.5.2 Computational screening identified novel RNA species . . . . . . . . . . . . 64
3.5.3 Experimental verification of predicted ncRNAs in Prochlorococcus Med4 . . 65
3.5.4 Yfr1 exists in diverse cyanobacteria . . . . . . . . . . . . . . . . . . . . . . . 69
3.5.5 Conservation of Yfr7/6Sa across the whole cyanobacterial lineage . . . . . . 70
3.5.6 Excursus to Synechocystis PCC 6803: A cis-encoded antisense RNA for isiA 78

4 Discussion 81
4.1 Promoter architecture and transcriptional regulation . . . . . . . . . . . . . . . . . 81
4.1.1 The conserved regulatory potential of four marine cyanobacteria . . . . . . 82
4.1.2 Circadian rhythm: clock proteins and other periodosome components . . . 85
4.1.3 Transcriptional signals of cyanophage P-SSP7 infecting Prochlorococcus Med4 86
4.2 New families of small RNAs in cyanobacteria . . . . . . . . . . . . . . . . . . . . . 87
4.2.1 Computational screening in marine genomes . . . . . . . . . . . . . . . . . . 88
4.2.2 Non-coding RNAs of Prochlorococcus Med4 and their orthologs . . . . . . . 89
4.2.3 Ubiquitous presence of 6S RNA in cyanobacteria and clues about its function 90
4.2.4 A novel antisense RNA to isiA mRNA in Synechocystis PCC 6803 . . . . . 91
4.3 Conclusion and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

Bibliography 95

A Appendix 113
List of Figures

1.1 Transmission electron micrograph of an ultrathin section of P. marinus SS120 (Bryant, 2003). 1
1.2 Depiction of the amount of chlorophyll present in the oceans and the amount of vegetation
on land (Bryant, 2003). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 False color image of sea surface temperature during the Atlantic Ocean meridional transect
cruise and cross sections along the transect (Johnson et al., 2006). . . . . . . . . . . . . . . 3
1.4 Relationships between Prochlorococcus and other cyanobacteria inferred using 16S rDNA
(Rocap et al., 2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Genome arrangement of Prochlorococcus podovirus P-SSP7 (Sullivan et al., 2005) . . . . . . 7
1.6 Synthesis and processing of T7 messenger RNAs (Dunn and Studier, 1983) . . . . . . . . . 8
1.7 Overview of the transcriptional regulatory network in E. coli (Martinez-Antonio and Collado-
Vides, 2003). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.8 The cyanobacterial periodosome model (Bell-Pedersen et al., 2005). . . . . . . . . . . . . . . 14
1.9 A model for fhlA-OxyS interaction (Argaman and Altuvia, 2000). . . . . . . . . . . . . . . . 17
1.10 Schematic representation of the proposed mechanism for TPP-dependent deactivation of
thiM translation (Winkler et al., 2002). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

2.1 Cultures of Prochlorococcus and Synechococcus. . . . . . . . . . . . . . . . . . . . . . . . . . 24

2.2 PCR result of 5’ RACE for psbD in Prochlorococcus SS120. . . . . . . . . . . . . . . . . . . 34
2.3 Pipeline for comparative prediction of non-coding RNAs (Axmann et al., 2005). . . . . . . . 41

3.1 Raster-score-filter method (Vogel et al., 2003a) and weblogo of Prochlorococcus SS120 -10
boxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2 Arrangement of ppiA, ycf4, psbD and psbC and alignment of upstream regions of psbD for
five cyanobacterial strains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.3 Comparison of the csoS1-rbcLS region from marine cyanobacteria. . . . . . . . . . . . . . . 45
3.4 Arrangement of genes around atpB and alignment of atpB upstream sequences for five marine
cyanobacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.5 Synteny of mviN, sfsA, amt1 and lytB and alignment of sfsA downstream sequences for five
marine strains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.6 Overview of orthologous protein-coding genes including σ and regulatory factors of four
marine genomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.7 Results of the PCR step during 5’ RACE experiments for lexA, umuD, PMM1427, urtA,
glnA, ntcA in Prochlorococcus Med4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.8 Arrangement of rpl27 and rpl21, encoding 50S ribosomal proteins, and kaiA, kaiB and kaiC
encoding the core clock proteins of cyanobacteria. . . . . . . . . . . . . . . . . . . . . . . . . 54
3.9 Alignment of coding regions of kaiA from Synechococcus PCC 7942, WH 8102 and 7803
together with the spacer regions between kaiB and rpl21 of Prochlorococcus strains, MIT
9313 and 9303. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.10 Alignment of the conserved rpl21 upstream regions of Synechococcus and Prochlorococcus. . 56
3.11 5’ RACE experiments for the cyanophage P-SSP7. . . . . . . . . . . . . . . . . . . . . . . . 59
3.12 Summary of identified sites and sequence comparisons for P-SSP7. . . . . . . . . . . . . . . 61
3.13 Small RNAs in marine Cyanobacteria (Axmann et al., 2005). . . . . . . . . . . . . . . . . . 64
3.14 Experimental screen for the presence of an RNA-coding gene in the guaB-trxA intergenic
region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.15 Determination of half lifes for Yfr 1, Yfr2 and Yfr5 to 7 in Prochlorococcus Med4 (Axmann
et al., 2005). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

3.16 Test of transcript accumulation of Yfr1-7 from Med4 (MED) under different conditions
(Axmann et al., 2005). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

3.17 Comparison of Yfr2, Yfr3, Yfr4 and Yfr5 from Med4. . . . . . . . . . . . . . . . . . . . . . . 70

3.18 Identification of Yfr1 RNA in diverse cyanobacteria. . . . . . . . . . . . . . . . . . . . . . . 71

3.19 Identification of Yfr7 RNA in different marine cyanobacteria. . . . . . . . . . . . . . . . . . 72

3.20 Identification of 6Sa RNA in different cyanobacteria. . . . . . . . . . . . . . . . . . . . . . . 73

3.21 6Sa/Yfr7 Parsimony tree based on the alignment of 13 sequences from diverse cyanobacteria
and genome location of ssaA in these strains. . . . . . . . . . . . . . . . . . . . . . . . . . . 74

3.22 Identification and transcript accumulation of Yfr7 in Prochlorococcus Med4. . . . . . . . . . 75

3.23 Knock-out construct for Yfr7 in Synechococcus WH 8102 and analysis of mutants. . . . . . 76

3.24 Accumulation of different transcripts from the isiAB region under iron limitation and during
high-light conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

3.25 Characterisation of an antisense RNA in Synechocystis PCC 6803. . . . . . . . . . . . . . . 79

4.1 Influence of IsrR on the accumulation of trimeric PSI and PSI-IsiA supercomplexes (Duehring
et al., 2006). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

2 Sequence logos of the best conserved elements obtained by phylogenetic footprinting analysis
of Prochlorococcus Med4, MIT 9313, SS120 and Synechococcus WH 8102. . . . . . . . . . . 113

3 List of high scoring clusters revealed by RNA prediction in marine cyanobacteria (Axmann
et al., 2005). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

4 6Sa/Yfr7 alignment of 13 sequences from diverse cyanobacteria. . . . . . . . . . . . . . . . . 122

List of Tables

2.1 DNA oligos used for PNK labelling and hybridisation to Northern blots. . . . . . . . . . . . 31

2.2 DNA oligos used for PNK labelling and hybridisation to Northern blots, which did not yield
signals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.3 PCR primer list. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.4 DNA oligos used for 5’ RACE experiments analysing ncRNAs. . . . . . . . . . . . . . . . . 34

2.5 DNA oligos used for 5’ RACE experiments for cyanophage P-SSP7. . . . . . . . . . . . . . . 35

2.6 DNA oligos used for 5’ RACE experiments. . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

2.7 Prochlorococcus Med4 gene-specific RT-PCR primers. . . . . . . . . . . . . . . . . . . . . . 37

3.1 Experimentally determined mRNA 5’ ends of Prochlorococcus SS120. . . . . . . . . . . . . . 43

3.2 Top five of predicted motifs and their best hits identified in Prochlorococcus Med4, SS120,
MIT 9313 and Synechococcus WH 8102 via phylogenetic footprinting analysis. . . . . . . . . 49

3.3 Overview of clock gene orthologs in Prochlorococcus Med4, SS120, MIT 9313, Synechococcus
WH 8102 in comparison to Synechococcus PCC 7942, the circadian clock model. . . . . . . 56

3.4 Summary of transcriptional and other signals mapped to the cyanophage P-SSP7 genome. . 58

3.5 Summary of identified ncRNA genes in Prochlorococcus MED4 and their orthologues in
three related strains of marine cyanobacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . 66

1 A core set of 35 regulatory factor and 5 σ factor genes conserved between Prochlorococcus
Med4, MIT 9313, SS120 and Synechococcus WH 8102. . . . . . . . . . . . . . . . . . . . . . 114

2 List of predictions for putative TF binding sites in marine cyanobacteria. . . . . . . . . . . 120

Chapter 1

Introduction

1.1 Shedding light on marine cyanobacteria

Today it seems, we know more about the outer space than about the oceans of our Earth
covering nearly 75 % of its surface. Thus, it was not surprising that no more than 20 years
ago, a new type of photosynthetic picoplankton was discovered in the open sea (Chisholm
et al., 1988) which since has changed the understanding of marine food webs as well as
of the global geochemical carbon cycle completely (Falkowski, 2002). Because of their
small cell sizes with diameters of only 0.5 to 0.7 µm and low cell densities, these tiny
cyanobacteria had not been discovered until sensitive shipboard flow cytometry became
available. A distinctive feature of all cyanobacteria known at that time was the presence
of a phycobilisome-type light-harvesting complex, for which, however, no evidence was
found in these organisms. This fact, together with the unusual combination of pigments,
and a typical prokaryotic ultrastructure, suggested that these picoplankters are free-living
relatives of the prochlorophyte Prochloron (Chisholm et al., 1988), which performs oxy-
genic photosynthesis using chlorophyll b, like land plants and green algae (Chlorophyta)
(Tomitani et al., 1999). Taking also the coccoid shape (Fig. 1.1) into account, the name
of the new genus was decided: Prochlorococcus (Chisholm et al., 1992).

Figure 1.1: Transmission electron micrograph of an ultrathin section of P. marinus SS120, taken from
Bryant (2003), showing the closely appressed thylakoids that distinguish this organism from marine Syne-
chococcus sp. (Photograph courtesy of William Li and Frederic Partensky).

Fossil and molecular studies suggested that the archetype of cyanobacteria thrived success-
fully for billions of years (Schopf, 1993). Photosynthesis evolved early in these organisms,
an activity that resulted in the enrichment of the planetary atmosphere in oxygen, so that
2 1. INTRODUCTION

about 1.5 billion years ago our vital atmosphere had been created. As an endosymbiont
the ancient cyanobacteria were transformed into an intracellular organelle - the chloroplast
- without which no eukaryotic photoautotroph can exist, among them all higher plants
(Douglas, 1998; Moreira et al., 2000). As a consequence of gene transfer from the endosym-
biont to the original host’s genome, about one fifth of all nuclear genes of extant land plants
might descend from the cyanobacterial endosymbiont (Rujan and Martin, 2001; Martin
et al., 2002). However, the known prochlorophytes, including Prochlorococcus marinus,
have been shown to be not the specific ancestors of chloroplasts (Tomitani et al., 1999).
Today all life on Earth depends on photosynthesis. Now we are recognising that marine
phytoplankton accounts for nearly 50 % of the net primary productivity of the biosphere
(Field et al., 1998) which was grossly underestimated until recently. Satellite-based re-
mote sensing (e.g., NASA sea-wide field sensor) has allowed more reliable determinations
of oceanic photosynthetic productivity to be made (Field et al., 1998; Falkowski, 2002;
Bryant, 2003); see Figure 1.2.

Figure 1.2: Depiction of the amount of chlorophyll present in the oceans and the amount of vegetation
on land, taken from Bryant (2003). Purple and blue represent low levels of chlorophyll, and green, yellow,
and red indicate progressively higher concentrations. Brown pixels show areas with little vegetation on
land, and blue-green pixels represent areas of densest vegetation (Bryant, 2003). [Image provided by
ORBIMAGE (Copyright 2003, Orbital Imaging Corporation) and processing by NASA Goddard Space
Flight Center.]

Cyanobacteria today form a huge and heterogenous group of prokaryotes which is different
in many features from other bacteria. They populate widely diverse environments such as
freshwater, the oceans, hot springs, the surface of rocks, desert soil or the Antarctic. The
group of Prochlorococcus and the closely related Synechococcus is abundant in the oceans
and belongs to the most important primary producers as it could be responsible for nearly
one-third of the primary biomass production on Earth (Fig. 1.2).
Total genome sequences are available for Prochlorococcus marinus SS120 (Dufresne et al.,
2003), Prochlorococcus Med4 and Prochlorococcus MIT 9313 (Rocap et al., 2003). These
strains are representative for distinct populations found in particular ecological niches
within the marine ecosystem. The genus Prochlorococcus is often present at high abun-
dances with more than 105 cells per ml in nutrient-poor areas of the world’s oceans split-
ting up in two major ecotypes (Fig. 1.4) - one being represented by the high-light-adapted
strains such as Med4, the other by low-light-adapted strains SS120 and MIT 9313 (Moore
et al., 1998; Partensky et al., 1999; West and Scanlan, 1999; Fuhrman and Capone, 2001).
1.1. SHEDDING LIGHT ON MARINE CYANOBACTERIA 3

These sometimes co-occurring ecotypes have not only different light optima for growth,
pigment contents and light-harvesting efficiencies but also different temperature optima
(Johnson et al., 2006), sensitivities for trace metals, and are each specifically affected by
cyanophages (Rocap et al., 2003; Moore et al., 1998; Sullivan et al., 2003). Nevertheless,
on the basis of their rDNA similarity they would be recognised as a single species as their
ribosomal DNA sequences differ by less than 3 % (Hagstrom et al., 2002).
Very recently, the variable Prochlorococcus ecotype distribution was correlated to en-
vironmental gradients along a meridional transect in the Atlantic Ocean, showing that
temperature plays a dominant role in regulating some ecotype distributions (see Fig. 1.3).
Other factors as light and competitor abundance (Synechococcus) can also be important,
while nutrient availability seems to play a minor role (Johnson et al., 2006).

Nitrate

Prochl.

eMIT9312
80 60 40 20 0
C

eMED4

eNATL2A

eMIT9313

Figure 1.3: Left panel: False color image of sea surface temperature during the Atlantic Ocean meridional
transect cruise (AMT13), represented by Johnson et al. (2006), during Sept/Oct 2003 overlaid with cruise
track and sampled stations (boxes). Reds are warm temperatures, blues are cold temperatures. Right
panel: Cross section along the transect (upper 200 m) of (A) nitrate concentration (µmol/kg), (B) total
Prochlorococcus (log cells/ml) and four dominant ecotype clade abundances (log cells/ml): (C) eMIT9312,
(D) eMED4, (E) eNATL2A, and (F) eMIT9313. Temperature contour lines are shown in white. The lower
limit of detection is ∼1 cell/ml (Johnson et al., 2006).

The next sequenced relative to the group of Prochlorococcus (Fig. 1.4) is the strain Syne-
chococcus WH 8102 (Urbach et al., 1992), which has evolved a unique type of swimming
motility (Palenik et al., 2003). Marine Synechococcus species, first described in 1979 (Wa-
terbury et al., 1979; Johnson and Sieburth, 1979), are typically less abundant in very
oligotrophic environments, but have a broader global distribution usually in the upper
4 1. INTRODUCTION

layer of the ocean, where white or blue-green light is available for photosynthesis (Ting
et al., 2002; Bryant, 2003). Known so far, all marine Synechococcus strains harbour chloro-
phyll a (and no chlorophyll b) and phycobilisomes (Waterbury et al., 1986), but a great
genetic diversity exits within this genus as well, with strains probably adapted to ecologi-
cal niches, whereby WH 8102 is more of a generalist with adopted strategies for conserving
limited iron stores and a reduced regulatory machinery (Palenik et al., 2003).
Compared to Synechococcus and other cyanobacteria, strains of Prochlorococcus are un-
usual in lacking phycobilisome light-harvesting complexes. Instead chlorophyll-protein
complexes known as Pcbs are used, which contain divinyl chlorophyll a and chlorophyll b
(La Roche et al., 1996; Bibby et al., 2001b; Hess et al., 2001).

------
*
*
------
--------

--------

*
*
-------------------------------
--------------------

-----------------

Figure 1.4: Relationships between Prochlorococcus and other cyanobacteria inferred using 16S rDNA,
shown by Rocap et al. (2003). Strains with dashed lines are sequenced, annotated and published (Dufresne
et al., 2003; Rocap et al., 2003; Palenik et al., 2003; Nakamura et al., 2002; Kaneko and Tabata, 1997;
Kaneko et al., 2001). For strains labelled with a star complete genome sequence is available at NCBI
database (Wheeler et al., 2005). Unfinished sequences by whole genome shotgun sequencing projects exist
for several other strains at NCBI (Wheeler et al., 2005).

The information of at least four fully sequenced and annotated genomes of closely related
and even so ecologically divergent marine cyanobacteria offers a great chance for compar-
ative studies. The conserved sequence motifs and patterns essential for regulation as well
as individual regulatory signals and specific gene contents responsible for niche differentia-
tion could be filtered out and studied in detail. All sequenced strains are available for cell
1.2. CYANOPHAGES ARE MIXING UP THE OCEANIC GENE POOL 5

culturing, and a genetic manipulation system has been developed for Synechococcus WH
8102 (Brahamsha, 1996). Thus, not only a computational analysis of these cyanobacterial
genomes appears promising, but these can be directly linked to corresponding experiments
in the laboratory to verify and improve the in silico predictions.
That could be a starting point to understand how these tiny and even so specialised organ-
isms could dominate the oceans for millions of years although environmental conditions
were and are changing. A major outcome of the total genome analyses of these cyanobac-
teria has been their small and compact genomes, being considered as the minimal genome
of a free-living photoautotroph. But, which regulatory potential does remain in such a
streamlined bacterium to enable responses to variations in temperature, light condition,
nutrient or metal? Today the global warming is faster than it was assumed initially -
to what extent the complex marine ecosystem will be affected? Can the regulatory and
genetic potential of these microbes sustain adaptation to a big global change? At which
point will it suddenly and unexpectedly crash, with a major impact for our biosphere?

1.2 Cyanophages are mixing up the oceanic gene pool

The analysis of four genomes of marine cyanobacteria has revealed an unprecedented de-
gree of genomic variation. What processes could alter microbial genotypes and enable
such extensive niche differentiation as observed within the group of slow-growing Pro-
chlorococcus and Synechococcus in the ocean? The annotation of the sequenced genomes
yielded INT family site-specific recombinases, suggesting that prophages were once inte-
grated. The comparison and analysis of very closely related genomes as Prochlorococcus
Med4 and MIT 9313 revealed numerous large and small-scale rearrangements, where the
break points between the orthologous gene clusters are often flanked by tRNAs (Rocap
et al., 2003). It was already suggested that tRNA genes (and tmRNA genes) are common
integration sites for phages and mobile elements (Williams, 2002).
Indeed, cyanophages were isolated that infect Prochlorococcus. Thereby, Podoviridae
normally infecting high-light-adapted Prochlorococcus were found to be extremely host-
specific (Sullivan et al., 2003). For example, P-SSP7 exclusively infects the high-light-
adapted Prochlorococcus Med4 strain. On the other hand, low-light-adapted Prochloro-
coccus as well as all Synechococcus strains yielded primarily Myoviridae (Sullivan et al.,
2003).

Genome sequencing of several cyanophages uncovered another fascinating feature of cy-

anobacterial viruses: Besides established "core" genes known for each phage group, the
cyanophage genomes contain not only a few host-homologous genes ’by accident’, but even
encode functional proteins, which are central to oxygenic photosynthesis and may be cru-
cial to maintain photosynthetic activity during infection (Mann et al., 2003; Lindell et al.,
2004; Millard et al., 2004). Some of the host genes that appear common in these phages
are suggested to represent "signature" genes which could be specifically characteristic of
oceanic cyanophages (Sullivan et al., 2005). For example, all marine cyanophage genomes
(except podovirus P60) sequenced so far contain a psbA gene that encodes a core reaction
center protein (D1) in photosystem II. Very recently it was shown that phage psbA and
hli (high-light inducible) genes are expressed during infection of Prochlorococcus and that
6 1. INTRODUCTION

they may increase phage fitness (Lindell et al., 2005).

Interestingly, the distribution of other phage-encoded signature genes is more sporadic
among the two viral families (Podoviridae and Myoviridae) which may reflect adaptations
for infection of photosynthetic hosts in low-nutrient oceanic environments (Sullivan et al.,
2005). Thus, both T4-like myoviruses P-SSM2 and P-SSM4 harbour genes (phoH, pstS)
that are likely to be important for responses to phosphate stress (Sullivan et al., 2005).

For bacteriophages an obligate lytic or temperate (lysogenic) lifestyle may be distinguished.

Infection by a temperate phage may either cause lysis or entry of a stable state within the
host (lysogeny) by integration of the phage genome into the host genome as a prophage,
which could be induced back into the lytic cycle in response to environmental conditions
(Mann, 2003).
Viral infection of marine unicellular cyanobacteria was first described in 1990 (Suttle et al.,
1990; Proctor and Fuhrman, 1990), and up to now, temperate phages have not been iden-
tified in complete genome sequences of marine Prochlorococcus and Synechococcus strains,
although some field studies dealing with lysogeny in natural populations suggest this is a
possibility (Ortmann et al., 2002; McDaniel et al., 2002).
However, the phenomenon of pseudolysogeny was observed by studying interactions be-
tween phage and nutrient-starved Synechococcus host (Ripp and Miller, 1997; Wilson et al.,
1996). Pseudolysogeny describes a phage-host relationship in which a phage-infected cell
grows and divides even though its virus is pursuing a lytic infection (Birge, 2000).

The marine Podoviridae known so far have an obligate lytic lifestyle (lacking known
lysogeny genes), although in the T7-like P-SSP7 genome a tyrosine site-specific recom-
binase (int) gene, which enables a temperate phage to integrate its genome into the host
genome, as well as a 42-bp sequence identical to the 5’ part of the leucine tRNA gene
in the host genome (Prochlorococcus Med4) was discovered (Sullivan et al., 2005). The
cyanophage P-SSP7 complete genome is available now (Sullivan et al., 2005) encoding 54
open reading frames (ORFs) within nearly 45 kb. It includes a T7-like RNA polymerase
gene as well as the host-like hli and psbA genes (Fig. 1.5).

Half of the translated ORFs in P-SSP7 could not be assigned a function (Sullivan et al.,
2005) which is a general problem for marine phage genomes. They have been described
as the largest untapped reservoir of genetic information (Paul et al., 2005), and further
sequences of marine phages as well as future studies are needed to understand the evolu-
tionary history and biological and ecological functions of the marine viral community.

1.2.1 T7-like transcription system - a characteristic strategy of infection

The cyanophage P-SSP7 genome is most similar to genomes of the T7-like phages (Sul-
livan et al., 2005). The lytic phage T7 infecting E. coli cells is one of the best studied
podoviruses. Members of this group (e.g. T7, T3, SP6) possess a unique strategy of infec-
tion, where the phage’s own transcription system plays an important role in phage gene
expression and also drives the efficient translocation of viral DNA into the host cells (Chen
and Schneider, 2005). The T7 genome contains bacterial (host-like, class I) promoters as
well as 17 promoters only recognised by their T7 RNA polymerase (RNAP) (Dunn and
1.2. CYANOPHAGES ARE MIXING UP THE OCEANIC GENE POOL 7

Figure 1.5: Genome arrangement of Prochlorococcus podovirus P-SSP7, described by Sullivan et al.
(2005). The ORFs are sequentially numbered within the boxes, and gene names are designated above the
boxes. Gene designations use T7 nomenclature for T7-like genes (Molineux, 2001) or microbial nomencla-
ture for non-phage genes. Class I, II, and III genes refer to those in T7 that belong to gene regions primarily
involved in host transcription of phage genes (class I), DNA replication (class II), and the formation of the
virion structure (class III). The ORFs are designated by boxes, and in this genome, all ORFs are oriented
in the same direction. Although the phage genome is one molecule of DNA, the representation is broken
to fit on a single page. Note that the P-SSP7 genome is most similar to genomes of the T7-like phages
(Sullivan et al., 2005).

Studier, 1983). The promoters for T7 RNAP were classified into three groups: class II,
class III or replication promoters depending on their location, temporal utilisation and
function during infection. Thus, T7 genes are transcribed from left to right and expressed
co-ordinately in three groups as illustrated in Figure 1.6a (Dunn and Studier, 1983):
(I) At first, the host RNAP initiates transcription at strong bacterial promoters located
near the left end of the bacteriophage DNA and transcribes the genome until the early
termination site (TE), resulting in a segment of about 20 % of the phage genome, which
includes the phage RNAP gene.
(II) During the next stage of infection, the newly-made phage RNAP specifically recognises
its promoters and transcribes about 2/3 of the phage genome until a strong termination
site (Tφ) without any accessory factors.
(III) Termination at Tφ is not completely efficient and would be lethal to T7 as there is
no promoter for T7 RNAP between this termination site and the following genes, which
specify structural proteins for T7, so these genes must be transcribed entirely by read-
through of Tφ. Thus, the class III genes are the last to be expressed.

Additionally, cleavage of T7 RNAs at specific sites is a prominent feature (Dunn and

Studier, 1973a,b, 1975): Ten RNase III cleavage sites were located inside the T7 genome,
8 1. INTRODUCTION

A B

Figure 1.6: Synthesis and processing of T7 messenger RNAs summarised by Dunn and Studier (1983).
(A) The T7 genes are represented by open boxes; the positions of transcription signals are given above the
genes, the RNase III cleavage sites below. The primary transcript from each promoter is represented by a
horizontal line, and sites of RNase III cleavage by short vertical lines. Apparently not all RNAs are cut at
the R3.8 and R13 RNase III cleavage sites, as indicated by parentheses. RNAs produced by ready-through
of Tφ are also represented. The scale bar below is in map units. (B) Two examples for RNase III cleavage
sites in T7 RNAs. Potential base-paired structures are indicated, as are the locations of the termination
codons for the genes immediately preceding, and the ribosome-binding sequences and initiation codons for
the genes immediately beyond the cleavage sites. The locations of the promoters for T7 RNAP are also
shown. The known positions of RNase III cleavage for the two (of five) cleavages in T7 early RNA are
indicated. The position of the secondary cleavage in R1.3 is indicated by a broken line (Dunn and Studier,
1983).

five in the early (Kramer et al., 1974; Rosenberg et al., 1974; Kramer and Rosenberg, 1976;
Rosenberg and Kramer, 1977; Robertson et al., 1977) and five in the late region (Dunn
and Studier, 1983). At these sites, the primary transcripts are processed by a host enzyme,
RNase III, to provide the mRNAs observed in vivo (Dunn and Studier, 1983). Ribonu-
clease III from E. coli participates in maturation and decay pathways by site-specifically
cleaving double-helical structures in cellular and viral RNAs, whereby the involvement of
base-pair sequence in determining cleavage sites is unclear (Zhang and Nicholson, 1997).
In the T7 genome, however, the promoter regions for T7 RNAP often overlap with se-
quences needed for the RNase III cleavage forming a characteristic pattern of base-pairing
(Fig. 1.6b).

The expression of T7 RNAP early during infection and the inactivation of the host poly-
merase, efficiently directs all gene expression in the cell to T7 DNA (Zhang and Studier,
2004). T7 RNAP appears to have multiple roles in selective transcription of T7 genes
as well as in replication, maturation and packaging (Zhang and Studier, 2004). During
the last stage of infection, T7 lysozyme binds T7 RNAP and inhibits transcription, and
stimulates replication and packaging of T7 DNA (Zhang and Studier, 1997, 2004). At
the end, T7 lysozyme cuts a bond in the peptidoglycan layer of the cell wall to release
1.3. ARCHITECTURE AND REGULATION OF A TRANSCRIPTION UNIT 9

newly-made phages.

For the circular marine P-SSP7 genome (Fig. 1.5) a similar architecture compared to
coliphage T7 was observed including the essential T7-like RNAP (Sullivan et al., 2005),
which may implicate a similar infection strategy as described above. On the other hand,
in a recent study about T7-like promoters and their RNAPs, no significant promoter sites
were found for P-SSP7 suggesting that this phage does not belong to the highly related
T7 group (including T7, T3, SP6) but rather to a more distantly related T7 supergroup
(Chen and Schneider, 2005). Chen and Schneider (2005) proposed, that P-SSP7 may
utilise a different transcriptional strategy to that known for T7, because P-SSP7 does not
contain both components of the T7-like transcription system (RNA polymerase and its
specific promoters). Further studies are needed focusing on P-SSP7 DNA transcription
during Med4 lytic infection to identify different stages of gene expression as well as their
responsible promoter sites.

1.3 Architecture and regulation of a transcription unit

A transcription unit is defined as the DNA between a promoter and a terminator that
serves as a template for a complementary strand of RNA which can be messenger RNA
(mRNA) encoding a protein, ribosomal RNA (rRNA), transfer RNA (tRNA) or small
non-coding RNA (ncRNA) with regulatory functions. The bacterial RNA polymerase
holoenzyme that synthesises these RNAs consists of two components - a core enzyme re-
sponsible for elongating the RNA chain and a σ factor required at the stage of initiation
for promoter recognition. Bacteria contain several σ factors, a strategy that is in itself
the first level of regulation of transcription initiation. Alternative σ factors activated by
adverse conditions (heat shock, nitrogen starvation) can direct the core enzyme to recog-
nise promoters with different consensus sequences. As new σ factors become active, old σ
factors are displaced, that turns genes off and dictates when its set of target genes is ex-
pressed, and the amount of factor available influences the level of gene expression (Lewin,
2000; Rojo, 1999). More than one σ factor may be active at any time, and the specificities
of some of σ factors overlap equipping the bacterial cell with a flexible regulatory network.

Transcription has been studied the best in Escherichia coli which uses seven different
species of σ factors to modulate promoter activation. Based on the abundant σ factor
of housekeeping genes (σ 70 reviewed by Ishihama (2000)) a consensus sequence of σ 70 -
dependent promoters has been established, and is defined by two consensus hexamers,
TTGACA and TATAAT, centred -35 and -10 base-pairs respectively upstream of the
transcription initiation site (Hawley and McClure, 1983).

While the basic principles of promoter recognition and geometry of RNA polymerase seems
to be similar for all eubacteria, the situation in cyanobacteria could be different by the
presence of a truncated RNA polymerase subunit β’ (γ) encoded by rpoC1 (Xie et al.,
1989) and an additional subunit β” (δ) encoded by rpoC2. These subunits correspond
to the N- and C- terminal parts of the β’ subunit of other eubacteria plus an additional
protein domain of 70 kDa, and are found in this split form in all cyanobacteria including
10 1. INTRODUCTION

Prochlorococcus as well as in plant plastids. Although RNA polymerases from enterobac-

teria and cyanobacteria share similar domains and subunits, it was shown that they are
not directly interchangeable (Schyns et al., 1998). It would not be the first case, where
known facts about E. coli cannot be extrapolated directly to any other bacterium, in this
case to a cyanobacterium. On the other hand, only a few studies existed about factors
required for the transcription of a gene in a cyanobacterium.
The first systematical study of transcription initiation sites was done in a marine cyanobac-
terium, Prochlorococcus Med4 where a subset of promoters was analysed to suggest a
consensus promoter sequence and to predict additional promoters in the whole genome of
Med4 (Vogel et al., 2003a). The computational analysis revealed some general features of
the cyanobacterial promoter architecture. The alignment of the experimentally identified
regions suggested a consensus promoter structure similar to other eubacteria whereby -10
region and TIS are spaced by about six base-pairs; transcription and translation start site
are spaced by 15 to 85 base-pairs in most cases. The -10 box itself exhibits three conserved
nucleotides: T(-12), A(-11) and T(-7) which partly represents the well-known consensus
TATAAT of E. coli and B. subtilis. The -35 regions do not possess overall conservations
but were found to define different subsets (Vogel et al., 2003a). One subset of promoters
with a -35 consensus similar to the known TTGACA sequence for house-keeping genes of
E. coli incloses the Med4 genes kaiB, rpl21, rps12 and ccmK (Vogel et al., 2003a).
This knowledge leads us to additional regulatory sites beside the core promoters such as
transcription factor binding sites (TFBS) and patterns responsible for a secondary struc-
ture of the 5’ untranslated region (5’ UTR) as in the case of recently found riboswitches.

1.4 Phylogenetic footprinting

Bacterial regulatory sites rarely occur as sequence patterns conserved over a certain length,
but rather as spaced, short words, what makes them more tricky to detect in the matrix of
millions of only four different letters A, T, G and C. Phylogenetic footprinting is the ma-
jor method for enriching for candidate regulatory elements (Bulyk, 2003). This technique
searches for conserved motifs upstream of orthologous genes from closely related species - a
search for ”islands of conserved sequences in seas of less conserved non-coding sequences”
(Pennacchio and Rubin, 2001). Sequence similarity is again the foundation for this bioin-
formatic method assuming that mutations within functional regions of genes accumulate
more slowly than mutations in regions without sequence-specific function (Wasserman and
Sandelin, 2004). The existing phylogenetic footprinting algorithms can be divided in three
components:

- defining suitable orthologous gene sequences for comparison

- aligning the promoter sequences of orthologous genes

- visualising or identifying segments of significant conservation

1.5. TRANSCRIPTIONAL REGULATORY NETWORKS 11

The great power of phylogenetic footprinting algorithms has been shown for organisms
of all kingdoms of life as the prediction of transcription regulatory sites in Archaea by
Gelfand et al. (2000), in 17 complete microbial genomes of E. coli and related γ-proteo-
bacteria by McGuire et al. (2000) and others, recently for the metal reducing bacterial
family Geobacteraceae (Yan et al., 2004), as well as for yeast (Cliften et al., 2001), mouse
and human (Loots et al., 2000; Gottgens et al., 2001), for example. Reviews of meth-
ods and available resources are given by numerous articles (Bulyk, 2003; Wasserman and
Sandelin, 2004; Ureta-Vidal et al., 2003; Frazer et al., 2003; Dubchak and Frazer, 2003).
Thereby, the initial and maybe the most challenging step is selecting a set of genomes with
the appropriate level of nucleotide similarity (evolutionary distance) of the sequences. One
has to balance the need to align orthologous sequences with the aim of having the func-
tional elements standing out (Cliften et al., 2001). The genomes of the four closely related
but differentially adapted marine cyanobacteria, Prochlorococcus Med4, SS120, MIT 9313
and Synechococcus WH 8102, may represent the right evolutionary distance to implement
a phylogenetic footprinting approach. Their genomes come close to the minimal genome
size for free-living autotrophs, so their regulatory machinery has been under intensive
streamlining pressure. Thus, a comparative analysis could provide clues about the min-
imal regulatory equipment of a marine picocyanobacterium mediated by already known
and hitherto unidentified genes and regulatory sequences like DNA binding sites of tran-
scriptional factors.

1.5 Transcriptional regulatory networks

Cells have to make decisions, when environmental conditions are changing. Their decision-
making process includes protein-DNA interactions defined by transcriptional factors (TF)
and their targets around promoters. How does the bacterial cell sense the changes in
environmental conditions which may start a signal-transduction pathway? Transcription
factors are mainly two-domain proteins, consisting of a DNA-binding domain along with
a regulatory domain (Madan Babu and Teichmann, 2003). The regulatory domains are
often sensors for small metabolites mediating precise activation or repression of the genes
belonging to the regulon of the TF. Other TFs are part of a two-component signal trans-
duction system where a CheY-like response regulator receiver domain is phosphorylated
by kinases.

The best known transcriptional network for any cell is the one of E. coli where about 300
genes (8 %) of 4405 identified ORFs are predicted or known TFs (Blattner et al., 1997;
Perez-Rueda and Collado-Vides, 2000). An overview of the complex network of currently
known interactions is given in Figure 1.7 where only seven regulatory proteins (CRP, FNR,
IHF, FIS, ArcA, NarL, and Lrp) directly modulate the expression of more than the half
of all genes in E. coli. On the other extreme, about one fifth of all transcription factors in
E. coli regulate only one or two genes (Martinez-Antonio and Collado-Vides, 2003). The
pleiotropic phenotype and the ability to regulate operons that belong to different path-
ways distinguish global regulators from local or dedicated regulators (Gottesman, 1984).
However, in many cases regulation occurs by multiple TFs and often a global regulator
acts together with more specific local regulators. The fact that three-quarters of TFs have
12 1. INTRODUCTION

arisen by gene duplication (Madan Babu and Teichmann, 2003) gives an idea how these
complex regulatory networks had been developed.

Figure 1.7: Overview of the transcriptional regulatory network in E. coli suggested by Martinez-Antonio
and Collado-Vides (2003). Regulated genes are shown as yellow ovals, TFs are shown as green ovals and
TFs considered to be global regulators are shown as blue ovals. The green lines indicate activation, red
lines indicate repression and dark blue lines indicate dual regulation (activation and repression) (Martinez-
Antonio and Collado-Vides, 2003).

In a regulatory network not only the number of interactions has to be considered but
also a hierarchy of different levels of regulation. In a bacterial cell where TFs regulate
TFs, including themselves in feedback loops (auto-regulation is frequent) and complex
relationships, a hierarchical organisation of the network can not be assumed from these
figures (Martinez-Antonio and Collado-Vides, 2003). In general, regulators work together
with other regulators in a given condition resulting in groups of co-regulated genes, which
provides the cell the flexibility to tune its transitions. Additionally, a single regulator can
be capable of regulating genes transcribed by different σ factors and σ factors regulate
genes subject to specific TFs. Summing all up, the cell is able to partition its genes by
alternative ways, either by relationships between different TFs, or in terms of classes of
genes of different σ factor promoters (Martinez-Antonio and Collado-Vides, 2003).

Although the bacterial regulation network alone is a very complex network, it can be
broken down into basic building blocks and ’network motifs’ can be defined (Shen-Orr
et al., 2002), revealing that there is only a small number of multi-level regulatory cascades
1.6. THE CIRCADIAN CLOCK OF CYANOBACTERIA IS UNIQUE 13

amongst the transcription factors (Madan Babu and Teichmann, 2003). A large part of
the network is composed of repeated appearances of three highly significant network mo-
tifs. This and related approaches help to define and to implement the basic computational
elements of biological networks generating models which may be supportive in gaining in-
sights into dynamic behaviours (Shen-Orr et al., 2002).

1.6 The circadian clock of cyanobacteria is unique

Daily oscillation mediated by an internal clock are one of the most fascinating biological
non-linear dynamics found within a single cell. Cyanobacteria are the simplest organ-
isms known to possess a true circadian clock, and in particular Synechococcus elongatus
is used as an instructive model system for circadian mechanisms in unicellular organisms
(Iwasaki and Kondo, 2004). The temporal coordination of internal biological processes
and adaptation to daily fluctuations in the environment play a critical role in survival
of diverse organisms, from bacteria to humans (Bell-Pedersen et al., 2005; Kucho et al.,
2005). Circadian clocks exist in all eukaryotes (except yeast) and were longtime thought
to be restricted to the eukaryotic kingdom. Today their existence has been demonstrated
for plants, animals and, among prokaryota, exclusively in cyanobacteria. The design prin-
ciples are conserved between these highly different groups of organisms, whereas their
components share no similarity. At the core of all circadian clocks is at least one internal
autonomous circadian oscillator containing positive and negative elements that form au-
toregulatory feedback loops (Young and Kay, 2001).

Defining properties of circadian rhythms include (Bell-Pedersen et al., 2005):

- a periodicity of about 24 hours, even in the absence of an environmental cycle

(free-running rhythm)

- the ability of the clock to be entrained in a time-dependent manner by envi-

ronmental stimuli

- the compensation of periodic length for changes in the natural environment,

in particular temperature compensation

Studies regarding the outright incompatibility of cyanobacterial nitrogen fixation and

oxygenic photosynthesis showed that temporal separation of disparate processes is ad-
vantageous for the cell and suggested an endogenous timing mechanism for cyanobacteria
(Golden et al., 1997). An authentic circadian clock has been confirmed in Synechococcus
elongatus PCC 7942 which fulfils all three criteria of a circadian process described above;
although PCC 7942 does not fix nitrogen.

In Synechococcus elongatus a pacemaker orchestrates a global rhythmic regulation of gene

expression and controls the timing of cell division (Golden and Canales, 2003; Johnson,
14 1. INTRODUCTION

2004), but the circadian timing-mechanism itself is independent of the cell cycle (Kondo
et al., 1997). The cyanobacterial pacemaker is based on the clock proteins KaiA, KaiB
and KaiC which are together with ATP sufficient to achieve a temperature-compensated
circadian rhythm of KaiC phosphorylation in vitro (Nakajima et al., 2005). The fun-
damental timekeeping mechanism involves interactions among clock proteins rather than
transcriptional control, culminating in the formation of a high-molecular-weight complex
during the night termed as the periodosome (Bell-Pedersen et al., 2005); see also Figure 1.8.

P Light
CikA

P
P P
P P P
P
CikR

KaiC
?

?
KaiA SasA
Late P
Receiver-like evening
domain
KaiB
Rhythmic output

P
SasR

Figure 1.8: The cyanobacterial periodosome model, reviewed by Bell-Pedersen et al. (2005). Environmen-
tal information, such as daylight, is transduced through the phosphorylation and activation of Circadian
input kinase A (CikA). CikA in turn phophorylates and activates its predicted binding partner, Circa-
dian input kinase R (CikR). Information is then transferred through protein-protein interactions to the
receiver-like domain of the circadian-clock protein KaiA. KaiA interacts with KaiC and stimulates au-
tophosphorylation of KaiC, which is hexameric. In the phosphorylated state, KaiC hexamers can form
a complex with other clock components. Synechococcus adaptive sensor A (SasA) joins the complex and
is thereby stimulated to phosphorylate its predicted binding partner, Synechococcus adaptive sensor R
(SasR). Phosphorylated, active SasR sends temporal information from the periodosome to the rest of the
cell to activate rhythmic gene expression, either directly or indirectly. Late in the evening, another protein,
KaiB, joins the periodosome introducing its disassembly and restarting the cycle. The modular events that
reactivate the cycle in constant environmental conditions have not been yet described (Bell-Pedersen et al.,
2005).

In a new model for the core clockwork in cyanobacteria rhythmic changes of chromosome
condensation underlie the rhythms of gene expression (Johnson, 2004). Recently, it was
suggested that the periodosome interacts directly with the bacterial nucleoid to affect su-
percoiling (Iwasaki and Kondo, 2004; Xu et al., 2003), but an experimental verification is
still missing. However, the supercoiling level in Chlamydomonas chloroplasts, evolutionary
a plant organelle derived from cyanobacteria, was found to be highest at the beginning and
lowest at the end of the light phase, indicating an increase in DNA supercoiling activity
1.6. THE CIRCADIAN CLOCK OF CYANOBACTERIA IS UNIQUE 15

at night (Salvador et al., 1998). Circadian changes of the chromosome topology in the
Synechococcus remain to be demonstrated.

Moreover, an intimate link to the metabolic status of the cell seems to exist. For example,
the light-dependent protein A contains redox-active iron-sulfur clusters and is, therefore,
sensitive to the redox state of the cell (Ivleva et al., 2005). LpdA co-purifies with perio-
dosome components and forms a complex with clock proteins in a circadian fashion (Ivleva
et al., 2005). It was shown that LdpA affects the absolute, as well as the light- and redox-
dependent, abundance of CikA, a key input pathway component, and furthermore the
absolute level of the clock protein KaiA. Thus, a novel input pathway to the circadian os-
cillator is suggested in which LdpA is a component of the clock protein complex that senses
a metabolic state of a cell for adjusting the period length according to light intensities, but
also to other environmental factors, such as nutrients and temperature (Ivleva et al., 2005).

Today, it is known that the cyanobacterial clock controls many physiological processes,
such as photosynthesis, amino acid uptake, carbohydrate synthesis, and the cell division
cycle (Golden et al., 1997; Iwasaki and Kondo, 2004). Although the entire genome seems
to be under clock control at the transcriptional level, different sets of genes are expressed
with distinct phase relationships, indicating that there are also other layers of control (Bell-
Pedersen et al., 2005). In some mutant backgrounds different periods can be observed for
different genes. Together, all the available data indicate that multiple circuits coexist in
the cell and that more than one oscillator is likely to function within the cyanobacterial
cell (Bell-Pedersen et al., 2005; Iwasaki and Kondo, 2004).

1.6.1 Roles of group 2 σ factors in circadian regulation

Cyanobacteria are unusual in having multiple, closely related, σ factors that are not es-
sential for cell viability (group 2) in addition to the essential, principal σ factor, RpoD1
(Imamura et al., 2003a). A model was suggested in which σ factors work as a consortium
to convey circadian information to downstream genes: Because each σ factor dissimilarly
affects transcription from specific cyanobacterial promoters, the cyanobacterial transcrip-
tional apparatus may oscillate in a circadian manner (oscillations in RpoD4 protein level
have been demonstrated) by changing the composition of the RNA polymerase holoen-
zyme for individual group 2 σ factors over the circadian cycle (Nair et al., 2002; Ditty
et al., 2003).

Inactivation of any of the four known group 2 σ factor genes (rpoD2, rpoD3, rpoD4 and
sigC ), either single or in pairs, alters circadian expression of the psbA promoter in Syne-
chococcus PCC 7942 (Nair et al., 2002). Two most striking examples are mutants lacking
one of the group 2 σ factors, SigC or RpoD3, which results in lengthening or shortening
the period of the gene expression rhythm of psbA without affecting the period of kaiBC
expression (Nair et al., 2002). These results reveal the possibility that multiple timing
circuits are coexisting at least in Synechococcus elongatus (Nair et al., 2002).

The group 2 σ factor RpoD3 of Synechococcus PCC 7942 was assigned to SigD in Syne-
chocystis PCC 6803 (Imamura et al., 2003b), which is light-induced and responsible for
16 1. INTRODUCTION

promoter recognition of the photosynthesis gene, psbA. Studies by Imamura et al. (2003a,b,
2004) in Synechocystis PCC 6803 revealed an antagonistic expression of the group 2 σ fac-
tors SigB and SigD. These are inversely regulated with respect to light/dark changes and
depend on the redox status of the photosynthetic electron transfer chain. Moreover, in sigB
and sigD knockout strains, the expression was significantly reduced of the dark-induced
lrtA as well as the light-induced psbA2/3 transcript, respectively. A possible model for
these observations includes repression of SigB expression by SigC under light conditions
(Imamura et al., 2003a). Furthermore, SigC contributes to global gene expression in the
stadium of post-exponential growth, involving the central signal transducer PII (encoded
by the nitrogen-related glnB gene) (Asayama et al., 2004).

Collectively, these observations and models raise the possibility that the cyanobacterial
circadian timing process is not absolutely dependent on the kai gene expression cycle
and that the "simple" cyanobacterial clock is more complex than was previously thought
(Iwasaki and Kondo, 2004).

In addition to the standard group 1 factor (σ 70 ) that is responsible for transcription

from a number of house-keeping promoters, all Prochlorococcus and Synechococcus strains
sequenced so far harbour several more group 2 σ factors. In Prochlorococcus strains
light/dark conditions induce cell synchrony (Jacquet et al., 2001). Distinct G1, S and G2
phases were found to characterise cell cycles of marine Synechococcus and Prochlorococ-
cus. It was suggested that cell division in marine Synechococcus is controlled by circadian
oscillators as it was described for Synechococcus elongatus PCC 7942 (Asato, 2003). For
the sequenced marine strains, Prochlorococcus Med4, SS120, MIT 9313 and Synechococcus
WH 8102, the existence of an internal autonomous circadian oscillator still awaits experi-
mental testing.

1.7 The new era of small RNA molecules - tiny but mighty
regulators
Small RNAs prove more influential than imagined: Short non-coding RNA molecules
known as microRNAs and short interfering RNAs (siRNA) may control more than one
third of our genes. This claim by Lewis et al. (2005) suggests that such RNA molecules
could play a role in almost every process from cell birth to cell death.
However, small non-protein-coding RNAs (ncRNAs) with important regulatory roles are
not confined to eukaryotes (Gottesman, 2005). Recent studies have shown that bacteria
also possess a significant number of regulatory ncRNAs. These are a heterogeneous group
of functional RNA molecules normally without a protein-coding function. They are fre-
quently smaller than 200 nucleotides in size, and act to regulate mRNA translation/decay
but can also bind to proteins and thereby modify protein function; for a recent review see
Gottesman (2004).
It has been known for many years that non-coding RNAs control replication and main-
tenance of prokaryotic extrachromosomal elements (Wagner and Simons, 1994; Gerhart
et al., 1998). A regulatory RNA, RNAIII, was found to play a critical role in Staphylo-
coccus aureus virulence; reviewed by Johansson and Cossart (2003). More recently, up to
1.7. THE NEW ERA OF SMALL RNA MOLECULES - TINY BUT MIGHTY REGULATORS 17

four different ncRNAs have been identified as crucial regulators in quorum sensing systems
of Vibrio species, including the regulation of virulence in V. cholerae (Lenz et al., 2004).
Moreover, ncRNAs are described as important factors in bacterial regulatory networks
that respond to environmental changes (Sledjeski et al., 1996; Altuvia et al., 1997). Ex-
amples are OxyS: oxidative stress, DsrA: cold shock, IstR: SOS response, RhyB: iron stress
and MicF: osmotic stress. All RNAs known so far are collected in the RFAM database
including sequence and folding information as well as alignments with sequences of related
species and software tools for the identification of new members (Griffiths-Jones et al.,
2005).

The bacterial small RNAs include two major classes. The largest family acts by base-
pairing with target mRNAs to modify mRNA translation or stability. For this mechanism,
an RNA chaperone protein, Hfq, seems to be essential, which is similar structurally and
functionally to eukaryotic Sm proteins: Hfq contains a conserved sequence motif, known
as the Sm1 motif, forms a doughnut-shaped structure, and has RNA binding specificity
very similar to the Sm proteins (Moller et al., 2002; Valentin-Hansen et al., 2004). It was
demonstrated in vitro for OxyS (Zhang et al., 1998, 2002) and for another ncRNA, Spot
42 (Moller et al., 2002), that they act by pairing with complementary sequences in their
mRNA targets (see also Fig. 1.9) and that Hfq is important for this pairing.

Figure 1.9: A model for fhlA-OxyS interaction, suggested by Argaman and Altuvia (2000). The Shine-
Dalgarno and the initiation codon AUG are underlined. The numbering of fhlA starts at the initiation
codon (AUG). The numbering of OxyS starts at the transcription start site. The arrows indicate point
mutations introduced by Argaman and Altuvia (2000). The fhlA nucleotides of the Rbs (5’ end stem-loop)
and the Cds (3’ end stem-loop) sites are in green and the complementary sites in OxyS are in purple
forming a so called kissing complex by pairing with each other.

The second major category of regulatory RNAs are those that act by binding to a protein
and modifying its activity. There are currently two known target proteins in E. coli regu-
18 1. INTRODUCTION

lated directly by ncRNA binding: CsrA, the carbon storage regulatory protein, regulated
by binding to CsrB and CsrC, two RNAs that inhibit CsrA (Weilbacher et al., 2003); and
RNA polymerase, changed in promoter selectivity by binding to the 6S RNA (Wassarman
and Storz, 2000).

Each ncRNA may regulate multiple target genes; for example OxyS acts as a global regu-
lator on more than 40 genes in E. coli, including such important regulatory genes as fhlA
(transcriptional regulator) and rpoS (alternative sigma factor) (Altuvia et al., 1998). The
converse situation has also been found: Three different ncRNAs, OxyS, DsrA and RprA,
converge on rpoS for regulation (Repoila et al., 2003).

1.7.1 How to find them?

Today, as a result of recent systematic searches, more than 70 ncRNAs are known in E.
coli, most of which had been overlooked by traditional genome analysis. Thus, they rep-
resent about 2 % of the number of protein-coding genes, at the moment. Within a very
short period, a variety of non-coding RNAs were found either by computational or exper-
imental genome-wide screens. However, computational detection of ncRNAs is far more
difficult than identification of protein-coding genes, because ncRNAs are typically short,
have widely varying motifs and are often characterised more by their secondary structure
than by their primary sequence (Huttenhofer et al., 2005). Some common properties are
needed to serve as a guide for computational studies.
The first global searches in E. coli were based on the observation that known small RNAs
mainly reside in intergenic regions and are often conserved in closely related species. Thus,
searches were successful outside of protein-coding genes (Wassarman et al., 2001) combined
with conservation of sequence characteristics expected for secondary structure (Rivas et al.,
2001). The basic idea implemented by QRNA (Rivas and Eddy, 2001) and some other
programs is that conserved structural RNA tends to show a pattern of compensatory mu-
tations consistent with the base-paired secondary structure (stem-loop).
Additionally, intergenic regions were scanned for promoters and rho-independent termina-
tors of transcription (a stem-loop followed by a row of Ts) with an orientation and spacing
consistent with a ncRNA (Argaman et al., 2001; Chen et al., 2002). Several other com-
putational studies exist, but not all have included experimental tests of their predictions
(e.g. ddbRNA by di Bernardo et al. (2003)).
The identification of ncRNA genes by experimental methods can be achieved by generating
specialised cDNA libraries of small transcripts (Vogel et al., 2003b; Kawano et al., 2005)
or microarray techniques (Wassarman et al., 2001; Tjaden et al., 2002). Direct cloning
of RNAs of a given size range may be useful for isolating ncRNAs expressed at a high
level under a given growth condition or in the case that microarrays are not available
(Gottesman, 2004). Small RNAs can also be identified by their association with a known
RNA-binding protein like Hfq (Zhang et al., 2003).
Another promising approach are genome-wide searches for binding sites of known classes
of regulatory proteins, which identified specifically regulated ncRNAs in Pseudomonas
aeroginosa and Vibrio cholerae (Wilderman et al., 2004; Lenz et al., 2004; Gottesman,
2005).
All the successful approaches used over the years were reviewed recently by Vogel and
1.7. THE NEW ERA OF SMALL RNA MOLECULES - TINY BUT MIGHTY REGULATORS 19

Sharma (2005),

- genome-wide searches based on the biocomputational prediction of ncRNA

genes

- global detection of non-coding transcripts using microarrays

- shotgun cloning of small RNAs

- co-purification with RNA-binding proteins, such as Hfq or CsrA / RsmA

- classical cloning of abundant small RNAs after size fractionation in polyacry-

lamide gels

providing a nice overview about the state of the art of RNA finding.

1.7.2 Are the newly identified ncRNAs functional?

Both computational and expression-based methods are productive in finding small reg-
ulatory RNAs, but each has its specific limitations. Many ncRNAs have been already
predicted and also verified, but for the majority of these RNAs little or nothing is known
about function. Thus, further experimental and computational approaches are needed
to determine whether these recently found transcription products are important parts of
regulatory networks and biochemical pathways or simply ’junk’ RNA transcribed as an
unplanned by-product in the cell (Huttenhofer et al., 2005).

1.7.3 Riboswitches: mRNA-enslaved ncRNAs

New findings indicate that riboswitches are robust genetic elements that are involved in
regulating fundamental metabolic processes in many organisms (for a review see Man-
dal and Breaker (2004)). Metabolite binding domains exist within several mRNAs that
serve as precision sensors for their corresponding targets. An allosteric rearrangement of
the mRNA structure (named riboswitch) is mediated by ligand binding, and results in
the modulation of gene expression without the obligate involvement of a protein factor
(Mandal et al., 2003); for an example see Figure 1.10. Numerous mRNAs in prokaryotes,
including cyanobacteria, carry complex folded domains, which are known as riboswitches,
within the non-coding portions of their polynucleotide chains.

TPP (THI element, Fig. 1.10) and cobalamin (B12-element (Griffiths-Jones et al., 2005))
are the two metabolite-binding riboswitches predicted for marine cyanobacteria, each with
not more than one count per genome. One might expect to identify additional cases for
genetic switches in these organisms by a more intensive, comparative search approach.
20 1. INTRODUCTION

Figure 1.10: Schematic representation by Winkler et al. (2002) of the proposed mechanism for TPP-
dependent deactivation of thiM translation. In the absence of TPP, the P8* pairing is formed between
the anti-SD element and the anti-anti-SD element. This conformation permits the SD sequence to interact
with the ribosome, and thus translation proceeds. In the presence of TPP (blue), the obligate formation
of the P1 stem sequesters a portion of the anti-anti-SD element, and therefore the complete P8 stem also
forms. This precludes ribosome access to the SD element, which inhibits translation. Complementary
sequence elements that form P1 and P8 are depicted in green and orange, respectively (Winkler et al.,
2002).

Two related and fascinating examples might be RNA thermosensors and T-box RNAs,
gene-control elements that share some characteristics with metabolite-binding riboswitches
(Mandal and Breaker, 2004). The 5’ untranslated region (5’ UTR) of the pfrA gene from
Listeria monocytogenes carries a hairpin structure that occludes the ribosome-binding
site. A shift in temperature from 30 to 37◦ C induces a fivefold increase in pfrA expression
(Johansson et al., 2002), presumably by melting local base-pairs and permitting access by
the ribosome. Thus, the RNA alone might serve as a molecular thermometer.
T-box RNAs are normally found in the 5’ UTR of genes that encode aminoacyl-tRNA
synthetases or related amino-acid-biosynthesis genes of gram-positive organisms. They
fold into a structure that selectively recognises a specific tRNA (Grundy and Henkin,
2003), and activate gene expression in the presence of non-aminoacylated tRNAs, thereby
boosting the expression of genes that are needed to maintain an adequate pool of charged
tRNAs. A scenario that can be assumed for marine cyanobacteria as well.

1.7.4 Are there regulatory RNAs in cyanobacteria?

Many of these versatile bacterial riboregulators found in enterobacteria use base-pairing
interactions to regulate the translation of target mRNAs. Because most of these antisense-
acting ncRNAs have only incomplete target complementarity, duplex formation frequently
depends on the activity of Hfq, an RNA chaperone. Only very recently, an hfq homologue
1.8. AIMS OF THIS WORK 21

was predicted in cyanobacterial genomes, including two of the strains used in this study,
Synechococcus WH 8102 and Prochlorococcus MIT 9313 (Valentin-Hansen et al., 2004). It
supports the idea that riboregulatory processes similar to those of enterobacteria should
exist in cyanobacteria as well.

At the begin of this work there was no information about the presence of regulatory RNAs
and their genes in marine cyanobacteria. Apart from rRNA and tRNA genes, only three
other well-characterised RNA genes were annotated by sequence similarity in each of the
four marine genomes, Prochlorococcus Med4, SS120, MIT 9313 and Synechococcus WH
8102. These encode the RNA components of RNAse P (M1 RNA), the signal recognition
particle (scRNA) and tmRNA (rnpB, ffs and ssrA, respectively). Although the Prochlo-
rococcus tmRNA has not been analysed experimentally so far, it was subject to several
in silico analyses, predicting it would consist of two separate molecules derived from a
common precursor (Gaudin et al., 2002; Keiler et al., 2000). Such a permuted gene struc-
ture producing a two-piece mature tmRNA results in a dramatically reduced number of
secondary structure elements. It remains unclear, if such a simplification in the structural
elements of this RNA species would bring any selective advantage. However, the ques-
tion raises, whether number and complexity of ncRNAs in these tiny, marine organisms
is generally reduced as observed for tmRNA and regulatory proteins. And if that will be
the case, what kind of ncRNAs might have escaped such an elimination and simplification
process?

1.8 Aims of this work

The global importance of cyanobacteria belonging to the two genera Prochlorococcus and
Synechococcus is indisputable these days. Environmental studies suggest that it is the spe-
cific niche adaptation into the complex and fluctuating oceanic environment, which has
shaped the physiology and underlying genetic information of these organisms, driven their
evolution and ecological success. Sequencing of multiple marine strains opened the great
opportunity to study the molecular background of this exceptional adaptation process in
a comparative genomics approach.

A successful comparative analysis, predicting conserved regulatory elements like DNA

binding sites or secondary structures, strongly depends on a well-chosen set of sequences
with reliable evolutionary distances. The marine genomes of Prochlorococcus Med4, SS120,
MIT 9313 and Synechococcus WH 8102 might represent just such a perfect set.
Thus, the major aim of the present study was to analyse these marine genomes compu-
tationally, whereby the resulting predictions should be tested directly in the laboratory.
Fortunately, all the four strains are growing in cell culture and for one marine strain, Syne-
chococcus WH 8102, a genetic manipulation system was described previously, providing
the minimum requirements for further biomolecular experiments.
In particular, experimental and computational promoter studies were aimed to get insights
into the transcriptional regulation of cyanobacterial genes. For this purpose, phylogenetic
footprinting algorithms has been successfully demonstrated to predict transcription fac-
tor binding sites within a given set of orthologous sequences. Additionally, the RACE
22 1. INTRODUCTION

technique appeared very useful, which has been shown to determine sites of transcription
initiation precisely.

Besides protein-coding genes, small non-coding RNAs possess a high regulatory potential,
but systematic searches for ncRNAs are still lacking for most eubacterial phyla outside
the enterobacteria. Thus, this work aimed at the question: Are small RNAs existing in
marine cyanobacteria and if they are, might it be possible to derive hints on their func-
tion? Recently, an effective method to score multiple alignments in terms of secondary
structure conservation was suggested (Washietl and Hofacker, 2004; Washietl et al., 2005).
Using a comparative genomics approach based on the published genome sequences, one
may predict candidates for ncRNAs in marine cyanobacteria. The expression of these can-
didate sequences needs to be tested under various growth and stress conditions that are
encountered in the natural environment. The conservation of ncRNAs between different
strains may provide information about an essential as well as a special ncRNA equipment
for each strain with prospect to its niche adaption.

Another fascinating phenomenon proven for freshwater cyanobacteria is a unique circa-

dian clock; otherwise found exclusively in eukaryotes but built of completely different
components. The most studied clock model strain Synechococcus elongatus is the closest
sequenced freshwater relative to the marine group - are they sharing a similar circadian
pacemaker?
Chapter 2

Materials and Methods

2.1 Experimental Part

2.1.1 Cultures, cultivation, plasmids
Cultivation of marine cyanobacteria
Cultures of Prochlorococcus and Synechococcus were grown in a chemically-defined artifi-
cial sea water medium (ASW): Prochlorococcus Med4, NATL2A-MIT and Synechococcus
WH 7803, RS9906, WH 8102. For cultures of Prochlorococcus SS120, MIT 9313, MIT 9312,
MIT 9211, MIT 9215, which were not growing well on the first medium, PRO99 medium
based on Atlantic seawater was used instead. The light conditions were 18 (Med4, 9312,
9215, WH 7803, RS9906 and WH 8102) or 10 (all other strains) µmol quanta m−2 s−1
white light at 23◦ C in a 12 h day - 12 h night cycle.
All cultures of marine cyanobacteria had originally been kindly provided by Dr. Frédéric
Partensky (CNRS, Roscoff, France) or Dr. Penny Chisholm (MIT, Cambridge, U.S.) and
existed at the begin of this work for at least two years in our laboratory.

Both media, ASW and PRO99, were prepared as described previously (Steglich, 2003;
Chisholm, 2006). The seawater (artificial sea water for ASW: Milli-Q water with 0.48 M
NaCl, 0.028 M MgSO4 , 0.027 M MgCl2 , 0.009 M KCl, 0.01 M CaCl2 ; Atlantic seawater for
PRO99) was sterile-filtered, autoclaved and allowed to cool overnight. Nutrient additions
were given to the filtered and autoclaved seawater, resulting in the final concentrations:
0.05 mM phosphate buffer, pH 7.5; 0.4 mM (NH4 )2 SO4 ; 1 mM HEPES, pH 7.5; 2 mM
NaHCO3 .
Metal additions were again the same for both media, ASW and PRO99, and minor modi-
fications resulted in the following final concentrations for trace metals in the medium: 1.2
mM EDTA; 1.2 mM FeCl3 ; 8 nM ZnCl2 ; 5 nM CoCl2 ; 90 nM MnCl2 ; 3 nM Na2 MoO4 ; 9
nM Na2 SeO3 ; 10 nM NiCl2 .

The use of 1 N HCl and Milli-Q washed and autoclaved glass-containers was essential
for successful preparation of media. All nutrient and metal additions were prepared sep-
arately as primary stocks, sterilised using 0.2 µm VacuCap 90 filters (PALL, Dreieich,
Germany) and stored at -20◦ C. The final media were stored at room temperature for up
to one month. All marine cyanobacteria were cultured in 40, 160 or 400 ml polycarbonate
24 2. MATERIALS AND METHODS

cell culture flasks (Nunc, Wiesbaden, Germany) as shown in Figure 2.1.

Figure 2.1: Different liquid cultures of Prochlorococcus or Synechococcus (left) and marine Synechococcus
WH 8102 on plate (right).

Environmental perturbations for Prochlorococcus Med4

Prochlorococcus Med4 was subjected to various changes in growth conditions by individu-

ally depleting nitrate, phosphate or iron in artificial seawater; a shift from approximately
10 µmol quanta m−2 s−1 white light into darkness or into 10 µmol quanta m−2 s−1 blue
light or into 50 µmol quanta m−2 s−1 daylight as high light condition, or the addition
of DCMU to a final concentration of 2 µM for the inhibition of photosynthetic electron
transport and the induction of severe oxidative stress; as well as temperature shifts to 15◦ C
or 30◦ C. Med4 cultures were concentrated ten-fold by centrifugation for 10 min at 9,000
rpm at 15-20◦ C and cell pellets were washed once with the corresponding depleted media
if necessary. The concentrated cultures were incubated for three hours at the respective
condition.

SN medium for marine Synechococcus

For strains of marine Synechococcus optimum growth was observed during cultivation on
liquid SN medium. This medium was described previously by Waterbury and Willey
(1988); here Atlantic seawater was the basis.
Solid SN medium (Waterbury and Willey, 1988) was prepared by using ultrapure, low-
melting point agarose (Invitrogen, Karlsruhe, Germany). For pour plating of single
colonies, marine Synechococcus strain WH 8102 was serially diluted in liquid SN medium
and 0.1 ml of the dilution was added to 35 ml of solid SN medium containing 0.6 %
(wt/vol) low-melting agarose at 37◦ C and poured immediately into a plastic petri dish
(Brahamsha, 1996). All plates were incubated at 23◦ C under 10 µmol quanta m−2 s−1
white light for 24 h and then moved to a higher intensity of 40 µmol quanta m−2 s−1
(Brahamsha, 1996). Colonies generally appeared from 10 to 14 days (Fig. 2.1), but it
took up to six weeks before mutant colonies became visible. If incubated for more than
two weeks, plates were wrapped in parafilm to avoid dry-out. Colonies were excised (at a
2.1. EXPERIMENTAL PART 25

diameter of about 1 mm) and transferred to liquid SN medium.

Cryopreservation
For cryopreservation of marine cyanobacteria, 50 ml cultures were spun for 15 min at
room temperature at 8,000 rpm. The pellet was resuspended in 1 ml original medium and
transferred into a cryogenic vial (Nunc, Wiesbaden, Germany) where 7.5 % DMSO (75
µl) was added. The vial was inverted several times and placed immediately into liquid
nitrogen and stored in a -80◦ C freezer.
Frozen cells were thawed in 37◦ C water bath until the half was liquid and transferred into
sterile medium to revitalise the culture.

Other cyanobacteria and RNA samples

Cells of Gloeobacter violaceus PCC 7421 and Synechococcus PCC 7942 were ordered from
Pasteur Culture Collection, Paris, France. Synechocystis PCC 6803 and Anabaena PCC
7120 were kindly provided by Dr. Annegret Wilde, Institute for Biology/Biochemistry,
HU Berlin, Germany. These strains were grown in BG 11 medium (Rippka et al., 1979)
under 30 to 40 µmol quanta m−2 s−1 white light at 23◦ C. For all other cultures, cell pellets
or RNA samples were kindly provided by:

Dr. Jan Kern, Max-Volmer-Laboratory for Biophysical Chemistry, TU Berlin,

Germany: Thermosynechococcus elongatus BP-1;

Holger Jenke-Kodama, Institute for Biology / Genetics, HU Berlin, Germany:

Nostoc punctiforme ATCC 29133;

Dr. Elke Dittmann, Institute for Biology / Genetics, HU Berlin, Germany:

Microcystis PCC 7806.

Dr. Debbie Lindell, MIT, Cambridge, U.S.: phage P-SSP7 RNA

Dr. Julia Holtzendorff, CNRS, Roscoff, France: RNA of synchronised Prochlo-

rococcus Med4 culture

For Synechocystis PCC 6803 the accumulation of different transcripts from the isiAB re-
gion was studied under iron limitation and high light conditions. Therefore, liquid cultures
of Synechocystis PCC 6803 were grown at 30◦ C in BG 11 medium (Rippka et al., 1979) un-
der iron replete conditions and continuous illumination with white light of 50 µmol quanta
m−2 s−1 to the logarithmic phase of growth. For high light conditions, cell cultures were
exposed to white light of 170 µmol quanta m−2 s−1 . In parallel, logarithmic cell cultures
were washed three times in iron-free media and grown further in iron deplete conditions
26 2. MATERIALS AND METHODS

for several days.

Escherichia coli strains and plasmids

E. coli strain TOP 10 (Invitrogen, Karlsruhe, Germany) cells were used for preparation of
eletrocompetent cells, transformation, blue-white colony screening and plasmid amplifica-
tion according to standard protocols (Sambrook and Russell, 2001). Thereby, the pDrive
cloning vector (QIAGEN, Hilden, Germany) was used for highly efficient cloning of PCR
products.

The pUC19 vector (Fermentas, St. Leon-Rot, Germany) is a small, high-copy number E.
coli plasmid which served here as a molecular marker after digestion with MspI (HpaII).

For genetic manipulation of marine cyanobacteria via conjugative biparental mating the
plasmid pMUT100 (constructed by cloning the kanamycin-resistance cassette isolated from
pUC4K into the Pst I site of pBR322 vector) as well as the helper system, E. coli MC1061
(pRL528, pRK24), were kindly provided by Dr. Martin Ostrowski, University of Warwick,
GB. All these bacterial strains and plasmids needed for transformation were described pre-
viously by Brahamsha (1996).

2.1.2 DNA and RNA oligos

All DNA oligos were ordered at Metabion, Martinsried, Germany or Sigma-Aldrich, Stein-
heim, Germany. Their Tm and putative secondary structures were determined by the
Oligonucleotide Properties Calculator (Cao and Kibbe, 2006). RNA oligos were delivered
by Metabion, Martinsried, Germany. Each oligo is described below in the corresponding
subsection.

2.1.3 Genetic manipulation of marine Synechococcus

Transconjugants of Synechococcus WH 8102 were obtained via conjugation, based on bi-

parental matings as suggested by Brahamsha (1996). Here, the original protocol for con-
jugation was followed (Brahamsha, 1996), with the exception of the final plating step:
After conjugation, the mating mixtures were cut out of the agarose with a sterile spatula
and transferred into liquid SN medium without antibiotic for three days. After this addi-
tional incubation step, kanamycin was added (15 µg per ml first and up to 25 µg per ml
after about one week of adaptation). At the earliest possible moment, mutant growth was
observed on liquid medium with antibiotic, they were plated to obtain single colonies as
described above.
2.1. EXPERIMENTAL PART 27

2.1.4 Isolation of nucleic acids

Isolation of plasmid DNA from Escherichia coli
Plasmid DNA was isolated using the QIAprep Spin Miniprep Kit (tip 20 and tip 100; QIA-
GEN, Hilden, Germany) following the manufacturers instructions. The low-copy plasmid
pMUT100 needed a special procedure described by QIAprep Midiprep for low-copy plas-
mids and cosmids in the manual.

Isolation of Prochlorococcus and Synechococcus DNA

The genomic DNA was purified as previously described (Franche and Damerval, 1988). A
culture pellet from 400 ml original culture was resuspended in 2 ml ice-cold SET buffer.
Addition of 1/4 vol 0.5 M EDTA and 20 % SDS (final concentration of 100 mM and 2
%, respectively) and approximately 0.5 mg proteinase K (Fermentas, St. Leon-Rot, Ger-
many) resulted in cell lysis after incubation at 50◦ C for 2 to 16 h with gentle agitation.
DNA was isolated by standard phenol-chloroform-extraction.

SET: 25 % sucrose, 50 mM Tris-HCl pH 7.5, 1 mM EDTA

Isolation of RNA
The cells were harvested by centrifugation for 15 min at 8,000 rpm and 4◦ C. Total RNA
of marine cyanobacteria was isolated as previously described (Garcia-Fernandez et al.,
1998). The pellet was resuspended in RNA resuspension buffer (0.5 ml for 400 ml cul-
ture), quick-frozen in liquid nitrogen and stored at -80◦ C until extraction. The frozen cell
pellets were thawed under addition of 3 vol Z6 RNA extraction buffer (Logemann et al.,
1987) for 30 to 60 min at room temperature and occasional shaking by hand. Acidic
phenol (Roth, Karlsruhe, Germany) was added (0.5 vol) and the cells were incubated for
another 60 min at room temperature, followed by phase extraction with 0.5 vol of chlo-
roform:isoamylalcohol (24:1). After shaking, the phases were separated by centrifugation
for 10 min at maximum speed. The aqueous phase was extracted then first with 1 vol
phenol:chloroform:isoamylalcohol (25:24:1) and then with 1 vol chloroform:isoamylalcohol
(24:1). Finally, the aqueous phase was transferred into a fresh tube, and the RNA was pre-
cipitated with 1 vol isopropanol for at -20◦ C overnight. After centrifugation at maximum
speed for 30 min and 4◦ C the pellet was washed with 70 % ethanol, dried for about 10
min, resuspended in water and re-precipitated with 3 vol EtOH:NaOAc (30:1) overnight
at -20◦ C. The RNA was pelleted again by 30 min of centrifugation, washed, dried and
dissolved in water.
The lysis conditions for MIT 9313 and WH 8102 were modified as these strains gave poor
RNA yields by the standard procedure. The resuspended cells from these strains were
homogenised in Z6 buffer (Logemann et al., 1987) by several freeze-thaw cycles using liq-
uid nitrogen over a time of 30 min, followed by the addition of 1 vol of acidic phenol and
incubation at 60◦ C for another 30 min.
28 2. MATERIALS AND METHODS

RNA resuspension buffer: 10 mM sodium acetate, 200 mM sucrose, 5 mM EDTA

Z6 buffer: 8 M guanidinium hydrochloride, 50 mM β-mercaptoethanol, 20 mM EDTA, 20

mM MES, pH 7 (NaOH) (Logemann et al., 1987)

NaOAc: 3 M sodium acetate, pH 4.8

To determine RNA stability in vivo, cells were treated with rifampicin, 200 µg/ml
(SIGMA, Munich, Germany) and filtered rapidly (i.e. within 60 s) through Supor 0.45
µm membrane filters (PALL, Dreieich, Germany) at different time points after treatment.
Filters were submersed in RNA resuspension buffer and quick-frozen in liquid nitrogen.
RNA was isolated by dissolving the filter in acidic phenol at 60◦ C followed by the standard
phenol-chloroform-extraction as described above.

An alternative protocol was used for the rapid isolation of many RNA samples in paral-
lel. Synechococcus and Synechocystis cell pellets were resuspended in Trizol (Invitrogen,
Karlsruhe, Germany) and frozen at -20◦ C. The tubes were transferred directly from the
freezer to a 65◦ C waterbath for 10 to 15 min, vortexing from time to time for lysis. The
phase extraction was done in phase lock gel tubes (Eppendorf, Wesseling-Berzdorf, Ger-
many) by adding 0.2 ml chloroform per ml of Trizol and by shaking for 15 s and resting
at room temperature for 3 min. The following centrifugation step was lasting for 12 min
at maximum speed which separated the phases. The upper aqueous phase which contains
the RNA was transferred to a new tube and precipitated by adding 0.5 ml of isopropanol
per ml Trizol used in the initial homogenisation.

Cyanobacterial cells such as Nostoc punctiforme, harbouring a very robust cell wall, were
disrupted by adding an equal volume of glass beads, 33.3 µl 20 % SDS solution and 583
µl acidic phenol to 500 µl concentrated cell solution, following several cycles of vigorous
agitation, freezing in liquid nitrogen and thawing in a waterbath. The centrifugation of
the mixture for 15 min at maximal speed at 4◦ C yielded an upper aqueous phase which
could be cleaned up by standard phenol-chloroform-extraction as described above.

Determination of concentration of nucleic acids

The concentration of nucleic acids was determined spectrophotometrically by measuring

the extinction (E) at the wavelength of 260 nm. An optical density (OD260nm ) of 1 corre-
sponds to 50 µg/ml dsDNA or 40 µg/ml ssRNA and 30 µg/ml of oligonucleotides. Proteins
absorb light of a wavelength of 280 nm, therefore the ratio of E260:E280 can be used as an
indication of the purity of the DNA. The ratios should be between 1.8 and 2. The qual-
ity and quantity of nucleic acids was further examined optically in an ethidium bromide
stained agarose gel.
2.1. EXPERIMENTAL PART 29

2.1.5 DNA gel electrophoresis, Restriction, Ligation

Double stranded DNA molecules of 80 bp to 10 kb were separated on 1.8 to 1.0 % agarose
gels containing ethidium bromide (0.2 µg/ml gel) in 1 x TAE electrophoresis buffer after
addition of 1/6 volume of 6 x DNA loading buffer under a field strength of 5 to 10 volt/cm.
For DNA fragments smaller than 80 bp, 3 to 4 % NuSieve GTG agarose (Cambrex Bio
Science, Verviers, Belgium) could resolve fragments differing by as little as 1 bp. After
migration, the DNA was visualised on an UV transilluminator, gel imager (BIO-RAD,
Munich, Germany). The standard markers pUC19/MspI (HpaII) and λ/Pst I were used
to compare the sizes of the loaded DNA molecules.

6 x DNA loading buffer: 50 % (w/v) glycerol, 1 mM EDTA (pH 8.0), 0.005 % bromphenol
blue and 0.005 % xylene cyanol

50 x TAE buffer: 2 M Tris, 2 M boric acid, 100 mM EDTA, pH 8.3 (Sambrook and Russell,
2001)

0.1 to 10 µg DNA were incubated with 5 to 10 U restriction enzyme (Fermentas, St. Leon-
Rot, Germany) in a final volume of 10 to 40 µl using the provided buffers at the optimal
temperature for the enzyme.

100 ng linearised vector was incubated with a 3 fold molar excess of insert, 1 x T4 ligase
buffer and 1 U T4 DNA ligase (Fermentas, St. Leon-Rot, Germany) in a final volume of
10 µl at 4◦ C overnight, or 3 hours at room temperature, or for 1 hour at 37◦ C.

2.1.6 RNA gel electrophoresis, Northern blotting and Hybridisation

Agarose gel electrophoresis and capillary blotting
Total RNA was separated on denaturating 1.5 % agarose gels. The agarose was first dis-
solved in boiling water and cooled down to 60◦ C until addition of 0.1 vol 10 x MEN-Puffer
and 16 % (vol/vol) formaldehyde. Prior to starting the gel, samples were mixed with 1
vol RNA loading buffer and denatured for 10 min at 65◦ C and kept on ice until loading.
Migration occurred under a field strength of 10 volt/cm in 1 x MEN as electrophoresis
buffer. For size-comparison, the standard RNA ladders High Range and Low Range (Fer-
mentas, St. Leon-Rot, Germany) were included on the gel.

2 x RNA loading buffer: 95 % deionised formamide, 0.025 % (vol/vol) SDS, 0.5 mM

EDTA, 0.4 % ethidium bromide

10 x MEN: 200 mM MOPS, 50 mM NaOAc, 10 mM EDTA, pH 7 (Sambrook and Russell,

2001)

The gel-separated RNA samples were blotted with 10 x SSC buffer to a Hybond-N+
membrane (Amersham, Freiburg, Germany) by capillary blotting; UV-crosslinking and
30 2. MATERIALS AND METHODS

stripping (hot SDS procedure for re-use) as described in the manual of the membrane.

20 x SSC buffer: 3.0 M NaCl, 0.3 M sodium acetate, pH 7 (Sambrook and Russell, 2001)

PAA gel electrophoresis and wet blotting

Total RNA was separated in 10 % polyacrylamide-urea gels, 15 x 20 cm, in a vertical
gel electrophoresis apparatus, Protean II (BIO-RAD, Munich, Germany). Electrophoresis
conditions were about 5 volt/cm field strength for about 20 hours in 1 x TBE. Polyacry-
lamide gels were stained with ethidium bromide (0.3 µg/ml) in 1 x TBE buffer, rinsed
with water and analysed with a Lumi-Imager F1 system (Roche, Mannheim, Germany).
Transcript sizes were determined by correlation to MspI-digested DNA of plasmid pUC19
(Fermentas, St. Leon-Rot, Germany), which was radiolabelled like oligonucleotide probes,
see below. Gel-separated RNAs were transferred to Hybond-N+ membranes (Amersham,
Freiburg, Germany) by electroblotting in a wet blot tank (BIO-RAD, Munich, Germany)
overnight, following the manufacturers instructions; UV-crosslinking and stripping (hot
SDS procedure for re-use) as described in the manual of the membrane.
Three different ways were used to verify that the same amounts of RNA samples were
loaded in Northern blots: first, by measurement of RNA concentrations; second, by direct
comparison of rRNA band intensities after staining by ethidium bromide; and third, by
control hybridisations using the 5S rRNA as an internal standard.

loading buffer 2 x RPA: 98 % deionised formamide, 2 mM EDTA, 0.1 % xylene cyanole,

0.1 % brom phenol blue

10 x TBE buffer: 8.9 M Tris, 8.9 M boric acid, 200 mM EDTA (Sambrook and Russell,
2001)

10 % PAA urea gel: 10 % Rotiphorese acrylamide/bisacrylamide stock solution 40 %

(Roth, Karlsruhe, Germany), 8.3 M urea, 1 x TBE, APS, TEMED (Sambrook and Russell,
2001)

Hybridisation
Following prehybridisation for at least 10 min in prehybridisation buffer at 45◦ C, oligonu-
cleotide probes, which were 5’ labelled by polynucleotide kinase (Fermentas, St. Leon-Rot,
Germany) with 30 µCi γ 32 P-ATP (Amersham, Freiburg, Germany), were added and usu-
ally hybridised at 52◦ C (depending on Tm of the oligo) for at least four hours. All DNA
oligonucleotides are listed in Tables 2.1 and 2.2. The membranes were washed in washing
solution I at 45◦ C for 10 min; washing solution II at 45◦ C for 5 min; and shortly in washing
solution III at ambient temperature. Signals were detected and analysed on a Personal
Molecular Imager FX system with Quantity One software (BIO-RAD, Munich, Germany).
2.1. EXPERIMENTAL PART 31

name sequence (5’ to 3’) hybridisation to

ScoccusREV GCGACGCCGTTTTACCT 6Sa RNA in Thermosyn., Nostoc punctif.
and PCC 7942
PCC6803REV CACCACGCCGTTTTACCT 6Sa RNA in PCC 6803
NosGloeoREV CGCAACGCCGTTTTACCT 6Sa RNA in PCC 7120 and Gloeobacter
isrRrev GGCAGGCAAAGCCATGTATTGGGGG IsrR in PCC 6803
PMM3822n TCCCTGGTATGACTCGAACTGC mRNA of PMM3822n in Med4
5S GGCATTGAGCTATTTTCTCAGGGGGCT 5S rRNA
t_ser CGGAGAGGGAGGGATTCGAAC tRNA-Ser
tm5 CGGAATCGAACCGCTGTCCGA tmRNA (5’end)
tm3MITWH TGACGGGAGAAACGAACGATGTTGT tmRNA (3’end) in MIT9313 and WH8102
tm3MS GCTGTTTGACGGGAGAAACTAACGAT tmRNA (3’end) in Med4 and SS120
sc CCTTCCGRTCCTGACCAGRTTT scRNA
y1M CGATTTGGTGTGTGAGGAGTATGG Yfr1 in Med4
y1MIT CGGTGTGTGTGAGGAGTGTGT Yfr1 in MIT9313
y1WH GGGGTGTGAGGAGTGTGTGG Yfr1 in WH8102
y_genM CCACTGTTTCAGTAAAACCTCTCCTAC Yfr2-Yfr5 in Med4
y2aM ATAGAGGTCTTTCCTTGGTTTAA Yfr2 Med4
y3aM ATTTCTAGAGATCTTTATATAGTTTTACT Yfr3 Med4
y4aM TCTAGAGGTCTTTTTCTGGTTTT Yfr4 Med4
y5aM AAGGGGCCTAAATTGGTTCTA Yfr5 Med4
y2to5M GTGTTTCCTTGTTTCCACTG Yfr2-Yfr5 in Med4
y2S GTGTTTCCTAGTTTCCACTTTT Yfr2 in SS120
y2MIT GTGTTTCCTTGTTTCCACTC Yfr2 in MIT9313
y6MS GTTGAACGTTTTRGTAGCKGTTGCTAC Yfr6 in Med4 and SS120
y7MS CCTCAAGTCGAAAAAGAGTCAGATCAGA Yfr7 Med4 and SS120
y7MS_2 GTCGAAAAAGAGTCAGATCAGAGCAC Yfr7 Med4 and SS120
y7MIT9312 GTCGAAAGAAAGTCAGATCAGAGCAC Yfr7 MIT9312
y7MITWH TCGAAAGAGAGTCAGATCAGAGCAC Yfr7 MIT9313, WH8102 and WH7803

Table 2.1: DNA oligos used for PNK labelling and hybridisation to Northern blots.

For hybridisation of mRNAs, the HexaLabel DNA labelling kit was used (Fermentas, St.
Leon-Rot, Germany) by following the manufacturers instructions. Therefore, the labelled
triphosphate α32 P-dCTP (Amersham, Freiburg, Germany) was chosen. The DNA tem-
plate (about 350 bp in length) for the radioactive DNA labelling was obtained by standard
PCR, described below. The steps and solutions of the hybridisation procedure were the
same as for oligonucleotide probes, described above. Only hybridisation and washing tem-
perature were slightly increased to about 55◦ C and 50◦ C, respectively.

prehybridisation buffer: 50 % deionised formamide, 7 % SDS, 250 mM NaCl, 120 mM

Na-PO4 , pH 7.2

washing solution I: 2 x SSC, 1 % SDS

washing solution II: 1 x SSC, 0.5 % SDS

washing solution III: 0.1 x SSC, 0.1 % SDS

32 2. MATERIALS AND METHODS

name sequence (5’ to 3’) no signal to

ScoccusFW AGGTAAAACGGCGTCGC 6Sa RNA in Thermosyn., Nostoc punctif.
and PCC7942
PCC6803FW AGGTAAAACGGCGTGGTG 6Sa RNA in PCC6803
NosGloeoFW AGGTAAAACGGCGTTGCG 6Sa RNA in PCC7120 and Gloeobacter
isrRfw GGATCTAATGTTGGCTGACTGACTAG putative complement IsrR-sequence in
GCG PCC6803
y1S TTCTGCAGCAATCTAAATTTTTAAGA guaB-trxA intergenic region for the pres-
GAAGAAAAATAA ence of Yfr1 in SS120
ID273MSt44rev CCCGAAACAGTCAGATGTG putative t44 RNA in Med4 and SS120
ID273WHt44rev CCCGAAACGGGCAGG putative t44RNA in WH8102
ID273MITt44rev CCCGAAACGGACAGGT putative t44RNA in MIT9313
ID217Mrev CAGGGGTTGATGGGTC cand. RNA CLID217 in Med4
ID217Mfw GACCCATCAACCCCTG cand. RNA CLID217 in Med4
ID102Mfw CCACAGAAACTACTACTACTATATAA cand. RNA CLID102 in Med4
ATTCATTATTAG
ID102Mrev CTAATAATGAATTTATATAGTAGTAG cand. RNA CLID102 in Med4
TAGTTTCTGTGG
ID185Mfw AGATCGTATTCATCCATGGTTAAGTA cand. RNA CLID185 in Med4
AAGTTAAA
ID185Mrev TTTAACTTTACTTAACCATGGATGAA cand. RNA CLID185 in Med4
TACGATCT
ID66Mfw TATGAGTGAATATGATAATGAATATC cand. RNA CLID66 in Med4
AATAGCAATATCAA
ID66Mrev TTGATATTGCTATTGATATTCATTAT cand. RNA CLID66 in Med4
CATATTCACTCATA
ID194Mfw GTAGGAGAGGTTTTACTGAAACAGTGG cand. RNA CLID194 in Med4
ID257fw CCCCATACTCCTCACACAC cand. RNA CLID257 in Med4
ID53Sfw GTAGCAACAGCTACTAAAACGTTCAAC cand. RNA CLID53 in SS120
ID53Mfw GCAACCGCTACCAAAACGT cand. RNA CLID53 in Med4
ID51Sfw GTCATGCAACTGGAAATGGATCATTAC cand. RNA CLID51 in SS120
ID51Mfw GTAATTGGAAACGAAAGCCCA cand. RNA CLID51 in Med4

Table 2.2: DNA oligos used for PNK labelling and hybridisation to Northern blots, which did not yield
signals.

2.1.7 Polymerase chain reaction (PCR)

DNA fragments were amplified by Taq DNA Polymerase (QIAGEN, Hilden, Germany)
in a standard PCR. A typical PCR cycle consisted of an initial denaturation at 93◦ C for
2-3 min followed by 34 three-step-cycles of denaturation at 93◦ C for 10 s, annealing at
about 60◦ C (depending on the primer set) for 30 s and elongation at 72◦ C for about 20
s (depending on the length of the amplified fragment). The final extension at 72◦ C for 5
min terminated the PCR. A list of all used PCR primers is given in Table 2.3.

The primer set, pDriveFW and pDriveREV, anneals on vector pDrive and was utilised for
amplification of the ligated insert, which was mainly used for screening of E. coli colonies
(’colony’ PCR).

For PCR systems that did not work well under standard conditions, a special protocol
was adapted, described in the manual, including Q-solution (QIAGEN, Hilden, Germany)
and an elongation at 72◦ C for 1 min per 1 kb amplified fragment. The Q-solution changes
2.1. EXPERIMENTAL PART 33

DNA oligo name sequence (5’ to 3’) organism/plasmid

pDriveFW ACGACGTTGTAAAACGACGG pDrive
pDriveREV TTCACACAGGAAACAGCTATGAC
WHYfr1suiSTARTfw GGTATGGCGACACCAAGCCCTGTC Synechococcus WH8102
WHYfr1suiENDrev TGGGCATTGCCCGCACACTCTTCACA
WHYfr7suiSTARTfw AATCCCCACTTTGTTTCCCCTTCTGGTTCTTC
WHYfr7suiENDrev TGGAACAGAGTCAGCTTCAAAAGGTCGTCG
WHYfr7suiIN197fw GCCCGATGACGCTCAGGTTGCTTG
WHYfr7suiIN177rev CAAGCAACCTGAGCGTCATCGGGC
Yfr7colonyPCRfw ATCCTGAACTGCTGTGTTGGTGACGGC
Yfr7colonyPCRrev TAAGAGGTCGCAGCCAGACGCTTCAAC
Kan782FW CGCCTGAGCGAGACGAAATACGCG pMUT100, pUC4K
Kan1541REV ACCTTCTTCACGAGGCAGACCTCAGC
isiAsondeFW CTGGGGCTTTGTTTCATACC Synechocystis PCC6803
isiAsondeREV TAAAATTCTGGCGGATAGGC
isi-5utr-fw TTGGGCGATCGCCAAAAATC (Vinnemeier et al., 1998)
isi-5utr-rev CTCTGTCCACCCGAGACCTA (Vinnemeier et al., 1998)

Table 2.3: PCR primer list.

the melting behaviour of DNA and can improve the PCR result.

2.1.8 Sequencing of DNA

The ’colony PCR’ fragments were purified on QIAquick spin columns (QIAGEN, Hilden,
Germany) and prepared for sequencing by adding water and 10 pMol primer pDriveREV.
Sequencing was performed by Dr. Martin Meixner (DLMBC, Berlin, Germany).

2.1.9 Rapid amplification of cDNA ends (RACE)

For RACE experiments, the RNAs were digested with RNAse-free DNAse I (Ambion,
Austin, Texas USA). 30 µg RNA were incubated for 15 min with 6 units (3 µl) DNase I, 1
x DNase buffer and 40 units (1 µl) RNase inhibitor (Fermentas, St. Leon-Rot, Germany)
at 37◦ C in a final volume of 40 µl. The DNase I was inactivated by incubation for 3 min at
room temperature with 4 µl of the provided glassbeads-based inactivation reagent. After
centrifugation, the RNA remained in the aqueous phase and was transferred to a fresh
tube.

Mapping of RNA ends was performed by rapid amplification of cDNA ends as described
previously (Bensing et al., 1996; Argaman et al., 2001) to determine experimentally the
sites of transcription initiation and termination in cyanobacteria.

5’ RACE
Primary transcripts in bacteria carry a 5’ triphosphate, which can be cleaved specifically by
tobacco acid pyrophosphatase, TAP (Epicentre, Madison, Wisconsin USA). The resulting
5’ monophosphate was subsequently ligated, using T4 RNA ligase (Epicentre, Madison,
34 2. MATERIALS AND METHODS

gene RT primer (5’ to 3’) PCR primer (5’ to 3’)

isrR6803 GGGCCGGAGCTCTAC GGCAGGCAAAGCCATGTATTGGGGG
CCCATTTTGCCAGCATCAGTAGCCTAGAAG
ssaA6803 CACCACGCCGTTTTACCT GGTGTAACCAGGGGGGAGTCATCAATTC
CTGCCATTATAACCCCATCCTTCGGCG
ssaA7120 CGCAACGCCGTTTTACCT CCAGGTGGGTACCGATTTTCCTCGTTTAAAG
CTCACTCATTATAGCTTTCAAAAACAGGTTCTACC
yfr1Med ACAGACTTAAAAAAGCCCGATAA GCGATTTGGTGTGTGAGGAGTATGGGG
yfr2Med ATAGAGGTCTTTCCTTGGTTTAA CTAAATCAAGTGTTTCCTTGTTTCCACTGTTTCAG
yfr3Med ATTTCTAGAGATCTTTATATAGTT AACCAAGTGTTTCCTTGTTTCCACTGTTTCAGTAAAA
TTACT
yfr4Med TCTAGAGGTCTTTTTCTGGTTTT CCGAATCAAGTGTTTCCTTGTTTCCACTGTTTC
yfr5Med AAGGGGCCTAAATTGGTTCTA CGAGTGTTTCCTTGTTTCCACTGATTTAATAAAACC
yfr6Med CTGTATGCTGTTTTTTTGCAGC GTTTTGGACGTATTTTTGTCCTCTGTAGATAAGTGTCATC
yfr6SS120 CATGCTTCTTTTCAACTACTGC GTTGAGTGTACTTTTGACCTCTGTAGGTTAGTGAC
yfr7Med AAGCCAGAAGAGGAGTCAA GCTTTCGTTTCCAATTACAGGATTGGTTTACGTCTACAT
CACATAAAGAAGAACTTGACCAAAGATCCAGTATGTAAATA
yfr7SS120 AAGCCAGAAGAGGAGTCAA TGATCCATTTCCAGTTGCATGACTGGTTTACTTCAC
CTTAAAAGGTCATCACCAACAAAACAGTTCTCTTGCATA

Table 2.4: DNA oligos used for 5’ RACE experiments analysing ncRNAs of Synechocystis PCC 6803,
Anabaena PCC 7120, Prochlorococcus Med4 and SS120.

Wisconsin USA), to the 3’ hydroxyl group of a RNA-oligonucleotide (5’ adaptor: GAU

AUG CGC GAA UUC CUG UAG AAC GAA CAC UAG AAG AAA), which was fol-
lowed by reverse transcription with a gene specific oligo (placed within the first 200 nt
downstream a start codon) and PCR amplification with a 5’ adaptor (ATA TGC GCG
AAT TCC TGT AGA ACG AAC ACT AG) and a nested gene specific primer (listed in
Tab. 2.4, 2.6 and 2.5).

M TAP+ TAP- U CIP

psbD SS120

Figure 2.2: PCR result of 5’ RACE for psbD in Prochlorococcus SS120. A single prominent band
was obtained with TAP-treated RNA (TAP+). The presence of a weaker band of the same size in the
control reaction (TAP-) indicates that a minor proportion of the primary transcripts were present in 5’
monophosphate form in the original transcript population. Additional controls are RNA without any
treatment (U) and 5’ dephosphorylated RNA using calf intestine phosphatase (CIP). The standard marker
pUC19/MspI (M) was used to compare the sizes of the loaded DNA molecules.

TAP-treatment is expected to yield a specific or at least strongly enhanced signal (lane

TAP+ in Fig. 2.2) for primary transcripts in the amplification step, as compared to
untreated RNA samples (lane TAP- in Fig. 2.2). Because of this selectivity for newly
initiated transcripts among the pool of 5’ ends, and the small amounts of RNA material
required, this technique was considered more suitable for this approach than traditional
primer extension analysis.

Altogether, three controls were included in all 5’ RACE experiments: RNA mock-treated
2.1. EXPERIMENTAL PART 35

gene RT primer (5’ to 3’) PCR primer (5’ to 3’)

P001 TCTCCATAATTGACCCGCTT CTTAATAAAGTCAGTCAGTCCATCCCAGTCAGT
P003 TCTCCCCCTTACCCATAG CCTTCTTACCGAATTTTCCTATGGTGTACTTAG
P004 TTGTGCTTGGGTTTCTCTCA GGAATCGTTTACAGGGAATAACTCAGGCTCG
P013 TCTTTACGTGGGGAGAATAG GGTTATCTTTGCAGTTATAGCTCCTTGCGATTCTG
P029 GCCTTGTCAGCATTACCC GGAAGGAACTTGTCATACGACCTGTGTAG
PMM1500 GATGAGATTTGACACCCTC CCATCCATATCCTTGCTCTGGGCCG
PMM1501 GGACCTAGATCTGATACATG CTATAAAGGCAGCATCAATACCTGGTAGGAC
PMM0684 TTGTAAATAAGAGATGGGTATAAAC CTGCCTTGCGGGTTAAGTATCCAGCC
PMM0819 CATTTATCCCAGGCTTTTTCC TGACTTGATGACTTGATTGCTGAGGGCTC

Table 2.5: DNA oligos used for 5’ RACE experiments for cyanophage P-SSP7 infecting Prochlorococcus
Med4.

in TAP buffer but omitting TAP (TAP-), RNA without any treatment (U) as well as 5’
dephosphorylated RNA (CIP) using calf intestine phosphatase (Fermentas, St. Leon-Rot,
Germany), see Figure 2.2.

3’ RACE
3’ hydroxyl ends of the RNA samples, either 5’ blocked by 5’ adaptor ligation or by dephos-
phorylation with calf intestine phosphatase (Fermentas, St. Leon-Rot, Germany), were
ligated to the 5’ monophosphate end of another RNA-oligonucleotide (3’ adaptor: pAAG
AUG AAU GCA ACA CUU CUG UAC GAC UAG AGC A ddC). Reverse transcription
of the ligated RNA molecules was initiated with a 3’ adaptor-specific oligo (GTG CTC
TAG TCG TAC AGA A), followed by PCR amplification, each with a gene specific (for
isrR of PCC 6803: GGA TCT AAT GTT GGC TGA CTG ACT AGG CG) and the 3’
adaptor primer (GTG CTC TAG TCG TAC AGA AGT GTT GCA TTC ATC). Using
the products from this PCR as a template, a second, nested PCR step was required to
obtain distinct PCR products. In this nested PCR, for each gene a nested gene specific
primer (for isrR of PCC 6803: CCC CCA ATA CAT GGC TTT GCC TGC C) and the
3’ adaptor primer was employed.

Upon cloning and sequencing of the amplification product, the first nucleotide downstream
of the 5’ adaptor RNA or the last nucleotide upstream the 3’ adaptor was assigned the
transcription initiation or termination site, respectively.

2.1.10 Quantitative RT-PCR

DNase conditions
For quantitative RT-PCR analysis, the RNAs were digested with RNAse-free DNAse I
(Ambion, Austin, Texas USA). 3 µg RNA were incubated for 20 min with 6 units (3 µl)
DNase I, 1 x DNase buffer and 40 units (1 µl) RNase inhibitor (Fermentas, St. Leon-Rot,
Germany) at 37◦ C in a final volume of 30 µl. The DNase I was inactivated by incubation
for 3 min at room temperature with 3 µl of the provided glassbeads-based inactivation
reagent. After centrifugation, the RNA remained in the aqueous phase and was trans-
ferred to a fresh tube.

Reverse transcription
36 2. MATERIALS AND METHODS

gene RT primer (5’ to 3’) PCR primer (5’ to 3’)

trxAMed ACTGCTCCAACAACAGTATC CTGATTCCATATTGACTTGCAACATTTGGATTCTCATC
trxASS120 ATCATTAATGTAGGGATGCTTC GCCATATTGACTTGCAACATTGGGATTCTCATC
ATGGCAAGAATGATTTTTTTCGCCAAAATAAATAAGAGCTG
trxAMIT TGGCTTTGGGCACTGCA CAGGGTGGGGATACTGCGAATACCG
GGTCGTTGATTTTGCGGAAATTTAGCAATCGTTTG
trxAWH8 CGTTGGGATGCTGCGG CCGTACTGACTGGCGACATTAGGGTTTTC
pcbASS120 AGTAGCTAAGTGAGGAAGAG GTACTGAAGGGTCAAAACGAGCTAATTCAAAAAGTGTGAA
GGGAAAGCGCATAAACACTAACGAAAAGAGCGA
psbASS120 TGGGTAGAAGTGAAGACCTAT GCACCTGAAATGATGTTGTTCCCATAAAGGAAAGATC
CCGCCAAGAGGAGTTTAGATGAAAGTTTGCAA
psbDSS120 GACTGTGCCCCATAGC GGGTACTAACAGCGGCAGTAAGGAAATTACATC
GGATACGCCTGAGACTTTAACAATTCAAGAACCAAG
petHSS120 TGTTTACAGGAACATTGGGGT ATTGCCTTTGGCGAGGAAGAAGGCTTTTCAG
CCAATATCTCAACTTTATTATTCGACATAGCCTTTTTCAAAG
kaiBSS120 GGGGTTGCAAGTATCTTATCT GCTAGCTGTGGGTTTTTAAGCACATCAATTACTTTCAAAG
TCTAAATCTTGACGGATTTAACCTCTAAGAGTGAGGATG
ftsZSS120 CTAGGCCTCGAGTTAATGT GAACTCTATTCTCAGCCGAAGATTGCAACAAAGC
GGTGCATTATGACTGGCTTTGGAGATCTGTTG
ntcASS120 GTCGAACAGCTCCTCTAC GCAGGGTCCCCGGGGAAAAAGATC
CCAATTCAAAGGACGATAAGAAGTGAAGCTAAAACC
cpeBSS120 CTTCTAGTGGGATAGCAATG CTCACAAACCATTCCTGATATAGCATCTGCAGC
lexAMed TTCAATTAAGCCACCAGCTG CCCATAATTGGGACTCCCTCAAAGATCTCATC
umuDMed CTGGTTTAGCAGTTAGACTC CTAAGAAAAAAAGTACTAAAAGGATTAGATATTAAATGCT
CGTTC
PMM1427 AGCTGCTTCTGGAGCAAATA CGACTATCTTTCCTCCAGTAAGTACAGCCC
glnAMed GAATTCTGGCTCTGGTCC AGGGTTCTCCACTTCTTGGCTCTTGAATAG
urtAMed TGTCCATCCACCAAATACAAC CTCAGCAAATGTTGGCCAGTCGGAAGC
psbDMIT GCTATGACCCATCGCATC CTTACAGCGGCGCTAAGGAAATTGCAGC
psbDWH8 CTGTGACCCATCGCATC AGACAGCAGCGGTGAGGAAGTTGCAAC
psbDWH7 CTGTGACCCATCGCATC GGTGGACACAGCAGCGGTGAGG
atpBWH7 CGAGAACGTTGAAGATTCG GTTGCTTCACCCACGGGAACGCTG
rpl21WH8 ATGGGCCATCACCTTCAG CTTGGAATCCTTCACCATCAGAACGTTTTCG
rpl21MIT GTGGTCCATCACCTTCAG CTTGCCGTCTTTGATCAGCAGTACCTTCTC
CCGCGTTTAGGCGCCTTTAGTCAAGATTG
rpl21SS120 ATGGATGTAACCATTACTCTTGT CTCTTGCCTGTGCCCATTCTTTGTGCG
CAAGATAACATCCTCACTCTTAGAGGTTAAATCCG

Table 2.6: DNA oligos used for 5’ RACE experiments analysing mRNAs of Synechococcus WH8102
(WH8), WH7803 (WH7), Prochlorococcus Med4 (Med), SS120 (SS120) and MIT9313 (MIT).

The First-strand cDNA Synthesis System for Quantitative RT-PCR (Marligen Biosciences,
Ijamsville, MD U.S.) was used to transcribe about 300 ng DNA-free RNA into cDNA, ini-
tiated with oligo (dT)20 and random primers, following the manufacturers instructions.

Primer design
Prochlorococcus Med4 gene-specific RT-PCR primers for the gene rnpB as well as for isiB,
glnA, pstS and hli8 were designed, based on the genome sequence using PrimerExpress
software (PE Applied Biosystems, Weiterstadt, Germany). The following parameters were
required: length of the PCR product: 80 to 120 bp, length of the primers: 20 to 25 nt,
melting temperature of the primers: 58 to 62◦ C and GC content: app. 40 %. A list of the
designed primer sets is given in Table 2.7.

Quantitative PCR
For PCR, 1 µl of the reverse transcription reaction product was used in a 25 µl reaction
with 1.8 pmol of each primer, 1 x SYBR Green buffer, 0.2 mM each dNTP, 3 mM MgCl2 ,
0.625 U Ampli Taq Gold Polymerase (PE Applied Biosystems, Weiterstadt, Germany).
PCR was performed in triplicates in a GeneAmp 7500 real time thermocycler (PE Applied
Biosystems, Weiterstadt, Germany) with one initial denaturation step at 95◦ C for 10 min
2.2. BIOINFORMATICS PART 37

gene FW primer (5’ to 3’) REV primer (5’ to 3’)

rnpB TTGAGGAAAGTCCGGGCTC GCGGTATGTTTCTGTGGCACT
isiB GATGAGGAAAGGTCTGGAACTG TTCTGTATAGGTAGATGAATCCCC
glnA GGAATACTGGTCGAACAGAAGAA CCTTATGTCTTGTGCGGTGTC
hli8 CTCTAACTGCTTTACTCGTTGC AGTCCGATCATTGCAAATCTACC
PMM0710, pstS GTACAGGTTGCAATGGGTATGG TGTGAAAGCCTTAGTAGTTCCTG

Table 2.7: Prochlorococcus Med4 gene-specific RT-PCR primers.

followed by 40 cycles of 20 s of denaturation at 95◦ C, 30 s of annealing at 60◦ C and 45 s

of elongation at 72◦ C.
For each sample, triplicate control PCRs were performed using 1 µl of a reverse tran-
scription reaction without reverse transcriptase. To evaluate possible sample-to-sample
variations in RNA concentrations, the concentration of the RNA subunit of ribonuclease
P (rnpB) was used as an internal standard to normalise the mRNA levels.

Quantification
In a first step, the average of the CT values of the triplicate PCR reactions was calculated
(mCT). If the expression of one gene is described relative to the expression of an internal
standard, then the difference between the mCT of the investigated gene and the mCT of
the internal standard was calculated: mCT(mRNA) - mCT(rnpB) = dCT. The equation
2−dCT describes the initial cDNA amount which corresponds to the normalised amount of
mRNA. To compare the transcription profiles of the same gene under different conditions
(e.g. high light to normal light), one has to calculate the difference between the dCT value
of the control, normal light, and the dCT value of the high-light culture. The transcript
amount (cDNA amount) is then calculated from 2−ddCT .

2.2 Bioinformatics part

2.2.1 Genome data
Cyanobacterial genomes were downloaded from NCBI GenBank (Wheeler et al., 2005)
and for Synechococcus WH 7803 from Genoscope (Partensky, 2006).

2.2.2 Software tools

All used software tools are freely available. The multiple alignments were performed with
CLUSTALW 1.81 (Thompson et al., 1994). The sequence logos were created with We-
bLogo (Crooks et al., 2006), described by Schneider and Stephens (1990); Crooks et al.
(2004). All individual secondary structure predictions were done using MFOLD (Zuker,
2003). For visualisation of genomes and gene arrangements, the Artemis software (Berri-
man and Rutherford, 2003) appeared very useful.
All other scripts and tools implemented for more complex algorithms are described below
and are available on request.
38 2. MATERIALS AND METHODS

2.2.3 Promoter prediction

The whole genome prediction for putative promoter sites of marine cyanobacteria was
based on the raster-score-filter method, described previously by Vogel et al. (2003a). A
set of upstream regions (300 bp in length) was created using only non-coding sequences
by masking the annotated coding regions. The information of experimentally determined
and aligned -10 boxes was compressed in a scoring matrix. A search was done for promot-
ers which exhibited features similar to the experimentally verified set, and the quality of
the newly found -10 boxes inside uncharacterised regions was measured by means of this
scoring matrix. The prediction was divided into three steps:

- raster, a search for 6 nt as a potential -10 region based on the known promoter
structure in bacteria, -35 region - space - -10 region - space - TSS - space -
translation start site

- score, determining the weights of the potential boxes with a scoring matrix

- filter, using a filter (cut-off value) to reduce the false positive rate of the
predicted boxes

Scoring matrix
A set of binding sites can be used to create a scoring matrix with the nucleotides A, T,
C and G as columns and the binding site positions as rows. Each entry of the matrix is
determined by formula:
pi Nobserved
log2 = log2 [bits]
p0 Nrandom
with pi as the observed frequency and p0 as the expected frequency or Nobserved as the
number of observed nucleotides and Nrandom as the number of expected nucleotides.
The determined score of one position within the binding site is the sum of the scores of
one column of the score matrix and equals 2 bits minus the Shannon entropy H1 . Thus,
if a position has a overall score of 2 bits, it is completely conserved; and zero bits stand
for no conservation (equal distribution).
The entries of a score matrix can be extended by adding pseudocounts (small number of
counts) to the columns of the matrix to increase the variability either to avoid zero counts
or to add more variation than was found in the sequences used to produce the matrix
(Mount, 2001):
Nobserved + Npseudocount
log2 [bits]
Nrandom + 4Npseudocount
It is a useful method, particularly suitable for small sets of known sites.

2.2.4 Phylogenetic footprinting

Phylogenetic footprinting is the major method for enriching for candidate regulatory ele-
ments by searching for conserved motifs upstream of orthologous genes from closely related
2.2. BIOINFORMATICS PART 39

species (Bulyk, 2003). For Prochlorococcus Med4, MIT 9313, SS120 and Synechococcus
WH 8102 a systematic intergenomic comparison was implemented to detect conserved sites
between their orthologous upstream regions. The algorithm was developed in cooperation
with Dr. Szymon Kielbasa using free software tools, such as BLAST (Altschul et al.,
1990), GLAM (Frith et al., 2004) or MySQL (Axmark et al., 2006).

The phylogenetic footprinting algorithm can be divided in the following main steps:

- download of genbank files (Wheeler et al., 2005) and data management of

annotated genomes by MySQL (Axmark et al., 2006)

- defining orthologous gene sets by pairwise BLASTp (Altschul et al., 1990)

comparison (E-value < 10−5 ), definition of reciprocal best hits and a single-
linkage procedure (Tatusov et al., 1997, 2003)

- aligning the promoter sequences of orthologous genes using GLAM (Frith

et al., 2004), a local alignment algorithm

- visualising and identifying segments of significant conservation with Graphviz

(Ellson and North, 2006) as well as construction and comparison of position-
specific weight matrices (PSWM) (Kielbasa et al., 2005)

- genome-wide searching for entries of candidate regulatory elements (Kielbasa

et al., 2001) and definition of regulons

Palindromicity of all predicted sites was evaluated via PSWM comparison and chosen as a
main criteria, as it is known to be correlated with the motif regulatory function (McGuire
et al., 2000).

Additionally, all σ factors were identified by using The SEED database (Overbeek et al.,
2005). Regulatory factors including transcriptional regulators were defined by comparison
to the set of TIGRFAMs/Pfams (Fraser, 2006) in the functional role category "Regulatory
functions" (Subroles: DNA interactions, Protein interactions, Small molecule interactions,
Other).

2.2.5 Prediction of ncRNAs

To identify candidates for experimental investigations, a comparative computational ap-
proach was developed in cooperation with Philip Kensche (Kensche, 2004; Axmann et al.,
2005) that was based on sequence and structure conservation and used the program ALI-
FOLDz (Washietl and Hofacker, 2004). The genome sequences of Prochlorococcus SS120,
Med4, MIT 9313 and Synechococcus WH 8102 were downloaded from NCBI GenBank
(Wheeler et al., 2005). A summary of the computational screening is given in Figure 2.3.
40 2. MATERIALS AND METHODS

A detailed description of the algorithm can be found by Kensche (2004); Axmann et al.
(2005); the main steps are described below.

It was assumed that homologous RNA structures would show a reasonable degree of conser-
vation on the sequence level for the given set of genomes. BLASTn version 2.2.8 (Altschul
et al., 1990) was used to screen for local sequence conservations within intergenic spacer
regions (IGRs) longer than 49 bp.
To take advantage of a multi-genome comparison, the pairwise sequence alignments were
transformed into multi-sequence clusters via single-linkage clustering. Sequences that pro-
duced a significant blast hit (E-value ≤ 10−10 ) for a given query were collected into initial
clusters.
Finally, each cluster was aligned using CLUSTALW version 1.81, default parameters
(Thompson et al., 1994) and the resulting alignments were scored by ALIFOLDz (Washietl
and Hofacker, 2004). The Z-score cut-off of -4 used by Washietl and Hofacker (2004) was
chosen as a soft cut-off. For all structure computations, folding temperatures were set to
24◦ C, which is the approximate habitat temperature of the marine cyanobacteria studied
here (Partensky et al., 1999).

Despite any structural conservation, any RNA in principle may encode for a peptide. The
necessary reading frame as defined in this analysis consisted of at least ten consecutive
codons starting with either one of the four possible start codons ATG, GTG, TTG or
ATT and finishing with TAA, TAG or TGA. If a reading frame was present, the possible
conservation of the encoded peptide sequence amongst other cyanobacteria was evaluated
by alignments (tBLASTn). Only in the case of a conserved open reading frame, the RNA
was considered to be coding and deleted from the list of possible ncRNAs.
2.2. BIOINFORMATICS PART 41

A B

Figure 2.3: Pipeline for comparative prediction of non-coding RNAs (Axmann et al., 2005). (a) Intergenic
sequences (IGRs) longer than 49 base-pairs were gathered from four Prochlorococcus and Synechococcus
genomes and locally aligned using BLASTn. Because of the initial asymmetric local alignment using
BLASTn (see Figure 2.3b for a summary of significant BLASTn hits between the strains Prochlorococcus
Med4 (MED), MIT 9313 (MIT), SS120 (SS) and Synechococcus WH 8102 (WH)), all candidate sequences
were reverse-complemented. Redundancy in this data set was reduced by unifying those hits from each
genome that showed a reciprocal overlap of 85 % or greater. This candidate set was used as both query
and subject in another local alignment step (BLASTn considering only the query strand as possible subject
strand). Sequences that directly produced a significant blast hit (E-value ≤ 10−10 ), or were connected by
a chain of such hits, were gathered into clusters (’single-linkage clustering’). Both genome strands were
screened; thus, the pipeline produced 310 pairs of clusters in both forward and reverse complementary
orientation. After an additional unification step of overlapping sequences within each cluster, the resulting
clusters and their complement clusters were scored using ALIFOLDz (Washietl and Hofacker, 2004). (b)
The number of BLASTn high-scoring segment pairs for each query and subject combination of intergenic
regions is given for a BLASTn E-value cut-off of 10−5 and after import of high-scoring segment pairs with
an E-value of 10−10 or lower (in parentheses). MIT, Prochlorococcus strain MIT 9313; SS, Prochlorococcus
strain SS120; WH, Synechococcus sp. WH 8102, MED, Prochlorococcus strain Med4 (Axmann et al., 2005).
Chapter 3

Results

3.1 Cyanobacterial promoters

The 5’ RACE (rapid amplification of cDNA ends) is a very sensitive detection method for
bacterial transcription initiation sites (TIS) by differentiating them from RNA process-
ing sites (Bensing et al., 1996). This technique was successfully tested to identify TIS in
Prochlorococcus Med4 (Vogel et al., 2003a) as the experimental basis for a computational
promoter prediction. It was of special interest to check if this method could be transferred
to another cyanobacterial strain and if different results would be obtained for distinct sets
of genes, possibly linked to their divergent lifestyles.

-35 16 bp -10 6 bp TSS 15-85 bp START

• Raster:

• Score: scoring matrix W TATAGT

A -3.424 1.277 0.036 0.036 -0.102 -3.424

C 0.407 -1.915 0.085 -1.915 0.670 -1.915
G -0.915 -1.915 -0.330 0.893 0.670 -1.915
TATAGT

A T
T 0.969 -3.424 0.036 -0.102 -0.616 1.277 Score: 4.3 2

T
• Filter: cutoff value of 2.5 score of 4.3 > 2.5 cutoff value
bits

A
0C
G
TATAGT putative -10 box
TAA
G
-10
-12

-11

-8
-9

-7

Figure 3.1: Raster-score-filter method (left) described by Vogel et al. (2003a) and weblogo of -10 boxes
(right) of 10 promoter sequences of experimentally determined promoters in Prochlorococcus SS120. The
positions -12 (T), -11 (A) and -7 (T) were found for the majority of TISs, see also Table 3.1.

The complete genome sequence of the extremely low-light-adapted Prochlorococcus SS120

was described in 2003 (Dufresne et al., 2003) including a manually verified genome annota-
tion which could be used here to design gene specific primer-sets for RACE experiments in
SS120. In parallel, an in silico prediction revealed a total of 3,130 transcription start sites
for 1,289 noncoding upstream regions of the 1,930 predicted protein-coding or RNA genes
in SS120 by using the raster-score-filter method (Vogel et al., 2003a), see also Figure 3.1,
3.1. CYANOBACTERIAL PROMOTERS 43

with a slightly modified scoring matrix adapted to the AT-content of SS120 non-coding
regions. The complete prediction can be downloaded from cyanolab (Hess, 2006). By anal-
ogy to Med4 (Vogel et al., 2003a), about 40 % of these sites were estimated to be functional.

The experimental validation by the RACE technique for eight randomly chosen genes in
SS120 showed that promoter elements are very similar to the ones described for Med4
previously (Vogel et al., 2003a). Again a well conserved -10 region could be identified with
a tANNNT like consensus motif, shown in Figure 3.1. As the computational algorithm
was mainly based on this -10 region, the in silico prediction was correct for most of the
-10 boxes upstream of the eight genes pcbA, psbA, psbD, petH, kaiB, cpeB, ftsZ and ntcA
in SS120, listed in Table 3.1.

promoter sequence position

cpeB gtaccatttctgctcttctagaagaggcattagagtTAGGTTggacagCG -63/-64
ftsZ1 actttcttgaaaaacccaacagatctccaaagccagtCATAATgcaccAA -31/-32
ftsZ2 cataatgcaccaaaaatcattcatctaaatcccgagTATGGTgatggctA +7
kaiB1 acaagataacatcctcactcttagaggttaaatccgtCAAGATttagAtA -36/-38
kaiB2 agaggttaaatccgtCAAGATttagAtAtatcaatTAAAGTctttctCtA -14/-16
ntcA gtcatttttgatacaagaggttttagcttcacttctTATCGTcctttgAA -13/-14
pcbA gtattgaaattagtccccttattttcgctcttttcgtTAGTGTttatgcG -24
petH tcataacgtgaactttgaaaaaggctatgtcgaataaTAAAGTtgagatA -32
psbA ttcttggttcagtacgcagttttgcaaactttcatcTAAACTcctcttGG -55/-56
psbD tgataaaaaatttccccgacttggttcttgaattgtTAAAGTctcaggcG -30

Table 3.1: Experimentally determined mRNA 5’ ends of Prochlorococcus SS120. Putative -10 regions
are in upper case letters as well as the first transcribed nucleotide. The given positions are calculated
relatively to the annotated start codon of the protein-coding gene.

In 2003 genomes were sequenced of another low-light-adapted strain, Prochlorococcus MIT

9313 (Rocap et al., 2003), as well as Synechococcus strains WH 8102 (Palenik et al., 2003),
and recently WH 7803 (Partensky, 2006). All these strains were successfully analysed
by RACE experiments, demonstrating the good practicability of the method for marine
cyanobacteria. The huge amount of genomic and RACE data allowed a detailed compar-
ative analysis of promoter regions for the different cyanobacterial strains. Subsequently,
some of the most interesting results from this analysis are shown exemplarily.

3.1.1 psbD promoter region

Photosynthesis is the major energy source for an oxyphotoautotrophic bacterium. Prochlo-
rococcus Med4 and SS120 possess the most reduced genomes of a free-living photosynthetic
organism known to date. Therefore, nearly all photosystem II (psb) genes are single-copy,
in contrast to other cyanobacterial genomes, which contain multiple copies of psbD, psbA
and other genes (Dufresne et al., 2003).
By comparing the promoter regions of the different psbD (photosystem II D2 protein)
genes (Fig. 3.2), the highest similarity could be detected between PMM1157, Pro1254,
44 3. RESULTS

psbD psbC

ppiA ycf4

Syn7803 : TACAACGGATTTCAGCTGAGAGCGA-TCAGCCCCGCAGTGCCCGTTGAGATTGTTAAGGTCCCCAG-GTATCCCACAGCCT-39nt-ATG
Syn8102 : TACAAGGGATTTCAGCTGATGCAAA-CAGCCCCCGCGACACCGGCTGAGATTGTTAAGGTCGCTGACGTATCCCACAGCCT-39nt-ATG
Pro9313 : TACAAGGGGTTTCAGCTCATGGCAACTCTCTCCCAAAGGCCCCTTTGGGATTGTTAAAGTCTTCGG-GTATCCCACAGCTT-41nt-ATG
ProSS120 : AACAACGGATTTCATCTGATAAAAAATTTCCCCGACTTGGTTCTT--GAATTGTTAAAGTCTCAGGCGTATCCCAAAAGCT-29nt-ATG
ProMed4 : TACAAGGCATTTCATAT-ATACAGATATTTCCCCACTTTACTCTCAATCTTTGTTACAGTCTC-ATCGTATCCCGAAGCTA-26nt-ATG
- 60 - 35 - 10 TIS psbD

Figure 3.2: Arrangement of ppiA, ycf4, psbD and psbC and alignment of upstream regions of psbD.
The green colour is indicating genes involved in photosynthesis. The arrangement of the four genes is
conserved between five cyanobacterial strains. The alignment of upstream regions of psbD (photosystem II
D2 protein) exhibits conserved blocks around TIS, -10, -35 and -60 region counted from the experimentally
verified and aligned TIS (labelled by an arrow and described for Prochlorococcus Med4 by Vogel et al.
(2003a); Prochlorococcus SS120 and MIT 9313, Synechococcus WH 8102 and WH 7803 this work). The
total length of the intergenic spacer between ycf4 and psbD varies around 200 bp within the five compared
genomes.

PMT1179, SYNW0677 (WH 8102) and ORF1665 (WH 7803), although the strains WH
7803 and WH 8102 harbour an additional psbD gene with a verified and functional TIS.
For all four most related copies, highly conserved regions occur at the TIS, the -10 and
partly at the -35 region but interestingly also at the -60 region counted from the mapped
and aligned TIS. An extended view of this genome-section (Fig. 3.2) revealed a very
conserved microsynteny with a head to head arrangement of the psbD and ycf4 (photo-
system I assembly protein) gene with a distance of about 200 bp, which might represent a
bidirectional promoter region and would allow a co-transcription of the genes needed for
photosynthesis.
In agreement with these results, a reciprocal BLASTp search verified this set of psbD
genes as the orthologous one. The additional psbD copies of Synechococcus do not possess
the strong conservations within their promoter regions, which were found for the four or-
thologous regions. Therefore, these additional copies might be regulated in a different way.

3.1.2 csoS1-rbcLS promoter region

csoS1-rbcLS and atpB are further examples for highly conserved promoter regions and
gene arrangements (Fig. 3.3 and Fig. 3.4). rbcLS encodes the ribulosebisphosphate
carboxylase (large and small chain), the key enzyme of photoautotrophic carbon assim-
ilation. The gene csoS1 (or ccmK ), and four additional genes (csoS2-csoS3-orfA,B for
carboxysome peptides A and B) located immediately downstream of rbcLS encode com-
ponents of the carbon concentrating mechanism. The alignment of five upstream regions
of different strains (Fig. 3.3) reveals high conservations for the TIS, -10 and -35 elements,
and again, several conserved positions at the -60 region.
3.1. CYANOBACTERIAL PROMOTERS 45

csoS1 rbcL rbcS

HAM1 family

Syn8102 : GGACAGTGATCTCGGGAGTGTGCCCAATTGCTTGACGGGAAGGGATCTGGCGGACCAATGTCCCGAGCGAACATTCCACCTTTT-13nt-ATG
Syn7803 : GGACAGTGATCTCGAGAGGTCTCTGAATCGCTTGACGGCCAGGCCTCTGGCAGAGCAGTCTCGGCGGCGAACATCCT-CCTCCT-13nt-ATG
Pro9313 : GGATGGTGATCTCCGTTTCGGCTGAATTTGCTTGACGGAAGTTGCCTGCGCAGAGCATTCTCCTTCCCGAACATTCAATCTCCT-14nt-ATG
ProMed4 : AAACATTAATAGAGTATTTAGTAAAAAATGCTTGACTTATCAGTACGTTATGGACCATTCTTCGGATTGAACATTCCACATTTA-29nt-ATG
ProSS120 : ATAAGGTGATAGTGTTTTTGATGAATTCCCCTTGACTTAATAGTATCTCAGCCGTCATTGTCCTTGATGAACATTCCATCCTTT-16nt-ATG
- 60 - 35 - 10 TIS csoS1-rbcLS

Figure 3.3: Comparison of the csoS1-rbcLS region from marine cyanobacteria. The csoS1-rbcLS operon
is located in all cases head to head to a gene encoding a HAM1 family protein. The experimentally
identified TIS (labelled by arrows) of csoS1-rbcLS for Prochlorococcus Med4 (Vogel et al., 2003a) and
for Synechococcus WH 7803 (Garczarek et al., 2001) can be easily aligned with upstream sequences of
Prochlorococcus SS120 and MIT 9313 and Synechococcus WH 8102 suggesting that the TIS, -10 and -35
region are conserved for these five strains. The total length of the intergenic spacer is 300 to 350 bp in the
five compared genomes.

3.1.3 atpB promoter region

The arrangement of groESL, an operon encoding a cyanobacterial chaperone, on the for-
ward strand and the genes atpB, atpE encoding the beta and epsilon subunits of the
proton-translocating ATPase on the reverse complementary strand strand is the same for
the five strains investigated. The length of the intergenic region between groES and atpB
is about 220 bp, so that a co-transcription and therefore co-regulation might occur.

Again, within the upstream regions highly conserved positions can be found around the
TIS and the -10 box. However, in this example nearly randomly distributed nucleotides
appear at the -35 or at the otherwise often conserved -60 region, whereas a large 19 base-
pairs long sequence block occurs between the -35 and -60 site.
Comparison of the predicted translation products of cyanobacterial groES and atpB demon-
strated a very high amino acid identity with cognate chaperonins and ATPases (beta
subunit), respectively, from bacteria and chloroplasts; their phylogenies confirmed the en-
dosymbiotic origin of chloroplasts from cyanobacteria (Curtis, 1987; Webb et al., 1990;
Morden et al., 1992).

3.1.4 sfsA downstream region

In contrast, for other examples non-coding sequences were observed without any significant
degree of conservation despite a high degree of microsynteny of the respective genes, e.g.
the non-conserved sequences downstream sfsA (Fig. 3.5), where a conserved arrangement
with mviN, amt1 and lytB was found. SfsA is similar to a sugar/maltose fermentation
stimulation protein; mviN encodes a predicted membrane protein homologous to the vir-
ulence factor MviN; Amt1 belongs to the ammonium transporter family; and LytB is
46 3. RESULTS

atpB atpE
groEL groES

Syn8102 : ACCAGCTCCAGGCTGTGCGGTTGACCGAACCGGGGGA--GTGCGGCTCA--GCAGTGGTAAGGTTGCGCCGCTCTGAATGGG-39nt-ATG
Syn7803 : CCATGGATTCGGC-GTGCGGGTGACCGAACCCTTTG---GCCCCGCTCAAGGCGGTGGTATCGTTGCGCCGCTCGAGAAGGG-35nt-ATG
Pro9313 : ACTAGTGGATACCTGTGCGGGTCACCGAACCCTGAGA--AGATCTCGCCATGCAGTGGTAATGTTGCGCCGCTCTGAAAGGG-42nt-ATG
ProMed4 : ACCACTCTTTAACTGTGCGGTTGGCCGAACCCAGTTAACGAATATACCCTCGGGATGGTAATATTTCGCATT-----ATGAG-40nt-ATG
ProSS120 : ACTACAACTGTTCTGTGCGGTTGACCGAACCGCACAACATACCTTGCCGGCAGGGTGGTAGTCTTGCGCGGCT--GAGTGAG-32nt-ATG
- 60 - 35 - 10 TIS atpB

Figure 3.4: Arrangement of genes around atpB and alignment of atpB upstream sequences. This region
exhibits a conserved synteny in Prochlorococcus Med4, SS120 and MIT 9313, Synechococcus WH 82012 and
WH 7803 with a head to head orientation of groES and atpB. By aligning the 5’ UTRs of atpB conserved
boxes appear around the TIS and -10 region and between -35 and -60 region. The total length of the
intergenic region between groES and atpB is about 220 bp.

described as a penicillin tolerance protein.

sfsA amt1 lytB

mviN

ProSS120 : ---TTCTTTAGAATATTTTTGTCTT--AATTACAAAACTAAGTAAACCTATAGAAACTAGAGCCAA-GAGTGT--ACCTCCTGCGATCATTACATTGGATCGTTCAGA--
Pro9313 : --CACAGCAAAAAGAGCTCCAGAT----GTAACAAAATGAACTTTCACCCCCGCAAATAGACCTAA-AAGCGTCCACCAATCCCATATTTCGTATCAACATGAACAT---
Syn8102 : GATCTTGTATCAATTAACACACCTA--AGGGCGAATCTGCGCTGCACTTTGGGTGAGTTCAGTCGG-ATACACC-ATCCGGCGTGACCAGCAGCTCTGCACTCT------
ProMed4 : -----TTTTTTGGGATTTTAAGTTG--ATTTAGAAAAAATACCTCAATTTAACAAAGATTATTTGGTATACAACAATTAAATTGGTATAAACGACTACTAGACATAC---
Syn7803 : -------TAATGGCATCAACGGCTACCAACCGCCAAGCCGAACGCACCTTTCTGGGGT--GTTTGGTAT-CAACAACTACTTGAGGGGTCCGCTTGGGCAGCCACCTTTG

Figure 3.5: Synteny of mviN, sfsA, amt1 and lytB and alignment of sfsA downstream sequences for five
marine strains. The adjacent non-coding sequences downstream sfsA of Prochlorococcus Med4, SS120 and
MIT 9313, Synechococcus WH 82012 and WH 7803 are shown. A conserved synteny with a head to head
orientation of mviN and sfsA followed by amt1 and lytB is visible. By aligning the downstream sequences
of sfsA no conserved regions are detectable. The total distance between sfsA and amt1 varies highly from
about 100 bp (WH 8102) to 300 bp (SS120).

Thus, one can expect that not all comparable non-coding sequences of these five marine
strains align and contain conserved blocks just by chance. Therefore, the phylogenetic
distance might be assumed as far enough between the related non-coding regions given by
this set of marine cyanobacteria, so that the detection of conserved regions as meaningful
biological signals inside otherwise non-conserved sequence parts becomes possible.
A highly conserved promoter region might indicate a likely similar transcriptional regula-
tion for the corresponding orthologous genes. On the other hand, several non-conserved
3.2. PHYLOGENETIC FOOTPRINTING 47

but regulatory important strain-specific sites, developed during a particular adaptation

process, will be excluded by a comparative approach.

It seems obvious that not only protein-coding information appears to be conserved be-
tween genomes but also the essential regions for promoter recognition and transcription
initiation. Interestingly, the strong conservation of the promoter regions between the dif-
ferent strains included mainly the transcriptionally important areas such as TIS, -10 or -35
region. Other nucleotides obviously follow different evolutionary constraints as there exist
non-conserved parts between the conserved blocks of important promoter sites. Thus, a
promising idea was to search more deeply for conserved elements upstream of orthologous
genes to identify essential sites for regulation like cis elements recognised by transcrip-
tional regulators.

3.2 Phylogenetic footprinting

A systematic intergenomic comparison of upstream nucleotide sequences from sets of or-
thologous genes (phylogenetic footprinting) was implemented to identify conserved regions
and regulatory motifs other than the core promoter elements. Therefore, free software
tools as BLAST (Altschul et al., 1990) or GLAM (Frith et al., 2004) were used to identify
groups of orthologous genes and to analyse their upstream sequences by a local alignment
algorithm. All intermediate data were managed by MySQL database server and in coop-
eration with Dr. Szymon Kielbasa.

Med4 SS120
1415
1716 CDS 1882 CDS
incl.5 σs incl. 5 σs
and 44 RFs and 42 RFs

1293 COGs
1360 incl. 5 σs and 1445

35 RFs (18 TFs)

1444 1405

MIT 9313 WH 8102

2273 CDS 2526 CDS
incl. 7 σs 1706 incl. 6 σs
and 66 RFs and 66 RFs

Figure 3.6: Overview of orthologous protein-coding genes including σ and regulatory factors, RFs, (in-
cluding transcriptional factors, TFs) of four genomes, Prochlorococcus Med4, MIT 9313, SS120 and Syne-
chococcus WH 8102, defined by best reciprocal BLASTp comparison with a cutoff E-value of 10−5 . The
4-way comparison resulted in 1293 clusters of orthologous groups (COGs) with at least one entry per
genome (red box). Numbers of pairwise orthologs are underlined. In black boxes total numbers of CDS
features of each single genome are displayed.
48 3. RESULTS

3.2.1 Orthologous gene sets

The starting point for a phylogenetic footprinting analysis is the definition of the set of
orthologous protein-coding genes between the genomes of interest. Therefore, the recip-
rocal best BLASTp hits were identified based on the annotated CDS features of the four
genomes Prochlorococcus Med4, MIT 9313, SS120 and Synechococcus WH 8102, similar
to the system described by Tatusov et al. (1997, 2003). The 4-way BLASTp comparison
with a cutoff E-value of 10−5 resulted in 1293 clusters of orthologous groups (COGs) with
at least one entry per genome (Fig. 3.6). As a subset of these 1293 COGs, σ factors were
identified by using the information of The SEED database (Overbeek et al., 2005). Regu-
latory factors (including transcriptional factors) were defined by comparison to the set of
TIGRFAMs/Pfams (Fraser, 2006) in the functional role category "Regulatory functions".
Thus, a core set of five σ and 35 regulatory factors including about 18 transcriptional
factors was suggested as the regulatory potential mediated by proteins conserved between
the four marine cyanobacteria (Appendix, Tab. 1).

3.2.2 Predicted TF binding sites and regulons

For each COG, its set of upstream sequences was extracted. The local alignment algorithm
GLAM (Frith et al., 2004) detected the best conserved element between the four strains
for each set of upstream regions. All GLAM-identified elements were translated into ma-
trices and compared to each other. Thereby, clusters of similar matrices were observed,
suggesting binding sites for trans-acting factors, which control more than a single gene.
The Graphviz package (Ellson and North, 2006) appeared to be very efficient for the vi-
sualisation of calculated similarities between the matrices. Additionally, palindromicity of
all predicted sites was evaluated, since it correlates frequently with a regulatory function
(McGuire et al., 2000).
The best conserved elements present among different orthologous sets were extracted and
realigned for construction of 21 position-specific weight matrices (PSWM); their weblogos
are listed in the Appendix, Figure 2. Finally, these 21 scoring matrices were used to search
for candidate regulatory elements in all upstream regions of all four genomes.
The results of the PSWM-search were analysed in detail by assigning the downstream
genes to known pathways or regulons. Towards this goal, the genome annotation and an
intensive literature search were informative. This final evaluation revealed three best mo-
tifs with analogy to already known sites for certain cyanobacteria: NtcA, LexA or ArsR,
described in detail below.
These three predicted recognition sites similar to NtcA, LexA or ArsR are spaced and
palindromic DNA motifs, and the further focus was kept on predicted elements, which
possess these properties as well. Together with two additional best motifs their predicted
binding sites and putative regulons are listed in Table 3.2. The complete list of results of
the PSWM-search for the five best motifs is shown in the Appendix, Table 2.
3.2. PHYLOGENETIC FOOTPRINTING 49

palindrome sequence logo best hits upstream of

ATCAA-N6 -TTGAT gap; arsR; pstS; phoB;

(Med4: phoR; SS120: crp)

ATAAG-N5 -CTTAT ham1; csoS1-rbcLS; petF ;

psbA; (SS120: hli12)

TAGTACA-N2 -TGTACTA recA; umuD; umuC ; (SS120:

sbcDC )

TGT-N10 -ACA ntcA; spt,agt; glnA; glnB;

urtA

GTCAG-N6 -CTGAC conserved hypothetical pro-

teins

Table 3.2: Top five of predicted motifs identified in Prochlorococcus Med4, SS120, MIT 9313 and Syne-
chococcus WH 8102 via phylogenetic footprinting analysis. Only those hits are listed, which are highest-
scoring in non-coding upstream regions and possess a predicted function for at least two orthologous
downstream genes, indicating a putative regulon (hits found in one genome only are put in parentheses).

1. Motif: ArsR-like

Within the shared upstream region of arsR and gap, encoding a bacterial regulatory pro-
tein of the ArsR (arsenical resistance regulator) family and a glyceraldehyde-3-phosphate
dehydrogenase respectively, a conserved inverted repeat was identified. The found motif
ATCAA-N6 -TTGAT (Tab. 3.2) is identical to one of two direct repeats (spaced by 13 bp)
of the ArsR site described for Synechocystis PCC 6803 by Lopez-Maury et al. (2003). The
motif, identified here, often appears in two repeats but spaced by only two letters.
Interestingly, the PSWM-search predicted this site for several phosphate-depending genes:
phoB (two-component response regulator, phosphate), phoR (two-component sensor histi-
dine kinase, phosphate sensing), pstS (ABC-type phosphate transport system periplasmic
component) and other genes encoding for putative transporters. A comparison to the
DNA binding sites of PhoB, an eight-base motif repeat (TTAACCTT-N3 -TTAACCAT)
suggested by Su et al. (2003) - similar to the Pho box of Synechocystis (Suzuki et al.,
2004), revealed no similarity to the putative marine ArsR-like site, and moreover, PhoB
binding sites predicted by Su et al. (2003) did not co-appear with it (with one exception:
in Med4 upstream of pstS).
Arsenic, one of the most important global environmental pollutants, is widely distributed
and also contaminating seawater. One of its chemical forms, arsenate, is a phosphate
analogue, which might compete with phosphate in transport and in metabolic pathways
(Thiel, 1988). Many organisms contain genes involved in arsenic resistance (ars genes).
Thus, several cyanobacterial species are able to grow in the presence of high concentrations
of arsenate and in low-millimolar concentrations of arsenite (Thiel, 1988; Lopez-Maury
50 3. RESULTS

et al., 2003). In Synechocystis PCC 6803 the arsBHC operon is responsible for arsenic
sensing and resistance (Lopez-Maury et al., 2003). The complete arsBHC operon, regu-
lated by ArsR in Synechocystis, is not present in the marine cyanobacteria analysed here,
but two genes, arsB (arsenite transporter gene) and arsC (arsenate-reductase gene), exist
separately within some of these marine genomes. The possible role of arsenic for marine
cyanobacteria is still unknown, and no information is available about its uptake/export,
toxicity, function and means of regulation there. Thus, a regulatory connection to the
phosphorus-circuit remains highly hypothetical.
However, a complex transcription regulation was already observed during phosphate-
starvation in other bacteria, which is presumably achieved by extensive cross-talk among
multiple transcription factors (Birkey et al., 1998; Sun et al., 1996; Verhamme et al., 2002).
For cyanobacteria, including Synechococcus WH 8102, a comparable complex network was
suggested, in which the tandem DNA recognition by PhoB, a two-component signal trans-
duction transcriptional activator, represents an important part of the P-regulon (Blanco
et al., 2002; Suzuki et al., 2004; Su et al., 2003).
One might assume a repression of the phosphate-depending genes caused by ArsR bind-
ing in the promoter region similar to an effect observed in Anabaena variabilis, where
a preincubation of phosphate-starved cells with arsenate caused subsequent inhibition of
phosphate transport, suggesting that intracellular arsenate inhibited phosphate transport
(Thiel, 1988).

2. Motif
The GLAM-detected motif for the spacer region of ham1 and csoS1-rbcLS, ATAAG-N5 -
CTTAT, was found further upstream of the -60 region, described for the csoS1-rbcLS
promoter before. Additional entries were observed upstream of petF and psbA. Only in
SS120 a site was predicted adjacent to hli12, a high-light inducible gene. Thus, this puta-
tive cis element might regulate genes needed for photosynthesis as well as rbcLS, the key
enzyme of photoautotrophic carbon assimilation.
In Synechococcus sp. PCC7002 a CO2 response element and the corresponding factor
trans-acting on the rbcLS promoter were observed previously (Onizuka et al., 2002). Al-
though the AT-rich element found there is not very similar to the here identified palin-
dromic motif for marine strains, the negative regulatory role of its trans-acting factor, me-
diating the rbcLS transcription in response to CO2 levels in the fresh-water strain PCC7002
(Onizuka et al., 2002), might characterise a mechanism existing in marine cyanobacteria
as well.

3. Motif: LexA
The recognition site of another regulator, LexA, has been identified in recent years for
several bacterial clades, demonstrating that the binding sequences of gram-positive and
cyanobacteria are closely related (Mazon et al., 2004b). Further experimental and phylo-
genetic analyses revealed the existence of two independent evolutionary lines for the LexA
recognition motif, suggesting gene loss and lateral gene transfer during the evolution of
lexA, a gene governing such a complex regulatory network as the SOS system (Mazon
3.2. PHYLOGENETIC FOOTPRINTING 51

et al., 2004a). In E. coli most of the genes belonging to the SOS regulon are repressed
by LexA. RecA, the positive regulator, activated by DNA damage, facilitates the autocat-
alytic cleavage of LexA, resulting in the expression of genes that are normally under the
negative control of LexA (Walker, 1984).
LexA boxes were described previously for the cyanobacterial counterparts of the main SOS
genes, lexA, recA, uvrA and ssb, in Anabaena PCC 7120 and Nostoc punctiforme ATCC
29133 (Mazon et al., 2004b). In contrast, for Synechocystis PCC 6803 only one LexA
box was predicted by Mazon et al. (2004b), which is located upstream of the lexA gene.
Moreover, the Synechocystis LexA protein contains modifications in important residues in-
volved in the autocleavage process, which is especially important for LexA in its function
as a repressor (Mazon et al., 2004b). Thus, based on detailed studies with Synechocystis
lexA, recA and ruvB genes, Domain et al. (2004) argued against the involvement of LexA
in the regulation of DNA repair in cyanobacteria. Recently, a complete novel function
was reported for LexA, where it is not acting as a repressor but as an activator for the
transcription of the hox-operon in Synechocystis, recognising two cyanobacterial LexA con-
sensus sequences much further upstream of the core promoter (Gutekunst et al., 2005).
The LexA-binding sequence detected here for the marine strains is a palindromic and
spaced motif with the improved consensus TAGTACA-N2 -TGTACTA as shown in Table
3.2. This consensus sequence is highly similar to the motif TAGTACT-AA-TGTTCTA,
described for Anabaena PCC 7120 and located upstream lexA (Mazon et al., 2004b) as
well as to the cyanobacterial LexA box discussed in detail under evolutionary aspects by
Mazon et al. (2004a). Within the four Prochlorococcus and Synechococcus strains studied
here, a putative LexA site was found upstream of the genes recA - encoding the bacterial
DNA recombination protein, umuC and umuD - putative SOS mutagenesis proteins. Only
a lower-scoring site was observed for the regulator encoding gene, lexA itself. Additionally,
in SS120 this site was predicted upstream sbcDC, two genes for DNA repair exonuclease
and ATPase, not conserved between the four marine strains.

4. Motif: NtcA

Another motif, GT-N6 TGNTACA, identified here by phylogenetic footprinting, was ob-
viously similar to the palindromic consensus sequence TGT-N10 -ACA recognised by the
transcriptional regulator NtcA. The motif predicted here for marine Prochlorococcus and
Synechococcus includes additionally the flanking A/T-rich sequences and the conserved
TG (or CA) dimer described previously (Vazquez-Bermudez et al., 2002). The best scor-
ing hits of the genome-wide search for additional entries, using a PSWM for the ’marine’
NtcA motif, are listed in Table 3.2: High-scoring sites were detected upstream of genes
known to be regulated by NtcA or involved in nitrogen metabolism as ntcA - encoding
the regulator itself, spt,agt - serine:pyruvate/alanine:glyoxylate aminotransferase, glnA -
glutamine synthetase, glnB - nitrogen regulatory protein P-II and urtABCDE - putative
urea ABC transport system.
52 3. RESULTS

5. Motif
The last of the five best clusters identified in this study possesses the conserved palindrome
GTCAG-N6 -CTGAC. Most of the high-scoring sites for this motif are located upstream of
genes with unknown function (conserved hypothetical proteins) indicating a new binding
site for a new regulon in marine cyanobacteria, which is unknown so far and likely dissim-
ilar from other bacteria.

3.2.3 Experimental verification of transcription initiation next to puta-

tive binding sites
5’ RACE experiments in Prochlorococcus Med4 were used to locate the TIS of genes, for
which putative binding sites had been predicted via phylogenetic footprinting. For two of
the best scoring motifs, NtcA and LexA, three genes were chosen, respectively. The TIS
of urtA, glnA, ntcA as well as for lexA, umuD and PMM1427, a conserved hypothetical
ORF, were mapped by RACE experiments. The results are shown in Figure 3.7.

PMM1262_lexA PMM1427_ORF PMM0936_umuD PMM0920_glnA PMM0970_urtA PMM0246_ntcA

M TAP+ TAP- M TAP+ TAP- M TAP+ TAP- M TAP+ TAP- M TAP+ TAP- M TAP+ TAP-

Figure 3.7: Results of the PCR step during 5’ RACE experiments for lexA, umuD, PMM1427 (left
panel) and urtA, glnA, ntcA (right panel) in Prochlorococcus Med4. For each gene one single TIS appears
except for ntcA, which exhibits two signals in the TAP-treated (TAP+) line. Overlay of the putative LexA
recognition sequence (upper case letters) and the determined TIS (indicated by an arrow) for lexA, umuD
and PMM1427 (left panel). Overlay of the predicted NtcA binding site and the mapped TIS upstream of
urtA, glnA, ntcA (right panel).

NtcA can act as both an activator and a repressor, as it is also the case for other regulators
of the CAP family (Kolb et al., 1993; Herrero et al., 2001). In Med4, NtcA might act as
a repressor if it binds to the -10 region, which could be the case for glnA and the second
TIS (P2) for ntcA (Fig. 3.7). An activating function of NtcA is described by binding at
the -35 region (Herrero et al., 2001), which might happen for urtA and the first TIS (P1)
of ntcA (Fig. 3.7).
The putative negative control of glnA by NtcA in Med4 appeared unusual, because in
3.3. CIRCADIAN RHYTHM AND KAI GENES 53

fresh-water cyanobacteria this gene was often found to possess a NtcA-activated promoter
(Herrero et al., 2001). On the other hand, a second TIS might exist for glnA, which was
not active under the conditions chosen for the RACE experiment. Additionally, an up-
regulation of glnA expression was clearly demonstrated by RT-PCR for nitrogen-depleted
Med4 cultures (by a factor of 5.8 compared to standard conditions), indicating a second
NtcA-activated promoter site or a regulation different from those cyanobacteria investi-
gated so far.
The ntcA gene itself has been found to be autoregulatory in several cyanobacteria. A
negative feedback control, which is typical for bacteria, is suggested here for Med4 as well
by the location of the NtcA recognition site at the -10 region of ntcA P2. If NtcA is
activated upon nitrogen depletion, it is becoming competent in promoting the recognition
of NtcA-dependent promoters (Herrero et al., 2001) and, therefore, its own expression
increases (likely initiated at P1 in Med4). However, recent studies suggest an even more
complex network for the nitrogen control in cyanobacteria, as NtcA seems to respond to
a signal of the C-N balance of the cell rather than to the ammonium concentration (Her-
rero et al., 2001). Additionally, a cooperation with other factors might be possible. The
group 2 sigma factor SigC of Synechocystis PCC 6803 was already shown to be involved
in nitrogen-dependent promoter recognition (Asayama et al., 2004).

All three mapped TIS for lexA, umuD and PMM1427 are part of the predicted LexA
binding site (Fig. 3.7). Thus, in case of LexA binding these genes would be repressed and
might represent a part of the putative SOS regulon consequently.

3.3 Circadian rhythm and kai genes

The basic components described for the circadian clock of cyanobacteria are the clock
proteins encoded by kaiA, kaiB and kaiC. The basic timing of circadian rhythm is medi-
ated by KaiC phosphorylation as shown by in vitro experiments (Nakajima et al., 2005).
The influence of transcriptional regulation is controversially discussed in the literature.
Recently, the sufficient promoter region of kaiBC and the role of a constitutive negative
element inside the kaiA coding region of Synechococcus PCC 7942 was described, suggest-
ing these sequences to be essential for the circadian oscillation (Kutsuna et al., 2005).

A first step in this study was to identify these clock genes in the given set of marine
cyanobacteria. Surprisingly, all Prochlorococcus strains sequenced so far do not possess a
complete kaiA gene, although kaiB and kaiC are present at a similar genomic location as
in the Synechococcus strains which possess all three clock genes (Fig. 3.8).

In all these cyanobacteria, the two- or three-gene kai operon is found on the reverse com-
plementary strand upstream of the ribosomal protein operon starting with rpl21, coding
for the 50S ribosomal protein L21. However, a detailed and comparative sequence analysis
of the genomic location for the five marine cyanobacteria and the freshwater model strain
for circadian clock, Synechococcus PCC 7942, clearly revealed a genome reduction during
which the kaiA gene became deleted step by step: For the two Synechococcus strains,
WH 8102 and 7803, the full-length kaiA gene is annotated, although the distance to the
54 3. RESULTS

TIS rpl21 TIS kaiBC

Syn7942 rpl27 rpl21 kaiA kaiB kaiC
~ 2.7 Mb

Syn8102 rpl27 rpl21 kaiA kaiB kaiC

2.43 Mb

Syn7803 rpl27 rpl21 kaiA kaiB kaiC

2.37 Mb

Pro9313 rpl27 rpl21 kaiB kaiC

2.41 Mb

TIS kaiB
ProSS120 rpl27 rpl21 kaiB kaiC
1.75 Mb

TIS kaiB
Genome Reduction
ProMed4 rpl27 rpl21 kaiB kaiC
1.66 Mb

Figure 3.8: Arrangement of rpl27 and rpl21, encoding 50S ribosomal proteins, and kaiA, kaiB and
kaiC encoding the core clock proteins of cyanobacteria. From Synechococcus to Prochlorococcus a genome
reduction is clearly visible in deleting the kaiA coding region with a complete gene loss in the smallest
genomes, Med4 and SS120. The grey arrows indicate the identified promoter regions for kaiBC (Kutsuna
et al., 2005), for kaiB and rpl21, this study, for Med4 (Vogel et al., 2003a) and for a predicted TIS of rpl21
in PCC 7942.

adjacent rpl21 gene is already reduced, compared to PCC 7942. For Prochlorococcus MIT
9313 only short relics of the kaiA coding region were detected (Fig. 3.9), and for Prochlo-
rococcus SS120 and Med4 the complete coding region of kaiA is missing, resulting in a
newly created intergenic spacer region between kaiB and rpl21. The reduction process in
the kai gene operon, furthermore, correlates with decreasing genome sizes caused by an
ongoing compactation starting from about 2.7 Mb for Synechococcus PCC 7942 to 1.7 Mb
for Prochlorococcus Med4, the smallest known cyanobacteria (Fig. 3.8).

For the marine Synechococcus, no TIS for kaiB could be detected by RACE experiments.
Also all attempts mapping TIS for kaiA or kaiC failed. Only secondary mRNA 5’ ends
resulting from RNA processing rather than initiation of transcription were found, indicat-
ing low expression and/or fast degradation of these mRNAs. The use of RNA samples
from synchronised cultures collected at the maximum expression rate of the respective kai
gene might improve the results of the RACE experiments.
In contrast, for the Prochlorococcus strains, SS120 and Med4, the transcription initiation
sites of kaiB were mapped easily by 5’ RACE. A well conserved motif could be detected
next to the TIS of kaiB in Med4 and SS120 around position -60 of the promoter region,
suggesting an important regulatory site of the kaiB promoter. Further analysis of this
region identified this motif to be conserved between all six strains considered here (Fig.
3.10), although only Med4 and SS120 share a common promoter region for kaiB and rpl21
(Fig. 3.8). Furthermore, promoter mapping for the adjacent rpl21 gene revealed this site
belonging to the 5’ UTR of rpl21. The motif includes an inverted repeat, functionally it
might correspond rather to the mRNA leader region of rpl21 than constitute a part of
the promoter for the kai operon. For instance, it might play an important regulatory role
3.3. CIRCADIAN RHYTHM AND KAI GENES 55

for this ribosomal operon. Indeed, from enterobacteria it is known that functional boxes
within the mRNA leader sequences mediate the autogenous control of ribosomal protein
expression (Lindahl and Zengel, 1986; Zengel and Lindahl, 1994). Thus, a comparable
regulatory mechanism for ribosomal genes may operate in marine cyanobacteria as well.
In the case of Med4 and SS120 the promoter region of rpl21 is teethed together with the
promoter of kaiB as their core promoter elements overlay in part: TIS and -35 region
overlap in Med4; -10 and -35 region in SS120. Thus, a combined expression of both genes
may be assumed, which was already verified by RT-PCRs on RNA samples of synchronised
Med4 cultures (Julia Holtzendorff, pers. communication).

Pro9303 : ------------------------------------------------------------------------------GG
Pro9313 : --------------------------------------------------------------------------------
Syn8102 : MARPGLTIALLLTTPNLVDACQQWLPDTRYHSIVLSGPHQGQEQLDLVSTLEAQQEEIDAVVVEQQLLDASSRDQLLGRG
Syn7803 : MARPALTIALLLTSKELVDACTQWLPVNRYKPVDLHQAASGE---GLLEVLAHQREAVDAVVIEQSLLDETTREGLRRGG
Syn7942 : ---MLSQIAICIWVESTAILQDCQRALSADRYQLQVCESGEMLLEYAQTHRDQIDCLILVAANPSFRAVVQQLCFEGVVV

Pro9303 : LRRVQARNLQSRAVKRRTTLTDLA*LAIERQTKIIPRLGAFSQD*TLISERQNVGTKPSLIMALSCGSTHQHSLFFAVSG
Pro9313 : --DAFRLGI-CRAGLLSTAPL*PIWPDSQSREKKGSS-RV+APLVKIER*SLRVKTSVPSRR*SRLSLAAQHTSIL*CLL
Syn8102 : LLFPAVVVGEMKGHVDYHAEELHLAEDQLAQLGYTVDAAISRFLRQG-------RADGRSDDDGLASVDKLSRRLQERLG
Syn7803 : LLFPAVVVGELMGRVDYHPEEVHLPVDQLEQLGYNVDAAISRFLRQGQKEARPEDGSAPSSAESASSAWKLSSRLQERLG
Syn7942 : PAIVVGDRDSEDPDEPAKEQLYHSAELHLGIHQLEQLPYQVDAALAEFLRLAPVETMADHIMLMGANHDPELSSQQRDLA

Pro9303 : VLPALI-CLWILWLQA+NGS---DQLVAASFL-S*GEDFVDAEIDR*VIFEILALFAWP+IAS*SALFAD*SIT+VVEIH
Pro9313 : SLVFYQP*FT*GCCGRRSRE---SLISLLQPQLS*GEVFVDAEIDA*VIF+ILALFPLP+IAS*SALFAA*SIT+FVEIH
Syn8102 : YLG----VFYKRDPSRFLGS---LPTEERRELLESLQRTYRDLLI-SYFSDPAASNQALESFVNTAFFSDLPITRTVDIH
Syn7803 : YLG----VFYKRDPSRFLAN---LPPDEQRELLRSLQRTYRDLLI-SYFRDPAAANQALESFVNTAFFGDLPITRTVEIH
Syn7942 : QRLQERLGYLGVYYKRDPDRFLRNLPAYESQKLHQAMQTSYREIVLSYFSPNSNLNQSIDNFVNMAFFADVPVTKVVEIH

Pro9303 : IDLIDAFWQQFKLELHERDSSSGVLFALIDLSNHLWVMLQRWLSAEVPLSLISISSDLEDASEMQL-----
Pro9313 : MDLIDAFWQQFRLELHERESSPGVLFALLNLSKHLWAMLQHSLSAEVPLR*YQPVPI+-------------
Syn8102 : VDQIDEFWKQLRLEGNKSEFLQDYRLALLDVMAHLCEMYRRSIPPDIPLSGLASGRHRREADLPDAPEVSS
Syn7803 : MNLIDEFWKQLRLEGHKNDFLQDYRLALLDVMAHLCEMYRRSIPPEV-----ALAVPSEQRRSLMDSEVSS
Syn7942 : MELMDEFAKKLRVEGRSEDILLDYRLTLIDVIAHLCEMYRRSIPRET------------------------

Figure 3.9: Alignment of the annotated coding regions of kaiA from Synechococcus PCC 7942, WH 8102
and 7803 together with the long spacer regions between the genes kaiB and rpl21 of two low-light-adapted
Prochlorococcus strains, MIT 9313 and 9303. A few amino acids conserved between all five strains are
visible, whereby a putative reading frame for MIT 9313 and 9303 is interrupted by several stop codons
(labelled by * and +), indicating the step by step deletion of the kaiA coding region in Prochlorococcus
genomes.

Additional factors beside the Kai proteins of the cyanobacterial clock model are the cir-
cadian input kinases, CikA and predicted CikR, as well as the adaptive sensors, SasA
and predicted SasR, which represent the key input and output pathway components, re-
spectively. An important link to the metabolic status of the cell as well as a novel input
pathway to the circadian oscillator is given by the light-dependent protein A (LpdA), re-
cently described in Synechococcus PCC 7942 (Ivleva et al., 2005). A gene of the output
pathway in the circadian system might be cpmA, whose mutation produced a modified
circadian phasing phenotype for PCC 7942 (Katayama et al., 1999).

It was demonstrated earlier that strains of marine Synechococcus and Prochlorococcus

exhibit a cell synchrony under dark/light conditions (Jacquet et al., 2001), although the
existence of a real circadian clock has never been shown so far. Here, the complete genome
56 3. RESULTS

Syn7942 : kaiA-106nt-CCACAGTTTGACAGTTCTCAAAAGTTTCTGAGATGATAGAGGGCTGTTGTTTAAAAGGCGGCTCGGAACAGGTTTCTT-53nt-rpl21
Syn8102 : kaiA--24nt-CTGAATGTTGACGACAGGGACCTCGATGATCGAAGATCTTTGTTTGTCTTTC-GATTGCGGTTCAGGCTGAATCGGTT-29nt-rpl21
Syn7803 : kaiA--20nt-CTGAATCTTGACGAAGGGGACAGCAAAGGCCGAAGATCTTTGTTTGTCTTTC-GATTGCGGTTCAGGCCAAATCGGTT-29nt-rpl21
Pro9313 : kaiB-523nt-TTCAATCTTGACTAAAGGCGCCTAAACGCGGGAAGATCCTTTTTTCTCTCTC-GATTGCGAGTCAGGCCAAATCGGTC-65nt-rpl21
ProMed4 : kaiB--36nt-ATAAACCTTGACTAATTACACTATAAAATAAGAAGATAAACAAACATTCAA--GACTGCTAAAAAGGCTTAATCAAAT-33nt-rpl21
ProSS120 : kaiB--38nt-CTAAATCTTGACGGATTTAACCTCTAAGAGTGAGGATGTTATCTTGTCTTTC-GATTGTGTTACAGGCTTAATCGGAC-48nt-rpl21
- 35 - 10
TIS rpl21

Figure 3.10: Alignment of the conserved rpl21 upstream regions of Synechococcus and Prochlorococcus.
Conservation of TIS (often thymine), -10 (GAaGAT) and -35 (cTTGAC) region is visible for all strains.
These core promoter elements are highly similar to known E. coli standard promoters (except the thymine
at TIS). Red arrows indicate a conserved inverted repeat within the rpl21 5’ UTR, possibly indicating
functional equivalence to the control elements of enterobacterial ribosomal operons (Lindahl and Zengel,
1986; Zengel and Lindahl, 1994).

sequences were used to screen for genes, that encode orthologs of other clock proteins or
transmitting in- and output signals, described above. Among these periodosome compo-
nents not only kaiA is absent in Prochlorococcus strains, but also the gene for the circadian
input kinase, cikA, is missing in all marine strains analysed (Tab. 3.3). The genes kaiB,
kaiC, sasA as well as lpdA and cpmA could be detected in all marine strains with reason-
able e-values by BLASTp comparison to PCC 7942 coding sequences.

PCC 7942 Med4 SS120 MIT 9313 WH 8102

core kaiA - - - (*) SYNW0548
clock kaiB PMM1343 Pro1424 PMT1419 SYNW0549
genes kaiC PMM1342 Pro1423 PMT1418 SYNW0550
pathway in cikA - - - -
components out sasA PMM1077 Pro1121 PMT1099 SYNW0753
signal in lpdA PMM1560 Pro1714 PMT1730 SYNW2065
transduction out cpmA PMM1278 Pro1352 PMT0358 SYNW1608

Table 3.3: Overview of orthologous genes in Prochlorococcus Med4, SS120, MIT 9313 and Synechococcus
WH 8102 identified by BLASTp with known genes of Synechococcus PCC 7942, the circadian clock model.
(*) In MIT 9313 only short relics of the kaiA coding region exist (Fig. 3.9).

The putative circadian clock of marine strains appears to be reduced by deletion of the
key input pathway component, CikA, although another input pathway given by LpdA
could compensate the absence of CikA. Additionally, in Prochlorococcus the coding region
for the clock core component kaiA is deleted. Thus, a reduced periodosome might be
found for a reduced and tiny genome of the slow-growing Prochlorococcus cell that divides
on average once in 24 hours. The coupling of kaiB to ribosomal gene expression (rpl21)
was suggested here, which may provide further evidence for a simplified circadian regu-
lation mechanism in Prochlorococcus. Future experiments have to solve the question for
an autonomous circadian oscillator existing in marine strains. The analysis of the protein
components as well as studies on synchronised cultures could provide further insights into
the marine pacemaker.
3.4. TRANSCRIPTIONAL REGULATION OF THE CYANOPHAGE P-SSP7 57

3.4 Transcriptional regulation of the cyanophage P-SSP7

The podovirus P-SSP7 is very host-specific in infecting the high-light-adapted Prochloro-
coccus strain Med4 exclusively. Recently, the genome sequence of P-SSP7 became available
(Lindell et al., 2005) and allowed bioinformatic and experimental studies. Thereby, sev-
eral similarities to the well studied coliphage T7 were observed (several T7-like class I, II
and III genes including T7-like RNA polymerase), but also intriguing differences (host-like
genes: hli and psbA; int gene and a possible integration site). Whether the transcriptional
regulatory mechanism of T7 can be assigned to P-SSP7 is in dispute, because no signifi-
cantly similar promoter sites were found for P-SSP7, suggesting that this phage does not
belong to the otherwise highly conserved T7 group (Chen and Schneider, 2005). Thus,
a detailed analysis of P-SSP7 promoter regions in comparison to those from T7 phages
might give further insights into the transcriptional order of the P-SSP7 genome.

Each intergenic region (> 50 bp) of coliphage T7 contains one or more transcriptional
signals. All together, three strong early promoters and the early termination site for E.
coli RNA polymerase as well as 17 promoters and one termination site for the T7 RNAP
were studied intensively (Dunn and Studier, 1983). Thus, for cyanophage P-SSP7 tran-
scriptional signals were assumed to exist in intergenic regions as well. Their prediction
was based on a matrix (-10 box) for Med4 promoters (Vogel et al., 2003a), a matrix for
T7-like promoters (Imburgio et al., 2000) and using the TransTerm algorithm (Ermolaeva
et al., 2000), which detects rho-independent transcription terminators by searching for
stem-loop-structures (inverted repeats) followed by a row of T’s in the genome. In an
experimental approach, the 5’ ends of P-SSP7 RNAs were mapped using RACE (Bensing
et al., 1996), a rapid amplification of cDNA ends, which allows a sensitive detection of
transcription initiation sites and a differentiation from RNA processing sites.

The results are compiled in Table 3.4 including ten predicted bacterial promoter signals
(-10 box), seven predicted termination sites (terminator) and six RNA 5’ ends mapped
experimentally (5’ end). Surprisingly, no T7-like promoters were detected in P-SSP7 as
well as no other frequently occurring sequence motifs, which is in agreement with findings
of Chen and Schneider (2005), who suggested the transcriptional strategy for P-SSP7 to
differ from the otherwise closely related T7 group. On the other hand, the DNA of P-SSP7
was shown to be transcribed from left to right during infection (Lindell, pers. commu-
nication) like T7 DNA, indicating a transcription of phage class I genes from bacterial
promoter sites located near the left end of P-SSP7 DNA. Indeed, a Med4-like promoter
was predicted near the left end, upstream of gene 1, although it could not be verified
experimentally. This initial transcription might be terminated at predicted sites between
gp12 and gene 3 or upstream of gp1, encoding RNAP, which is different in the T7 model,
where the early expression of the phage RNAP is needed for the following stages of infec-
tion.

Non-coding gaps between coding sequences usually contain the transcription signals and
also RNase III cleavage sites (Dunn and Studier, 1983). Thus, genes were chosen for
58 3. RESULTS

signal DNA pattern position intergenic region

-10 box cagaat 52..57 upstream gene 1
motif ggttcaattcctctcccatcaattgc 91..116 upstream gene 1
5’ end t 113 upstream gene 1
motif ggttcgactccctcccatccaattgc 1650..1675 gp12; gene 3
5’ end t 1672 gp12; gene 3
terminator ccctgcaatgagcaggg 1675..1691 gp12; gene 3
-10 box cacact 2591..2596 gene 6; gene 7
-10 box caagtt 2629..2634 gene 6; gene 7
-10 box taattt 3566..3571 gene 10; gp0.7
-10 box tagagt 3577..3582 gene 10; gp0.7
-10 box caatgt 3585..3590 gene 10; gp0.7
terminator gcgattggtcaaaagccagtcgc 5006..5028 int; gp1
5’ end c 5032 int; gp1
-10 box taattt 5034..5039 int; gp1
5’ end c 5045 int; gp1
5’ end c 5060 int; gp1
-10 box tatgct 19430..19435 psbA; gene 28
motif ggtgcgagtcctccctactcaatttg 19725..19750 gene 28; gp10
5’ end t 19749 gene 28; gp10
terminator gggaccttcgggtccc 21031..21046 gp10; gp11
terminator ggaagctgcattagatgtggcttcc 24626..24650 gp12; gene 32
terminator gcacccgcaagggtgc 39402..39417 gene 49; gene 50
terminator gcccttcagggc 41165..41176 gene 50; gp19
-10 box tacttt 42978..42983 gp19; gene 52
-10 box cactgt 42988..42993 gp19; gene 52
terminator cccccgtaggggg 44063..44075 downstream talC

Table 3.4: Summary of transcriptional and other signals mapped to the cyanophage P-SSP7 genome.
Signals for Med4 promoters (-10 box) were predicted based on a matrix (Vogel et al., 2003a), terminators
by using the TransTerm software (Ermolaeva et al., 2000) and 5’ ends by RACE experiments. The resulting
DNA patterns are given as well as their absolute position in the genome and in the respective intergenic
region. A conserved motif appeared by aligning sequences around detected 5’ ends of gene 1, gene 3
and gene 29 (gp10). "gpX" are T7 designations for the homologous genes, and "gene X" designations
correspond to the gene order for P-SSP7 as set out in Figure 1.5.

RACE experiments, which possess non-coding regions and/or predicted signals upstream,
to identify 5’ ends emerging from transcription initiation or processed sites. The resulting
list contains 20 phage genes analysed experimentally:
gene 1, 3, 4, 8, 11 (gp0.7), 13 (gp1: RNAP), 14 (gp2.5), 17 (gp5), 19 (gp6), 20 (nrd), 24
(gp8), 26 (hli), 27 (psbA), 29 (gp10), 30 (gp11), 32 (gp13?), 36 (gp17), 50, 51 (gp19), 52
(designation given in Fig. 1.5).

RACE experiments for P-SSP7 revealed specific signals for gene 1, gene 3 (the same for
gene 4) and gene 29 (gp10). These 5’ ends (Fig. 3.11) might originate rather from RNase
cleavage than transcription initiation as no specific differences were found whether TAP
treatment was included or not. Furthermore, the mapped first nucleotide in each case
turned out to be a uracil (or thymine in DNA) - a pyrimidine base, and not the purine
usually found at an RNA primary 5’ end resulting from initiation of transcription. An
appropriate cyanobacterial -10 box similar to the tANNNT consensus sequence was absent
in all three cases. Instead, an alignment of the sequences around these three sites resulted
in a new motif, which is shown in Table 3.4 and discussed in detail below.
3.4. TRANSCRIPTIONAL REGULATION OF THE CYANOPHAGE P-SSP7 59

In T7 phages it was shown that the primary phage transcripts are processed at ten RNase
III cleavage sites to provide the mRNAs observed in vivo (Dunn and Studier, 1983). RNA
cleavage might be assumed for P-SSP7 as well, although a prediction of these sites is diffi-
cult, because a specific sequence pattern for RNase III is not known. On the other hand,
signals predicted by TransTerm (Ermolaeva et al., 2000) include inverted repeats causing
double-stranded regions, which might be recognised by RNase III as well.

3.4.1 Mapping of 5’ ends by RACE experiments

Total RNA of Prochlorococcus Med4 and phage P-SSP7 was extracted at different time-
points during lytic infection (Lindell et al., 2005) and kindly provided by Dr. Debbie
Lindell as the basis for 5’ RACE experiments. The primer sets for several phage genes
(see above) as well as for a few host genes were chosen by using the annotated genome
sequence of P-SSP7 (Sullivan et al., 2005) and Med4 (Rocap et al., 2003) to determine the
5’ ends of their transcribed mRNAs.

M TAP+ TAP- M TAP+ TAP- M TAP+ TAP-

gene001 T=1 gene003 T=1 gene004 T=1

M TAP+ TAP- M TAP+ TAP- M TAP+ TAP-

gene029 T=4 PMM0684 T=6,7 PMM0819 T=6,7

Figure 3.11: 5’ RACE experiments for the cyanophage P-SSP7. Signals of maximum strength were
detected for gene 1, gene 3, gene 4 and gene 29 (gp10) (designation given in Fig. 1.5) in the TAP-treated
(TAP+) but also for the non-treated (TAP-) control sample, indicating processed but stable RNA 5’ ends.
The signals for 5’ ends of the host (Prochlorococcus Med4) genes PMM0684 and PMM0819 are of lower
strength and represent unprocessed RNA ends in the TAP+ variant (absent signal in the control variant,
TAP-, for the same length).

Only for about the half of chosen genes RACE signals were observed (genes 1, 3, 4, 8,
13 (gp1), 17 (gp5), 20 (nrd), 29 (gp10)). Thereby, two different types of 5’ ends may be
distinguished:
The first group, upstream of gene 1, gene 3 (same 5’ end as for gene 4) and gene 29
(gp10), see Figure 3.11, revealed signals of maximum strength by RACE experiments for
TAP+ and TAP- (control) variants, indicating processed 5’ ends with only one phosphate
residue. Nevertheless, the three experimentally identified sites could be aligned to define
a putative motif, shown in Figure 3.12b.
60 3. RESULTS

For all the other analysed sites (except for gene 13, gp1: RNAP; see 3.4.3, p. 62) no
conserved region was identified, including those similar to known bacterial promoters or
described phage promoters.

For the phage encoded hli and psbA genes, a signal could not be detected by RACE ex-
periments, suggesting that they are either not transcribed alone, or their transcription is
too weak for detection. Alternatively, they may be co-transcribed with the essential phage
capsid genes. Indeed RT-PCR products spanning the intergenic regions between the phage
psbA and capsid genes suggests that they are co-transcribed (Lindell et al., 2005).

Experiments using Prochlorococcus Med4 DNA arrays identified host genes up-regulated
during infection (Lindell, unpublished): Are these Med4 genes transcribed by the phage
RNAP or the host RNAP? The analysis of the 5’ ends of host genes resulted in two per-
fect bacterial promoters for PMM0684 and PMM0819 (Fig. 3.11 and Fig. 3.12). Thus,
a transcription by host RNAP is very likely: The -10 box fits perfectly to the previously
derived model for Med4 promoter elements (Vogel et al., 2003a); the -35 region has a
perfect match to the TTGAC sequence in the -35 region of E. coli, which also was the
case for a subgroup of genes in Med4, like rpl21 and ccmK. Interestingly, the upstream
regions (Fig. 3.12) as well as the coding regions (not shown) of these two host genes are
very similar to each other.
5’ RACE experiments for three other Med4 genes, PMM0368, PMM1500 and PMM1501,
gave no interpretable signal.

3.4.2 Putative motif upstream of genes 1, 3 and 29

The alignment of regions around the mapped 5’ ends of gene 1, gene 3 and gene 29 (gp10)
revealed a conserved motif, listed in Table 3.4 and shown in Figure 3.12. Interestingly, ad-
ditional hits for the putative motif, which might also represent the possible RNAP binding
site, do not exist within intergenic regions of the P-SSP7 genome, which was surprising as
the related RNAP of phage T7 has 17 highly conserved promoters. On the other hand,
the motif identified here is very uncommon: It has only weak similarity to the known T7
promoters, although the phage P-SSP7 is classified as T7-like and harbours the gene 13
(gp1) encoding a typical T7-like RNAP. The RNAP of T7-type phages is well-known to
recognise highly conserved motifs. On the other hand, a computational search for more
closely related sequence motifs failed to identify any candidates which might be more suit-
able as promoter sequences. Additionally, all three detected 5’ ends of genes 1, 3 and 29
begin with an uracil base (thymine in DNA), which is the least likely case for all known
transcription initiation sites.

It is already known that 5’ ends for mRNAs of T7-like phage genes originate from multiple
processes including bacterial and phage transcription as well as RNase III cleavage (Dunn
and Studier, 1983). Especially, because the observed RACE signals for P-SSP7 gene 1,
gene 3 and gene 29 (gp10) indicate RNA processing, it is likely that these 5’ ends emerge
from RNase cleavage. Inverted repeats are located next to the identified 5’ ends (Fig.
3.12a). These are likely to form double-stranded regions, which might serve as recognition
site for RNase III. One of five RNase III cleavage sites located in the early region of T7
3.4. TRANSCRIPTIONAL REGULATION OF THE CYANOPHAGE P-SSP7 61

A motif
gene001 : ATTCATGCGTGATCGAGAGGTTCAATTCCTCTCCCATCAATTGCTACTCACTGAGAGTAGCCAATTTTCAC

inverted repeat
motif
gene003 : TGCAAGGACATAAGGACGGGTTCGACTCCCTCCCATCCAATTGCCCTGCAATGAGCAGGGTTTACTGACTA

inverted repeat
motif
gene029 : TAAAGCCGATCAGGGAGGTGCGAGTCCTCCCTACTCAATTTGGCTTTTTGCCCTCCAAGGAGGATACCAAT

inverted repeat
TIS for bacterial promoter
- 35 - 10
gene013 : AGCGATTGGTCAAAAGCCAGTCGCTTTCTTAATTTATAACCTATCCACTAAAGGACAAATATCCAACTTAT

inverted repeat
B C
gene001 : AGAGGTTCAATTCCTCTCCCATCAATTGCTACT gene001 : AGAGGTTCAATTCCTCTCCCATCAATTGCTACT
gene003 : ACGGGTTCGACTCCCTCCCATCCAATTGCCCTG gene003 : ACGGGTTCGACTCCCTCCCATCCAATTGCCCTG
gene029 : GGAGGTGCGAGTCCTCCCTACTCAATTTGGCTT gene029 : GGAGGTGCGAGTCCTCCCTACTCAATTTGGCTT
T7 : --TAATACGACTCACTATAGGGAGA--------
motif : GGTNCNANTCCNNNCNNNNCNATT

P-SSP7 gp10: TCAGGGAGGTGCGAGTCCTCCCTACTCAATTTGGCTT gene001 : AGAGGTTCAATTCCTCTCCCATCAATTGCTACT

AACY011174 gp10: ACAGGGAGGTTCAAGTCCTCCCATCTCTATTGGAGCG gene003 : ACGGGTTCGACTCCCTCCCATCCAATTGCCCTG
AACY011067 gp10: TTGGGGAGGTGCAATTCCTCCCCTGCCTATTGGCGTT gene029 : GGAGGTGCGAGTCCTCCCTACTCAATTTGGCTT
T7 : ----------TAATACGACTCACTATAGGGAGA
D Yfr7

PMM0684 : AATTTTATTCCCAGAATCAAGAGAAAAAGGTAAA-TAAAAAATTCAATACTTGACGAACTA-TACTCACTACACTACAGTAGTAGTA-18nt-ATG
PMM0819 : AAATAGCTTTCCAGAATTAAGAGAAAAACTCAAAGTTAATAATTAAATACTTGACGATTTAATTACCATTGGCTTATATTAGTAGTA-18nt-ATG
- 35 - 10 TIS
PMM0820

Figure 3.12: (A) 5’ ends determined for the phage P-SSP7 gene 1, gene 3, gene 29 (gp10) and gene 13.
Experimentally mapped start sites are red and underlined, predicted termination sites (inverted repeats)
are marked by blue lines. (B) A consensus sequence is suggested for the upstream region of P-SSP7 gene
1, gene 3, gene 29 (gp10) (top). Motif searches inside a set of environmental samples (Venter et al.,
2004) revealed two hits similar to the motif identified for the gene 29 (gp10) upstream region in P-SSP7
(bottom). (C) Comparison of the conserved sites upstream of P-SSP7 genes 1, 3 and 29 (gp10) to the
consensus sequences of T7 promoter (Imburgio et al., 2000) revealed a very low similarity. (D) Upstream the
two host genes PMM684 and PMM0819 of Med4 a perfect bacterial promoter site was detected including
conserved -10 and -35 region. Adjacent coding sequences, gene for Yfr7 RNA and PMM0820, are indicated
by black lines above and below the sequences.

was shown to possess a pU at the 5’ terminus (Robertson et al., 1977), like it was found for
5’ ends of P-SSP7 genes 1, 3 and 29. Additionally, the site for gene 1 was observed at each
time-point during infection investigated, also at the early stage (T = 1, Fig. 3.11) at which
the phage RNAP may not yet exist. If this would be the case, that phage RNAP is not
transcribed early, then the promoter of gene 1 must be recognised by the host polymerase.

Thus, the motif might share more similarities to a RNase III cleavage site than to a T7
RNAP promoter site, and a verification of the motif origin in P-SSP7 could be highly
advantageous. However, experiments like in vitro studies about P-SSP7 RNAP and its
promoters have not yet been done and would be extremely challenging for this new phage-
host system.

Therefore, further computational searches were attempted by using a set of environmental

sequences from metagenomic analysis of the Sargasso Sea (Venter et al., 2004). An initial
BLASTp search was done to identify orthologous genes of P-SSP7 gene 29 (gp10, encoding
a capsid protein), which harbours the motif upstream of its coding region. For the four
62 3. RESULTS

newly found environmental gp10 homologs the upstream regions were aligned with the
P-SSP7 sequence (upstream region of gene 29): Two of the environmental upstream se-
quences appeared to be very similar to P-SSP7 including the motif (Fig. 3.12b). Although
these additional hits of the P-SSP7 motif can not be seen as a proof of its function, they
may serve as strong evidence that this conserved site possesses a biological meaning for
the phages, maybe for the transcription of the mRNA encoding the capsid protein, gp10.

By searching the Med4 host genome with the putative motif, not a single match was de-
tected within non-coding regions, whereas three hits were found inside tRNA genes and
tmRNA. One hit of the motif occurred at the 3’ end of tRNA-Leu2, the tRNA which was
suggested to serve as integration site for P-SSP7 (Sullivan et al., 2005). Another hit lies at
the 3’ end of tRNA-Gln1 and at the 3’ end of the short tmRNA piece (mature two-piece
tmRNA exists in Prochlorococcus, predicted by Gaudin et al. (2002) and shown in this
work (Fig. 3.13 and Axmann et al. (2005)).
Both tRNA genes, tRNA-Leu2 and tRNA-Gln1, are each closely located to a downstream
gene (ndhL and PMM1561, respectively), which might be co-transcribed with the tRNA,
resulting in a mixed tRNA-mRNA polycistronic transcript. Thus, a later processing step
at the tRNA 3’ end to release the tRNA would be required. A similar situation was
described for the metY-nusA-infB operon of E. coli, which is cleaved by RNase III in
a hairpin structure downstream from the tRNA, separating the tRNA from the mRNA
(Regnier and Grunberg-Manago, 1989).
The tmRNA of Prochlorococcus is encoded by one circularly permuted gene (Gaudin et al.,
2002) and is found in an unusual two-piece form in the cells (this work, Fig. 3.13 and
Axmann et al. (2005)). Thus, for the mature two pieces RNA processing is needed as well.
Taking all entries of the motif into account, one might suggest that the motif first ob-
served in P-SSP7 seems to possess a biological function not only in the phage but also in
its host, Med4, by representing a potential cleavage site for an RNase. In addition, one
has to conclude that the motif identified in this study is also the closest candidate for a
cyanophage-type promoter up the now.

3.4.3 Transcription of phage-specific RNA polymerase

Multiple 5’ ends of phage mRNAs can be explained by a combination of processes, includ-
ing bacterial and phage transcription and RNase III cleavage as known from T7 (Dunn
and Studier, 1983). For P-SSP7, upstream of gene 13 (RNAP) three different ends were
detected by RACE experiments (Fig. 3.12).
The region around one of these 5’ ends appeared to be similar to a bacterial promoter
site, which might represent a small hint for one of the most interesting questions about
this host-phage system: Is the phage RNAP transcribed and translated prior to the other
phage genes? The putative bacterial promoter upstream of gene 13 might indicate the
possibility that the phage RNAP could be transcribed initially by the host RNAP and,
moreover, independently of the other phage genes.
Another 5’ end found for gene 13 is located immediately downstream of a predicted ter-
mination site. This termination site consisting of an inverted repeat may also serve as
secondary structure for an RNase cleavage as it is described for several closely located
promoter and cleavage site structures in the coliphage T7 (Dunn and Studier, 1983), see
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 63

also Figure 1.6b.

3.4.4 Missing promoter site upstream gene 30

The fact that the RACE failed in detecting a 5’ end upstream gp11 (tail tubular protein)
in P-SSP7 can be explained by comparison to the phage T7. A transcription termination
site exists downstream gp10 (capsid protein) in T7 but no promoter for RNAP between
this termination site and gp11 and gp12, so that these genes have to be transcribed by
read-through of the termination site, which represents a part of the transcriptional strat-
egy in T7-like phages. In P-SSP7 a termination site was also predicted downstream gp10
(Tab. 3.4), so that the absent promoter site for gene gp11 might resemble the T7 case.

3.5 Small RNAs in marine and other cyanobacteria

3.5.1 Known and abundant RNAs of marine strains
Total RNA samples from the four marine cyanobacteria Prochlorococcus Med4, MIT 9313,
SS120 and Synechococcus WH 8102 were separated on high-resolution polyacrylamide gels
to get an overview of the presence of small RNAs. This analysis showed abundant RNA
molecules with sizes in the range 50 to 250 nt (Fig. 3.13). A particularly abundant class
of RNAs in the 70 to 90 nt size range indicates the location of tRNAs in this gel, which
was confirmed by hybridisation to the tRNASer [GCU]. The hybridisation signal for this
tRNA was located at the upper end of this abundant cluster of bands, consistent with
the fact that it is the largest annotated tRNA in these genomes. Several small RNAs
migrated above the tRNA cluster and very few below it (indicated by the weakly visible
bands below the tRNAs). These bands collectively indicated the occurrence of abundant
small mRNAs, ncRNAs and precursors to tRNAs and rRNAs (Axmann et al., 2005).

Small eubacterial RNAs besides the abundant transfer and ribosomal RNAs, however,
very rarely reach a concentration that allows direct identification in a gel. For known
RNA species and their possible precursors or degradation products, information on their
expression can be gained from hybridisation. Here, oligonucleotide probes were used for
the scRNA and tmRNA and, as controls, the 5S rRNA and tRNASer [GCU], which was
predicted to be the tRNA with the highest molecular mass. The lengths of the scRNAs
in the four strains vary between 90 and 100 nt, in accordance with the varying lengths
of the respective annotated ffs genes. The 5S rRNA was detected as a very abundant
RNA species together with two precursors. Furthermore, the results of these Northern hy-
bridisations confirmed that Prochlorococcus tmRNA is indeed composed of two separate
molecules (Gaudin et al., 2002).

Several additional bands in the investigated size range indicated the presence of addi-
tional abundant small mRNAs or ncRNAs. The lack of specific oligonucleotide probes for
hybridisation, however, made it difficult to get information about these. Thus, a com-
putational prediction was used to identify candidates for further testing (Axmann et al.,
64 3. RESULTS

Figure 3.13: Small RNAs in marine Cyanobacteria, shown by Axmann et al. (2005). About 10 µg of total
RNA from Prochlorococcus strains MIT 9313 (MIT), SS120 (SS1) and Med4 (MED) and from Synechococ-
cus WH 8102 (WH8) was analysed by staining a 10 % polyacrylamide gel with ethidium bromide (center)
and by Northern blot hybridisation with DNA-oligonucleotides directed against known RNA molecules
such as scRNA (ffs gene product), the separate 5’ and 3’ ends of tmRNA and, as controls, tRNASerin and
5S rRNA. Two distinct precursors of the 5S rRNA were detected. Selected bands have been labelled by
arrows in the hybridisation and in the gel picture and their sizes (nt, nucleotides) are indicated (Axmann
et al., 2005).

2005).

3.5.2 Computational screening identified novel RNA species

The algorithm for identification of new, small RNAs was developed in cooperation with
Philip Kensche (Kensche, 2004) and basically focused on sequence and structure simi-
larities. Again, the advantage of a multi-genome comparison, between Prochlorococcus
SS120, Med4, MIT 9313 and Synechococcus WH 8102, was used by transforming the
initial, pairwise sequence alignments (BLASTn) into multi-sequence clusters via single-
linkage clustering to increase the sensitivity of ALIFOLDz (Washietl and Hofacker, 2004;
Washietl et al., 2005), the final scoring algorithm for each cluster. Also, single clusters
were scored by ALIFOLDz, although the scoring method was found to be sensitive to the
number of sequences in the alignment. Thus, a Z-score cut-off of -4 was considered as a
soft cut-off for both, alignments and single clusters.
An overview of the computational screening is displayed in chapter Materials and Meth-
ods and a summary of the highest scoring clusters is given in the Appendix, Figure 3.
Detailed information on all clusters predicted by this method, including the positions of
all sequences, is available online (Kensche, 2005) and the terminology introduced there is
followed here subsequently.

Although the sequence similarities between the predicted RNA elements in cyanobacteria
and other organisms were weak, for many of the clusters, clues for their possible function
could be obtained from the literature. These included elements that, according to loca-
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 65

tion or structure, might be functionally related to enterobacterial mRNA leader regions

mediating the autogenous control of ribosomal protein and rRNA expression (clusters 5,
92, 227, 228) (Zengel and Lindahl, 1994; Lindahl and Zengel, 1986), the rpoBC leader
region (cluster 245) (Barry et al., 1980) and the likely terminator (cluster 226). It was
decided against direct experimental analysis of these elements, which are less likely to be
novel types of ncRNAs. Additionally, two possible riboswitches for thiamine pyrophos-
phate (cluster 2) (Winkler et al., 2002) and cobalamin (cluster 101) (Nahvi et al., 2004)
were excluded from further experimental investigations (Axmann et al., 2005).

3.5.3 Experimental verification of predicted ncRNAs in Prochlorococ-

cus Med4
In the following, it was focussed on the analysis of small RNAs of the high-light strain
Prochlorococcus Med4. In the remaining clusters, all candidate sequences from Med4 were
tested by Northern hybridisation. Thereby, three different ncRNAs and a group of four
related ones yielded strong signals with RNA preparations from Med4. Each of these seven
candidate regions was probed for transcripts from both strands. Because some of these
ncRNAs have a phylogenetic distribution beyond Prochlorococcus (see below), a new gene
designation was introduced, yfr (for cyanobacterial functional RNA-coding gene), and Yfr
for the respective RNAs. Each of these genes is discussed in detail in the following sections
(Axmann et al., 2005).

Yfr1: a small RNA encoded between guaB and trxA

The yfr1 gene was detected in three of the four cyanobacteria in the intergenic region sep-
arating guaB and trxA (Fig. 3.14). Although the two adjacent genes guaB and trxA are
located in a similar genomic arrangement in SS120, a yfr1 gene was not found at this or
any other genomic position nor indicated by a Northern hybridisation signal. This result
is in agreement with the high sequence divergence of the guaB-trxA intergenic spacer in
SS120 compared to Med4, MIT 9313 and WH 8102 (Axmann et al., 2005).

The direction of yfr1 is conserved between Med4, MIT 9313 and WH 8102. It is transcribed
in the same direction as the mRNAs from two close-by neighbouring genes, indicating the
possibility of co-transcription. Therefore, it was searched for the presence of specific tran-
scriptional initiation sites (TIS) for yfr1 and for trxA by rapid amplification of cDNA ends
(RACE). A conserved TIS was mapped for yfr1, indicating that this transcript originates
from a specific promoter (Fig. 3.14a) and reducing the likelihood that it is co-transcribed
with guaB. Transcription of the adjacent trxA gene, encoding the redox regulator thiore-
doxin, was found to initiate approximately 100 bp downstream of the 3’ end of the yfr1
gene (Fig. 3.14a); co-transcription of yfr1 with trxA is thus unlikely. In SS120, the lack of
the yfr1 TATA box, and the fact that the trxA TIS and TATA box are shifted upstream
by about 20 bp compared to the other three strains (Fig. 3.14a), lends additional support
for the absence of a yfr1 gene (Axmann et al., 2005).
66 3. RESULTS

Strain RNA gene Length of Adjacent protein- Orientation

name RNA in nt coding genes
MED4 yfr1 54 trxA and guaB ← ⇐ ←
yfr2 94 PMM0363 and PMM0364 → ⇒ →
yfr3 95 PMM0686 and PMM0687 → ⇒ ←
yfr4 94 PMM0404 and phdC ← ⇒ →
yfr5 89 PMM1027 and PMM1028 → ⇐ →
yfr6 244 PMM0659 and PMM0660 → ⇐ ←
yfr7 220 purK and PMM0684 → ⇒ →
SS120 yfr2 90 rpsU and Pro0591 ← ⇐ ←
yfr6 239 Pro1007 and purK → ⇐ ←
yfr7 175 Pro1007 and purK → ⇐ ←
MIT 9313 yfr1 57 guaB and trxA → ⇒ →
yfr2 87 PMT1567 and PMT1568 → ⇐ ←
yfr7 175 purK and PMT0671 → ⇒ ←
WH 8102 yfr1 56 trxA and guaB ← ⇐ ←
yfr2 n.t. overlapping SYNW1139
yfr3 n.t. SYNW1140 and SYNW1141 → ⇒ ←
yfr7 174 SYNW1306 and purK → ⇐ ←

Table 3.5: Summary of identified ncRNA genes in Prochlorococcus MED4 and their orthologues in three
related strains of marine cyanobacteria; n.t., not experimentally tested (Axmann et al., 2005).

Although direct information on cyanobacterial RNAs is scarce (Reyes et al., 1997; Golden
et al., 1986) and not a single study exists for marine cyanobacteria, the half-lifes of eubac-
terial mRNAs are frequently in the range of a few minutes. In contrast, Yfr1 is extremely
stable as a half-life of more than 60 minutes was measured after transcriptional arrest was
induced by rifampicin (Fig. 3.15).
The expression of many bacterial regulatory RNAs is stimulated by varying environmen-
tal cues as by the stress response in which these RNAs have to play a role. Therefore,
a variety of stress conditions and their possible impact on the accumulation of ncRNAs
were tested.

Figure 3.16 shows a series of Northern hybridisations with RNA samples from cells that
had been depleted of nitrogen, phosphate or iron, exposed to higher intensities of white or
of blue light, or treated with 2 µM 3-(3,4-dichlorophenyl)-1, 1-N-N’-dimethylurea (DCMU)
to induce oxidative stress or grown at elevated or lowered temperatures (30◦ C and 15◦ C).
Normalisation of loaded RNA used 5S rRNA as an internal standard to compensate for
small RNA sample loading differences; however, Yfr1 levels were unaffected by any of
these conditions (Axmann et al., 2005).
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 67

Figure 3.14: Experimental screen for the presence of an RNA-coding gene in the guaB-trxA intergenic
region. (A) Sequence alignment of the guaB-trxA (guaB: sequence not shown, located upstream of yfr1)
intergenic region visualises the conserved yfr1 gene labelled by the bar above the alignment and its tran-
scriptional initiation site in three of the analysed strains (MED, Med4; MIT, Prochlorococcus strain MIT
9313; WH8, Synechococcus WH 8102) but not in Prochlorococcus strain SS120 (SS1). Transcriptional
initiation sites (TIS) and the deduced -10 elements are indicated. (B) Northern blots show a signal for
Yfr1 at a size of 54, 56 and 57 nucleotides (nt) for Med4, WH 8102 and MIT 9313, respectively. No signal
with RNA from SS120 confirms the absence of this gene in this strain, as was predicted from the sequence
data.

Yfr2 to 5: A new family of related short RNAs

In top scoring cluster 194, a family of structurally highly similar RNAs (Yfr2, Yfr3 and
Yfr4) was predicted (Fig. 3). Subsequent local alignments identified yet another similar
sequence in Med4, and at least one homologue each in SS120, MIT 9313 and WH 8102
(Axmann et al., 2005).

Northern hybridisations with oligonucleotide probes specific for each of these candidate
genes in Med4 yielded distinct bands of 89 to 95 nt. RACE mapping of 5’ ends further
confirmed that all four loci are transcribed in this organism (Fig. 3.17). The RNAs Yfr2
through Yfr5 in Med4 and their homologues in the other genomes are each encoded by
distant genomic loci and the position of their genes is not fixed within the four investigated
genomes with respect to adjacent genes (Tab. 3.5). The sequence comparison shows that
Med4 Yfr2 and Yfr5 on one hand and Yfr3 and Yfr4 on the other hand are more similar
to each other (Fig. 3.17a). The predicted secondary structures of the Yfr2-5 ncRNA
family in MED4 are highly conserved with a GGAAACA repeat within the loop of the
predicted 5’ hairpin (Fig. 3.17c). Among the different tested environmental conditions,
the amount of Yfr2-5 was affected by temperature (up at 15◦ C and down at 30◦ C) as well
as by nitrogen limitation and incubation in blue light (Fig. 3.16) (Axmann et al., 2005).
68 3. RESULTS

0 4 8 16 32 64 128 MED 0 4 8 16 32 64 128 MED

Yfr6
Yfr7 Yfr2
Yfr1 Yfr5

5S 5S

Figure 3.15: Determination of half lifes for Yfr 1, Yfr2 and Yfr5 to 7 in Prochlorococcus Med4, shown in
Axmann et al. (2005). The time in minutes after the addition of rifampicin is given on top. Hybridisation
to 5S rRNA served as control (Axmann et al., 2005).

Yfr6: A long RNA in Med4 and SS120

The yfr6 gene was predicted in cluster 53 (Fig. 3). This cluster included nine different
sequences, among which only yfr6 in Med4 and SS120 may code for a functional RNA.
The seven other sequences each have only about 40 nucleotide positions from their respec-
tive 5’ untranslated region in common with Yfr6. That was sufficient to cluster all nine
sequences together, but these other seven sequences included mRNAs for open reading
frames. In contrast, Yfr6 from the two strains, Med4 and SS120, each have an extended
sequence and structural similarity to each other (Axmann et al., 2005).

In Med4, yfr6 is located between the hypothetical PMM0660 gene and PMM0659, the
latter encoding 322 amino terminal residues of a DNA ligase. The region is framed by
trnS and nrdJ (encoding a B12-dependent ribonucleotide reductase). In SS120, the nrdJ-
trnS region lacks the yfr6 gene, which instead is located 448 nt downstream of another
ncRNA gene, yfr7. Despite the different genomic locations, Yfr6 sequences from the two
strains show a nucleotide identity of approximately 70 % to each other. A Northern blot
signal for Yfr6 is restricted to Med4 and SS120 and no signal was found in WH 8102
and MIT 9313. This 244 nt RNA had a half-life of approximately 2 minutes in Med4.
In Med4, blue light and incubation in the cold elevated the expression of Yfr6 compared
to white light or darkness. In addition, expression was reduced upon nitrogen depletion
and under high light conditions (Fig. 3.16). The yfr6 locus could also code for a 33
amino acid peptide as there is a possible reading frame that is conserved between Med4
and SS120 that begins at nucleotide 97 of the Yfr6 transcript in MED4. This situation,
a relatively long transcript with strong structural potential and a very short centrally
located reading frame, resembles the RNAIII from Staphylococcus aureus, a riboregulator
from which the 26 amino acid delta-hemolysin peptide is also translated (Tegmark et al.,
1998). In the hyperthermophilic archaeon Sulfolobus solfataricus, recently as many as
13 sense strand RNA sequences have been found that were encoded either within, or
overlapping, annotated open reading frames (Zago et al., 2005).

Yfr7: A conserved RNA located downstream of purK

The yfr7 gene is located downstream of purK (encoding phosphoribosylaminoimidazole

carboxylase) in all four strains analysed here (Tab. 3.5). This conserved location of a
ncRNA gene downstream of purK gene was already observed in freshwater cyanobacteria
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 69

Figure 3.16: Test of transcript accumulation of Yfr1-7 from Med4 (MED) under different conditions,
shown in Axmann et al. (2005). The left side shows the Northern hybridisations for which the following
conditions were used: nutrient depletion (phosphate (P-), nitrogen (N-), iron (Fe-)); blue light for three
hours (3 h); controls under blue (Blue), white (White) and no light (Dark); oxidative stress mediated by
the application of DCMU; low (15◦ C) and high (30◦ C) temperatures; and high light intensity (50 µE). For
comparison, 5S rRNA was hybridised as an internal standard and the mRNA of gene PMM3822n which,
with a length of approximately 250 nucleotides, was taken as an example for a small mRNA. Additional
controls by quantitative RT-PCR for the genes isiB (Fe), glnA (N), pstS (P) and hli8 (high light) were
carried out to confirm the effects of nutrient depletion or high light. The amounts of these mRNAs were
enhanced by a factor of 79.7 (isiB), 5.8 (glnA), 2.8 (hli8) and 4.0 (pstS) under the respective treatment
compared to standard conditions. Yfr6 shows an inconstant signal; for example, at cold, blue/white light,
N-, Yfr2 to Yfr5 were hybridised with the consensus oligonucleotide y_gen (Fig. 3.17). The band intensities
were quantified and normalised to the amount of 5S rRNA as an internal standard (right).

for the 6Sa RNA encoded by the ssaA gene (Watanabe et al., 1997).

At first, the search strategy identified this gene only in Med4 and SS120 (Fig. 3), due to
the fact that in MIT 9313 and WH 8102 this corresponding region is located within an-
notated mRNA genes. These hypothetical genes, PMT0670 in MIT 9313 and SYNW1307
in WH 8102, are located on the forward strand. Their expression was not detected, but
strong signals were found for Yfr7, which is transcribed from the complementary strand.
The sequence of Yfr7 is highly conserved between the four strains (Fig. 4). Rifampicin
tests showed this RNA to be stable (half-life > 1 hour). In Med4, expression of Yfr7 was
not affected by conditions employed in Figure 3.16.

3.5.4 Yfr1 exists in diverse cyanobacteria

Further computational screening detected additional entries of a yfr1 gene in more dis-
tantly related cyanobacterial strains like Synechococcus PCC 7942, Thermosynechococcus
elongatus, Synechocystis PCC 6803. Again, this small RNA was found to be encoded
inside the guaB-trxA intergenic spacer. Compared to other eubacterial ncRNAs (Rivas
et al., 2001; Tjaden et al., 2002), Yfr1 is one of the shortest bacterial ncRNAs, with a
length of about 50 to 60 nt (Fig. 3.18b). No peptide reading frame within yfr1 is conserved
between any of the strains.
70 3. RESULTS

A TIS yfr2-5
y2a-5aM
TATA
Yfr2 MED : CATATTAAAAAGGTGTGTAGGAGAGGTTTTACTGAAACAGTGGAAACAAGGAAACACTTGATTTAGTTAAACCAAGGAAAGACCTCTAT---TAGGGGTCTTTTTTTT
Yfr5 MED : GAAATTTAAATTGTGTGTAGGAGAGGTTTTATTAAATCAGTGGAAACAAGGAAACACTCGATTTATTAGAACCAATTTAGGCCCCTT-----TAGGGGTCTTTTTT--
Yfr3 MED : TATGTTATATATGTGTGTAGGAGAAGTTTTACTGAAACAGTGGAAACAAGGAAACACTTGGTTTAGTAAAACTATATAAAGATCTCTAGAAATAGAGATCTTTTTT--
Yfr4 MED : TAGTTTATTGTTGTGTGTAGGAGAGTTTTTACTGAAACAGTGGAAACAAGGAAACACTTGATTCGGTAAAACCAGAAAAAGACCTCTAGAAATAGGGGTCTTTTTTT-
* 20 * 40 * 60 * 80 * 100
conserved loop

B C

Yfr2 Yfr5 Yfr3 Yfr4

95
94
Yfr2
89
MED

Figure 3.17: Comparison of Yfr2, Yfr3, Yfr4 and Yfr5 from Med4. (A) Sequence comparison of the yfr2
through yfr5 coding regions of MED4. Transcriptional initiation sites (TIS) and the deduced -10 elements
are indicated. The location of specific oligonucleotide probes y2aM, y3aM, y4aM and y5aM used in Figure
3.17b and in 5’ RACE and of the y_gen consensus probe used in Figure 3.16 is indicated by the lines with
black diamonds on the ends on top of the alignment. (B) Signals for the four individual non-coding RNAs
(ncRNAs) were detected in Northern blots using probes y2aM, y3aM, y4aM and y5aM. These probes have
a minimum of five mismatches to their non-target ncRNAs, making cross-hybridisations impossible. The
numbers indicate transcript lengths in nucleotides. (C) Prediction of secondary structure of MED4 Yfr2
by MFOLD (Zuker, 2003; Axmann et al., 2005).

Surprisingly, although Yfr1 sequences originate from diverse cyanobacterial strains, they
share extensive structural and sequence conservation: Two terminal loops separated by
a 16 to 19 nt unpaired region that contains a CACAC motif (Fig. 3.18). Consistently,
the 3’ located stem-loop element is formed by at least five GC pairs, and is followed by a
short stretch of U residues, indicative of a Rho-independent transcription terminator (Fig.
3.18). Thus, a small but stable and structurally highly conserved RNA was identified
here, which might exist in every cyanobacterial genome, with only one known exception:
Prochlorococcus SS120.

3.5.5 Conservation of Yfr7/6Sa across the whole cyanobacterial lineage

The high sequence conservation of Yfr7 between Prochlorococcus Med4, MIT 9313, SS120
and Synechococcus WH 8102 was used to define oligonucleotides that hybridised to this
RNA species in four additional, unsequenced strains of Prochlorococcus and in three ad-
ditional Synechococcus strains (Fig. 3.19). The signal pattern is very distinct as all three
Prochlorococcus strains, which are high-light-adapted ecotypes (Med4, MIT 9312, MIT
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 71

A C A C A C Med4 PCC7942
A C A C A C C
U
C A
A
Yfr1 U
C A
C
Yfr1
C A C A
20 C
C U C
U C
U U 40
C 40 C U
G C
A C A C G
C C G U C C
U G A C C G
G G U G C G G A
A A C 20 U G G C
C U U U G C
U U G A U C A
C U A U 60 G
G U C G
C G C C
C C U
A G U G U U
G U G
A U A
G U G G
A A
A C
A A

B
yfr1_Syn8102 : -GUGCGGGCAAUG---CCCACACACUCCUCACACCCCCC--GGCCCGGCGCGC-------UCGGGCUUU--
yfr1_Pro9313 : -GUGAGGGCUACGG--CCCACACACUCCUCACACACACC--GGCCCGACGAGC-------UCGGGCUUU--
yfr1_ProMed4 : -AUGGGGGAAAC----CCCAUACUCCUCACACACCAAAU--CGCCCGAUUUA--------UCGGGCUUU--
yfr1_PCC6803 : GGCGGAGACGUAAGUUUCCGUUCACUCCUCACACCACACUCCGCCUGGAUGAU-----GUUCGGGCGGUUU
yfr1_T.elong : AGCGGAGACACACGUUUCCGUUCACUCCUCACACCACACCCCGCCUGGACAGUUAUUGGUUCGGGCGGCUU
yfr1_PCC7942 : AGCGGAGACAAAUGUUUCCGUUCACUCCUCACACCACACUUCGCCCGGGCAAC-----GCUCGGGCGUUUU

Figure 3.18: Identification of Yfr1 RNA in diverse cyanobacteria. (A) Two examples for the highly
conserved secondary structure of Yfr1 predicted for Prochlorococcus Med4 and Synechococcus PCC 7942.
(B) Parts of the guaB-trxA intergenic spacer were aligned and represent sequence similarities and a CACAC
motif, which is located inside the unpaired region of Yfr1 (red box) of Prochlorococcus Med4, MIT 9313
and Synechococcus WH 8102 as well as in Synechococcus PCC 7942, Thermosynechococcus elongatus,
Synechocystis PCC 6803.

9215), showed two signals in hybridisation, one at approximately 200 nt and one at ap-
proximately 300 nt, whereas RNA from the four low-light-adapted Prochlorococcus (SS120,
MIT 9313, NATL2A and MIT 9211) and four Synechococcus (WH 8102, WH 7803, WH
8020, RS9906) strains gave a single signal at approximately 175 to 185 nt (Fig. 3.19).
These strains represent a large genetic diversity within the marine cyanobacterial radia-
tion (Fuller et al., 2003), thus the presence of real orthologues of Yfr7 in additional and
even more distant cyanobacteria appeared likely.

Indeed, in the freshwater cyanobacteria Synechococcus PCC 6301 and Synechocystis PCC
6803, the 6Sa RNA has also been described directly downstream of purK (Watanabe
et al., 1997). There is some structural similarity between Yfr7 and the 6Sa, which leads
one to assume that these RNAs are real homologues of each other. In addition, a recent
publication provided comparative structural information suggesting that the ncRNA Yfr7
described here and 6Sa RNA from the latter cyanobacteria have structural elements in
common with the 6S RNA of γ-proteobacteria, in particular a large internal loop (the cen-
tral bubble in Fig. 3.20c), a typical closing stem and terminal loop (Barrick et al., 2005).
This possibly indicates that the here described 6Sa and Yfr7 RNAs are the orthologues of
γ-proteobacterial 6S RNA and may have a similar role throughout the whole eubacterial
radiation.

A detailed sequence comparison of already annotated 6Sa genes in very diverse cyanobac-
teria,

unicellular Synechococcus PCC 6301,

72 3. RESULTS

N2 MED SS1 MIT WH8 9312 9211 9215 7803 8020 9906

300 nt

200 nt

Figure 3.19: Identification of Yfr7 RNA in different marine cyanobacteria. In Northern blots, a signal
for the predicted Yfr7 was detected with RNA from all four strains and seven additional strains from the
marine cyanobacterial radiation: Prochlorococcus NATL2A (N2), Med4 (MED), SS120 (SS1), MIT 9313
(MIT), MIT 9312 (9312), MIT 9211 (9211), MIT 9215 (9215) and Synechococcus WH 8102 (WH8), WH
7803 (7803), WH 8020 (8020), RS9906 (9906). The high-light-adapted Prochlorococcus strains are labelled
in red, low-light-adapted strains in blue, and Synechococcus strains are colour-coded in green.

unicellular, non-nitrogen-fixing Synechocystis PCC 6803,

unicellular, thermophilic Thermosynechococcus elongatus,
unicellular, unusual Gloeobacter violaceus PCC 7421 and
filamentous, heterocyst-forming, nitrogen-fixing Anabaena PCC 7120,

together with the described marine Yfr7 RNAs did not result in an alignment exhibiting
well conserved regions, which were assumed for the 6Sa stem structure.

Further analysis of the genome location and alignments including the complement se-
quences (Fig. 3.21) indicated that the ssaA gene was annotated on the incorrect strand
for Thermosynechococcus elongatus, Anabaena PCC 7120 and Gloeobacter violaceus PCC
7421; which was supported by Barrick et al. (2005) very recently.
Using the complement sequences of annotated ssaA genes for these three strains yielded
a significantly improved alignment of 6Sa RNAs. This initial alignment was extended by
adding BLASTn hits of 6Sa in Synechococcus PCC 7942 and Nostoc punctiforme. Addi-
tionally, sequences of the four analysed marine strains as well as their BLASTn hits in
Prochlorococcus MIT 9312 and Synechococcus WH 7803 were added to the 6Sa alignment
(Fig. 4), which was the basis for further phylogenetic studies as the Parsimony tree shown
in Figure 3.21. The 6Sa tree clearly separates the chosen cyanobacteria into two clusters:
The highest relatedness was found inside the marine group of strains and the freshwa-
ter/terrestrial group. Thereby, the location of the ssaA gene appeared to be conserved
downstream of purK except for a few cases: Synechocystis PCC 6803, Gloeobacter vio-
laceus PCC 7421 and Thermosynechococcus elongatus, which interestingly appear close to
each other within the 6Sa tree (Fig. 3.21).

The highly conserved regions of the 6Sa/Yfr7 alignment were used to design oligonu-
cleotides that hybridised strand-specific to 6Sa RNA in all sequenced and unsequenced
strains available for culturing, whereas probes against the incorrect orientation failed.
Thus, the expression of 6Sa RNA could be verified for all strains investigated by hybridi-
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 73

U G 100
U G
A C
C G
U U
C C
A A
C G G C
G C G C
G C G C
Synech. Thermos. Synech. Microcys. Anabae. Nostoc Gloeobac. C G
100 C G
C G
PCC7942 elong. PCC6803 PCC7806 PCC7120 punctif. PCC7421 A
A
C G
A A A
U
A A A A
200 nt U A
80 G C
U A
G C
C G
C A U
A C U
G
A G U
U
G C U G A
G C U A A
C G C
A C
80 C G
G C
U A U A 120
C
C G C U G
U U C G
G U G U A
U C A
U
C G A
A
A U A
A
G U 120 A U
A U C G C
60 G C C G
G C A U
60 A
G C G A A A
A A
U G CU A A G C
U 140
B U C U U AG C
C C U U
Synechocystis PCC 6803 U
G A U
G
U G CAC C U G G
U
U U G G A C C
C U G A U
C U C A A
U UG C
5‘ proximal TIS Additional 5‘ ends C U U C
G
U
C U A U UU A
M TAP+ TAP- M TAP+ TAP- A 40
G C CU U
WH 8102 40 U A
PCC 7120 U
U A
C G 140 U A
U G U A
A U G C
G C A U 160
U A U
C G C
U A U A
C A
C C G
G C C G
U A G U
Anabaena PCC 7120 G C U A
A U C
A A
U G G A
5‘ proximal TIS Additional 5‘ end U
U
U C G
G A
M TAP+ TAP- M TAP+ TAP- U C G G C
U
20 C G 20U G
G C U G
G
CA U G C
G C C G
U G A U
G C G U
A
G C 160 C G
U G G C
U G U G
G C C G 180
U A A U
G C C G
G
U G U A
C G U A
G U
G C
U G
C G A U
A U G C
U U A
C A

Figure 3.20: Identification of 6Sa RNA in different cyanobacteria. (A) Hybridisation of Northern blots
with radiolabelled probes designed for a wide range of cyanobacteria identified signals for 6Sa in each strain
investigated (from left to right: Synechococcus PCC 7942, Thermosynechococcus elongatus, Synechocystis
PCC 6803, Microcystis PCC 7806, Anabaena PCC 7120, Nostoc punctiforme and Gloeobacter violaceus
PCC 7421). (B) RACE experiments detected the TIS of 6Sa RNA in Synechocystis PCC 6803 and An-
abaena PCC 7120 as well as additional processed 5’ ends. (C) Prediction of secondary structure of the
Synechococcus WH 8102 ncRNA Yfr7 and of the Anabaena PCC 7120 6Sa RNA.

sation to Northern blots, shown in Figure 3.20a. Thereby, the correct orientation of the
ssaA gene, as described above, was experimentally proven for Thermosynechococcus elon-
gatus, Anabaena PCC 7120 and Gloeobacter violaceus PCC 7421.

Additionally, RACE experiments verified the expression of 6Sa in Synechocystis PCC 6803
as annotated and in Anabaena PCC 7120 from the complement strand and detected the
TIS for 6Sa as well as additional processed 5’ ends (Fig. 3.20b). Their determined pro-
moter regions upstream of ssaA harbour well conserved elements for the -10 (TATAAT)
and -35 region (TTGac), perfectly matching to the consensus sequences of standard pro-
moters in E. coli as well as a few promoters (like rpl21 and ccmK ) found in Prochlorococcus
Med4 (Vogel et al., 2003a).
74 3. RESULTS

Parsimony tree Genome location

93 c PCC 7120 purK ssaA alr2743

58 [Link]. purK ssaA ?

47
PCC 6803 slr1288 ssaA sll1166

100 c Gloeobact. glr3379 ssaA glr3380

gyrB ssaA tll0646

c [Link]

30 PCC 7942 purK ssaA ORF4

100
PCC 6301 purK ssaA syc0127_d

80 MIT 9313 purK ssaA PMT0671

98 WH 7803 purK ssaA ORF1222

44
WH 8102 purK ssaA SYNW1306

SS120 purK ssaA Pro1007

Med4 purK ssaA PMM0684

99
MIT 9312 purK ssaA ?

Figure 3.21: 6Sa/Yfr7 Parsimony tree is shown based on the alignment of 13 sequences from (top
to bottom) Anabaena PCC 7120, Nostoc punctiforme, Synechocystis PCC 6803, Gloeobacter violaceus
PCC 7421, Thermosynechococcus elongatus, Synechococcus PCC 7942, Synechococcus PCC 6301, Prochlo-
rococcus MIT 9313, Synechococcus WH 7803, WH 8102, Prochlorococcus SS120, Med4 and MIT 9312.
Additionally, the arrangement of the 6Sa/Yfr7 encoding ssaA gene and adjacent genes is illustrated for
the 13 strains.

Day-time and growth-phase dependent expression of Yfr7 in Prochlorococcus

Med4
RACE experiments of Med4 RNA samples detected two 5’ ends for Yfr7 (Fig. 3.22a). At
the proximal TIS, a 220 nt long RNA is transcribed, and further upstream inside the purK
gene, a 5’ end resulting from processing yields a 332 nt RNA (Fig. 3.22b).

A comparable situation was recently described for Bacillus subtilis, where two differen-
tially expressed 6S RNAs exist, 6Sa and 6Sb (Barrick et al., 2005). Each transcript is
encoded by its own gene, whereby 6Sb was identified as the orthologous RNA to E. coli
6S, whereas 6Sa appeared functionally diverged (Barrick et al., 2005). For both Bacil-
lus subtilis transcripts a different growth-phase dependent expression was demonstrated
indicating an additional fine tuning of the transcriptional response to nutrient limitation
(Barrick et al., 2005).
On the other hand, experiments showed Yfr7 to be stable in Med4 and its expression
was not affected by stress conditions including nutrient limitation (Fig. 3.16). Therefore,
further investigations were needed to find out if expression of Yfr7 is affected by changing
growth conditions as demonstrated for 6S in E. coli or Bacillus subtilis.

Thus, the expression of both Yfr7 transcripts was tested during growth of the Med4 strain
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 75

Prochlorococcus MED4

A B
5‘ proximal TIS Processed 5‘ end purK ssaA
M TAP+ TAP- M TAP+ TAP-
TGGCTCATATAACGTTTTTA-100nt-TAATATttacatA
processed 5‘ end TIS ssaA
(long transcript) (short transcript)
332 nt 220 nt

C
growth curve within 10 days growth curve within 24 hours
0 1 2 3 4 6 8 10 days 11:00 13:00 15:00 17:00 19:00 21:00 23:00 3:00 5:00 7:00 9:00 11:00

0.02 0.04 0.06 0.09 0.12 0.14 0.17 0.19 OD day night day

SsaA 11 13 15 17 19 21 23 3 5 7 9 11

long

SsaA
11 13 15 17 19 21 23 3 5 7 9 11

short
5S
rRNA

Figure 3.22: Identification and transcript accumulation of Yfr7 in Prochlorococcus Med4. (A) 5’ RACE
detected the TIS of the yfr7 gene as well as the processed 5’ end further upstream. (B) Genome location of
yfr7 downstream of purK is shown. Transcription starting at the proximal TIS resulted in a 220 nt short
RNA. The long transcript variant processed at a site inside the purK gene has a length of 332 nt. (C)
The accumulation of both Yfr7 transcripts during growth within 10 days (left panel) and within 24 hours
(right panel) in a synchronised culture is shown. Northern blots were hybridised with radiolabelled probes
specific for Yfr7 and 5S RNA. Band intensities were quantitated, corrected for 5S loading controls, and
normalised to the maximum level of the respective Yfr7 transcript (%). The levels of both Yfr7 transcripts
are different for each timepoint and are time-shifted. The 332 nt transcript peaks at OD600nm 0.14; the
220 nt RNA accumulates at the last time-points, day 8 and 10 (left panel). Within a day (right panel)
both transcripts are changing their amounts rhythmically with peaks in the morning for the 332 nt RNA
and at noon for the shorter 220 nt variant.

(Fig. 3.22c). Interestingly, the growth curve within ten days from OD600nm 0.002 to 0.19
revealed varying expression levels with different maxima for the long and the short Yfr7
transcript: The total levels of the processed 332 nt RNA are peaking at OD600nm 0.14,
probably the entry into the stationary phase under these growth conditions; at the latest
time points (day 8 and 10) the 220 nt RNA accumulated maximum.
Surprisingly, the analysis of RNA samples of a synchronised Med4 culture collected over 24
hours showed a rhythmic accumulation of Yfr7 with different minima and maxima for both
transcript variants (Fig. 3.22c): The 332 nt RNA decreases at the day-night-transition
and increases during the night obtaining highest levels in the morning; the shorter 220
nt RNA accumulates to a maximum at noon and increases to a minimum around midnight.

The differential expression of both Yfr7 transcripts during growth within ten days indi-
cates a well regulated transcriptional response to likely limiting conditions by peaking
during entry into stationary phase at day six (long transcript) or within stationary phase
at day eight (short transcript), similar to the situation in Bacillus subtilis.
Intriguingly, an effect of the daytime on Yfr7 accumulation was also identified: Again both
76 3. RESULTS

Yfr7 variants were peaking differentially within a rhythmic expression pattern. Thus, the
cyanobacterial ortholog of 6S RNA, Yfr7, could be involved in the regulation of cellular
processes depending on light/darkness and therefore on the circadian rhythm.

Knock-out constructs and Synechococcus WH 8102 mutants

Taking the new findings about Yfr7 and 6Sa into account, it appeared very tempting to
get insights into the precise function of the cyanobacterial 6S ortholog and to demonstrate
that it is also functionally related to the γ-proteobacterial 6S RNA. Thus, the yfr7 gene
of WH 8102 was targeted for a single-cross-over insertion. The transformation was suc-
cessful as the obtained WH 8102 transconjugants were growing on the chosen antibiotic,
kanamycin. Thus, the transformants were plated to get single colonies of one mutant
variant.

KanFW suiInREV site for yfr7

suiInFW KanREV knock-out
M wt mu wt M mu
A B possible insertion sites for pMUT100

5000 350 bp
cloned PCR fragment

500
purK ncRNA yfr7

insertion at insertion at colonyFW suiInFW colonyREV

site B in site A in suiInREV
mutants mutants

colonyFW colonyFW suiInFW

colonyREV
M wt mu
suiInREV
M wt mu
colonyREV
M wt mu
B
5000 Kan pMUT100
KanFW KanREV
Northern
500 blot
wt mu A

no wt in wt-site A in wt-site B in Yfr7 in wt Kan pMUT100

mutants mutants mutants and mu
KanFW KanREV

Figure 3.23: Knock-out construct for Yfr7 in Synechococcus WH 8102 and analysis of mutants. The yfr7
gene of WH 8102 was targeted for a single-cross-over insertion: A 350 bp DNA fragment was cloned into
pMUT100 knock-out vector; mutants were generated by bi-parental mating and insertion of the knock-out
construct into the WH 8102 genome at site A (yfr7 destruction) or site B (no yfr7 destruction). Mutant
(mu) and wild-type (wt) cultures were analysed by PCR and Northern blot hybridisation, which revealed
a mixed DNA population without a successful sorting out of a yfr7 knock-out.

Further on, the insertion was clearly verified by PCR, but it was not possible to isolate
Yfr7-deficient cells up to now: In any case, although an insertion of the foreign con-
struct (including the kanamycin-resistance cassette) into the targeted genomic location
was demonstrated (Fig. 3.23), it was not possible to isolate a pure knock-out strain
containing only one type of insertion. Additionally, hybridisation to Yfr7 on Northern
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 77

blots loaded with total RNA of wild type and mutants revealed signals virtually identical
in abundance and molecular weight (Fig. 3.23). Collectively, these data suggest that a
mixed mutant population is still existing or that a single WH 8102 cell harbours more
than one copy of DNA, resulting in the expression of sufficient amounts of Yfr7. Similar
results were observed for knock-out experiments for Yfr1 in WH 8102 (data not shown).
Thus, the gene disruption technique based on a single cross-over event might not be the
most appropriate one for short target genes. Since a minimum of 300 bp is needed for a
successful crossing over, the disrupting plasmid can integrate within or outside the target
gene. Therefore, screening for successful gene disruption (by PCR) is very time consum-
ing. However, since now Yfr1 and Yfr7/6Sa are identified in freshwater cyanobacteria like
Synechocystis PCC 6803 for which molecular tools are more developed than for marine
cyanobacteria, the function of these highly interesting RNAs can now be studied in fresh-
water strains.
78 3. RESULTS

3.5.6 Excursus to Synechocystis PCC 6803: A cis-encoded antisense

RNA for isiA
Studying the gene expression of isiA, encoding the chlorophyll binding protein IsiA, in
Synechocystis by the group of Dr. Annegret Wilde, a small RNA was detected that is
transcribed from this region. More detailed RNA analyses in the context of this work
showed that the accumulation of this RNA is inversely regulated to the mRNA level of
the stress-inducible gene isiA (Fig. 3.24). Therefore, this newly identified RNA, was
named IsrR for Iron Stress Repressed RNA. Strand-specific oligonucleotide probes as well
as RACE experiments clearly identified this transcript as an antisense RNA transcribed
from the complementary strand of isiA and from an own promoter (Fig. 3.25a,b).

Iron depletion High-light stress

isiAB
isiA

isrR

5‘ UTR
0 6 12 17 24 40 48h 0 2 17 24 41h

Figure 3.24: Accumulation of different transcripts from the isiAB region under iron limitation (left
panel). Cells were cultivated under iron replete conditions to the logarithmic phase of growth (0 h),
washed in iron-free media and grown further in iron deplete conditions for 6 to 48 h. The isiAB dicistronic
precursor transcript, the isiA mRNA and a transcript in the 5‘ UTR region of isiA are induced under
iron deplete conditions. The antisense RNA IsrR is inversely accumulated: IsrR is present under iron
replete conditions and disappears with increasing time of iron limitation. Induction of the mRNA for isiA
and isiAB and of the 5’ UTR transcript during high-light conditions (right panel). Cells were cultivated
under normal light conditions (0 h) and then transferred to high light for 2 to 41 h. The isiAB, isiA
and 5‘ UTR are induced with increasing time of incubation under high light whereas the amount of IsrR
decreases. The short transcripts were detected on Northern blots prepared from 10 % polyacrylamide gels
using strand-specific oligonucleotides for IsrR and a 5‘ UTR DNA fragment as radiolabelled probes. The
isiA and isiAB transcripts were separated on 1.3 % formaldehyde-agarose gels and hybridised with an
isiA-specific DNA fragment.

The length of IsrR and by that the precise nucleotide sequence was determined by RACE
experiments targeting the 5’ and 3’ ends. The position in the genome extends from 1518034
to 1518210 on the complementary strand. Thus, the 177 nt long IsrR is located in the
middle of the annotated coding region of isiA (Fig. 3.25a,d). The determined IsrR RNA
sequence can be predicted to fold into two extended stem regions each finishing with a
terminal loop (Fig. 3.25c).
In addition, another short transcript of about 160 nt is originating from the isiAB region,
representing a major part of the 5’ untranslated leader region (5’ UTR) of the isiA gene.
This 5’ UTR transcript was observed before (Vinnemeier et al., 1998), but without assign-
ing specific function to it.
The newly identified antisense RNA IsrR is present in high amounts under iron replete
conditions but absent under iron limitation. In contrast, isiA mRNA was not detectable
under iron replete conditions but with an increasing concentration after 17 h in iron de-
plete media (3.24). As the isiA gene is known to be responsive to high-light-induced stress
as well (Havaux et al., 2005), the response of cells transfered from normal to high light was
3.5. SMALL RNAS IN MARINE AND OTHER CYANOBACTERIA 79

tested for 2 to 41 hours. Induced by high light, the amount of isiAB dicistronic precursor
transcript, isiA mRNA and 5‘ UTR transcript are gradually increasing, whereas the IsrR
RNA signal is fading away rapidly (Fig. 3.24).

U
A C
A
A B 5‘ proximal TIS
G
A M TAP+ TAP-
C
U
U TIS isiA
C
G
C G C 80 isiA isiB
G C
U
G C
G C isrR
U
G C
G
60 C TIS isrR
G A U
U A
A
U C
A
U U
G G
C G C
C U
A U
U
C G
C C
G
C 100
D >isrR complement(1518034..1518210) 177nt

G
G
G
C
5‘e - AUAACCAUAAAUAACCAAAGGAUCUAAUGUUGGCUGACUGACU
U
C U A AGGCGCACCGUUUGGGUGGUGGCGUCGUACAAUCCCCCCCAAUACAU
G
A
A
G
C C GGCUUUGCCUGCCAACAGUAGAGCUCCGGCCCCCAGGAACAGGAG U
U G C
40
U G
G A GAU GACCC C
AAAAUUAU CCUAGUUGCUGGGGGUUUUC -3‘ CC
C A U A
U C
G A
G
U
U
A
A A
U G
C AG
U C
U
A 20 C
G 120
A A A CG
C G G
C C
A U C
G C
A U G C
G C
U U G A
G G G
A U U A
C A
A C G U U C
A
A G G
C AU GA
C G
U C 160 C A
C U 140
A U G
A A G A
C G U U
C C
A A U A U
A
C
A
C
A A

Figure 3.25: Characterisation of an antisense RNA in Synechocystis PCC 6803. (A) Location of the
isiA gene within the Synechocystis genome. The long distance between transcription start (red arrow) and
translation start site (green arrow) indicates the location of a previously described 5’ UTR transcript in
sense orientation (Vinnemeier et al., 1998)). Directly downstream of isiA the isiB gene is located encoding
flavodoxin. (B) The unprocessed 5’ end (lane TAP+) of IsrR detected by 5’ RACE indicating the TIS for
isrR. (C) RNA secondary structure prediction for IsrR by MFOLD. The red arrow points to the 5’ end
of the RNA detected in RACE experiments, the green arrows label the found 3’ ends, whereby the larger
green arrow represents the most frequently observed 3’ end. (D) Verified sequence of IsrR: The red letter
indicates the 5’ end (TIS), green letters are different 3’ ends.

Thus, at least four different transcripts originate from the isiAB genomic locus: The unpro-
cessed isiAB precursor transcript, the isiA monocistronic transcript, IsrR with maximum
accumulation under iron replete and low-light conditions, when isiA mRNA is missing,
and the 5’ UTR transcript which appears to be co-accumulated with isiA mRNA. For
each newly identified non-coding RNA it often appears to be challenging on which targets
it may act. In this example, the reciprocal relation between the amounts IsrR and isiA
80 3. RESULTS

mRNA suggests that they could react to each other via a titration mechanism.
Chapter 4

Discussion

4.1 Promoter architecture and transcriptional regulation

A unique comparative dataset is given by the highly related but even so diverse adapted
marine cyanobacteria, Prochlorococcus Med4, SS120, MIT 9313 and Synechococcus WH
8102. The fact, that these four genomes were already sequenced at the beginning of this
thesis, offered the great opportunity to study their transcriptional regulation based on
sequence comparisons. Firstly, combined experimental and computational promoter stud-
ies revealed similar features as known from E. coli: The marine cyanobacterial promoter
consists of TIS and -10 element, and in the minority of cases also the -35 region is con-
served (Vogel et al., 2003a). Additionally, a -60 element occurs for several analysed genes,
probably representing another promoter element, which could also be linked to the fre-
quent head to head orientation of genes resulting in an overlapping and often co-regulated
(bidirectional) promoter region (Korbel et al., 2004). On the other hand, the sequence
composition of spacer regions between conserved promoter elements appears to be more or
less random as well as not every comparable genomic region between the four strains pos-
sesses conservations in their non-coding parts. Thus, the set of four marine cyanobacterial
genomes might offer a reliable genomic distance highlighting conserved sequence elements
as biological relevant motifs among orthologous promoter regions within a context of oth-
erwise non-conserved sequence.

Beside putative transcription factor (TF) binding sites for the CRP-like regulator NtcA
(Palinska et al., 2000), known to mediate nitrogen control in cyanobacteria, not much was
known about the regulatory network for marine cyanobacteria, and genome-wide studies
did not exist about co-regulated genes (regulons) controlled by trans-acting TFs and their
cis encoded DNA binding sites. Thereby, a search for additional promoter elements was
tempting to get an overview of the transcriptional regulatory potential of these very spe-
cialised organisms. Nearly half of all genes in the genomes of marine cyanobacteria are of
unknown function. Photosynthesis and an autotroph lifestyle can be assumed to require
a set of regulatory circuits that is significantly different from what is known from model
organisms such as the intensively studied enterobacteria.
82 4. DISCUSSION

4.1.1 The conserved regulatory potential of four marine cyanobacteria

A phylogenetic footprinting approach for the four sequenced marine genomes was imple-
mented to reveal functionally relevant similarities between promoter regions. Thus, new
information was obtained about the fundamentals of transcriptional regulation for the
marine cyanobacterial strains each adapted to its particular habitat.

In a first step, a set of orthologous coding regions was calculated resulting in 1293 clusters,
which represents a number similar to other BLASTp comparisons of the four genomes, Pro-
chlorococcus Med4, SS120, MIT 9313 and Synechococcus WH 8102 (Hess, 2004; Dufresne
et al., 2005). Keeping in mind, that the total number of coding regions in the smallest
of the four genomes, Med4, is 1716, about three-quarter of all Med4 genes belong to this
conserved core, whereas only half of the Synechococcus WH 8102 genes are represented
by these orthologous clusters. Within this set of clusters, only five sigma and 35 putative
regulatory factors were found including about 18 potential transcriptional factors, which
constitute the predicted core set of regulatory proteins conserved between these four ma-
rine cyanobacteria.
This set of regulatory proteins as identified here is mainly limited by the Med4 genome,
which is the most compact of any free-living photoautotroph, suggesting that it may
possess only that part of the regulatory network of a cyanobacterial cell which is most
fundamental and widespread. Therefore, the presented data were expected to be relevant
not only for comparison to more closely related marine strains but also to many other
cyanobacteria.

About 20 motifs were detected with more than one entry per genome (Appendix, Fig. 2),
including sites similar to previously described consensus sequences of the regulators ArsR,
LexA and NtcA, found in freshwater cyanobacteria. Focussing on palindromic motifs and
following further genome-wide searches of additional entries for the putative binding sites
and assignment of their possible regulons, five best motifs were suggested:

(1) An ArsR-like consensus sequence is located within the spacer region of arsR and gap.
However, the arsBHC operon that is involved in arsenic sensing and resistance in Syne-
chocystis PCC 6803 (Lopez-Maury et al., 2003) was not found in the four marine strains.
Thus, the ArsR-like factor here may participate in the regulation of other genes and oper-
ons. Indeed, a system identical to Synechocystis becomes more unlikely, as the predicted
ArsR-like sites appeared upstream of genes like pstS, phoB (two-component response reg-
ulator, phosphate) and phoR, thought to be regulated by the amount of phosphate in the
cell. As both regulators, PhoB and ArsR, are encoded in each of the four marine genomes
studied here, a crosstalk between their regulons might be assumed with the exception of
SS120, where phoB is missing. On the other hand, a predicted Pho box by Su et al. (2003)
was not identified in upstream regions together with this marine ArsR-like motif.
Arsenate is chemically analog to phosphate, and for Anabaena variabilis an inhibition of
the phosphate transport by intracellular arsenate was already suggested (Thiel, 1988),
although the underlying mechanism is still poorly understood. Thus, the repression of
certain P-dependent genes in the presence of arsenate mediated by ArsR binding in their
promoter regions might represent a possibility.
Interestingly, a minimal cellular P content of Prochlorococcus and Synechococcus cells
4.1. PROMOTER ARCHITECTURE AND TRANSCRIPTIONAL REGULATION 83

under P-starved conditions is one of their most striking features, indicating very low
phosphate requirements (Bertilsson et al., 2003). Even in a non-limiting environment,
Prochlorococcus was found to retain an intracellular P limitation (Bertilsson et al., 2003).
Therefore, one could image, that the regulation of the phosphate uptake and content of
these cells has to be very specific. Future experimental efforts may gain further insights
into this highly complex regulatory network.

(2) Another palindromic sequence was observed within the upstream region of ham1 and
csoS1-rbcLS operon. Additional entries occurred for petF and psbA, encoding proteins
involved in photosynthesis. Thus, this predicted cis element might be involved in the
regulation of carbon fixation and photosynthetic light reaction. The mRNA level of psbA
(encoding the reaction centre II subunit D1) was shown to be tightly correlated with light
irradiance, with a minimum at night and a maximum at noon (Garczarek et al., 2001) and
also rbcL expression of the indigenous phytoplankton was greatest in the day and least at
night (Pichard and Paul, 1991). A more detailed study, using two cyclostats of Prochloro-
coccus culture, revealed a transcript accumulation of rbcL increasing from the beginning
of the dark period to a maximum in the early morning and afterwards decreasing sharply
down to nearly zero at the end of the light period (Bruyant et al., 2005). Thus, at least the
maximum transcript accumulation of rbcL and psbA appeared to be shifted about several
hours, which does not necessarily exclude a co-regulation.
Moreover, a CO2 response element and the corresponding trans-acting factor of the pro-
moter for rbcLS in Synechococcus sp. PCC7002 were described earlier, which possibly
regulate the rbcLS transcription in response to CO2 levels (Onizuka et al., 2002) - a mech-
anism which is quite similar to the regulation of the rbcS promoter in higher plants and
which might be assumed for the marine strains, studied here, as well. The RuBisCO
operon transcriptional regulator, which is found in all cyanobacteria so far (similar to
HTH-, LysR-type Ycf30 encoded on plastids of eukaryotic algae), might mediate this CO2
response or an additional regulatory circuit.

(3) The putative LexA site found for marine strains is highly similar to the previously de-
scribed consensus sequences of gram-positive and freshwater cyanobacteria (Mazon et al.,
2004b). Furthermore, the LexA regulon predicted here contains several genes known to be
active in the SOS response system of bacteria such as umuC and umuD and especially recA
and lexA, encoding the positive and negative regulator respectively, which might indicate
a mechanism surprisingly similar to the SOS system best known from E. coli (Walker,
1984).
RACE experiments showed that the predicted binding sites of LexA are located exactly
at the transcription initiation site of lexA, umuD and PMM1427 in Med4, indicating that
LexA itself might be negatively autoregulated and could act as the repressor for several
other genes.
Although today, there are different functions for LexA discussed in literature (Mazon et al.,
2004b; Domain et al., 2004; Gutekunst et al., 2005) and studies about Synechocystis (Do-
main et al., 2004; Gutekunst et al., 2005) raised the question if all cyanobacteria possess
an E. coli-type SOS regulon, the data obtained during this study for marine cyanobacteria
give evidence for a DNA repair system similar to the E. coli model.
84 4. DISCUSSION

(4) NtcA is a major regulator for nitrogen control in cyanobacterial cells (Herrero et al.,
2001). Those parts of the genome, which are repressed or activated by its presence, con-
stitute the N-regulon. Here, only a small but high-scoring subset of this putative regulon
was defined, including genes for major enzymes of nitrogen-metabolising pathways such as
spt,agt (aminotransferase) and glnA (glutamine synthetase) as well as important nitrogen
dependent transport systems like urtABCDE (urea transporter). The consensus sequence,
identified here, harbours additional features besides the often used TGT-N10 -ACA motif:
The flanking A/T-rich sequences and a conserved TG (or CA) dimer (Vazquez-Bermudez
et al., 2002). This more complex motif for marine cyanobacteria corresponds only partly
to the profile GTA-N8 -TAC suggested recently (Su et al., 2005). Thereby, only a subset
of genes was predicted here compared to the larger NtcA regulons described by Su et al.
(2005), in which genes involved in photosynthesis were additionally identified. Nitrogen
is an important bio-element and a more global regulation mediated by NtcA might be
assumed, that would also affect photosynthesis genes and others. On the other hand, it
remains to be proven experimentally, if NtcA is directly involved in a general response
of such a large set of genes or if the global influence observed during nitrogen-starvation
includes secondary effects as well.

(5) The last of the five top motifs consists of a 16 bp palindrome. The prediction of its
likely regulon revealed several genes with an unknown function indicating a completely
new binding site for a new regulon in marine cyanobacteria.

A core set of regulons for marine unicellular cyanobacteria is suggested here for the first
time. The comparison of four sequenced genomes gives new insights into the minimum net-
work of transcriptional regulation for strains within the marine ecosystem, but it does also
allow drawing conclusions for cyanobacteria in general: Two known regulators, NtcA and
LexA, appeared to be conserved over a wider evolutionary distance from freshwater to the
group of marine cyanobacteria - from the most primitive unicellular to the filamentously
growing complex species. The identification of NtcA and LexA in marine cyanobacteria
illustrates how the data set might be utilised for an identification of promoters and regu-
latory sequences in other cyanobacterial species.
In contrast, other factors like the one recognising the ArsR-like binding site, might have
evolved differentially and probably possess new functions and regulons adapted to the
marine environment. Further experiments and comparisons with high throughput gene
expression data will improve this initial regulatory network.
Moreover, one has to remark that the computational predictions of DNA binding sites
made here together with the experimentally tested examples can not serve as the entire
proof of their biological function. For this purpose, additional binding studies, e.g. DNA
affinity precipitation, DNase I protection or mobility shift assays, as well as detailed mu-
tational analyses of the appropriate promoter regions might follow in the next future.

Of course, this small set of transcription factors can not represent the complete one for
each strain, because it is missing several factors and their binding sites, which are not con-
served between all four strains. Additionally, the deeper focus on palindromic and highly
conserved motifs excluded all regulators, which recognise patterns of different geometries
and lower base-pair conservations like Fur (ferric uptake regulator protein) binding sites
4.1. PROMOTER ARCHITECTURE AND TRANSCRIPTIONAL REGULATION 85

and others.
Thus, combinations of varying genome-sets as input for the phylogenetic footprinting anal-
ysis might help to generate more focussed information on the specific regulation network of
a single strain or a set of few strains adapted to a particular niche in the marine ecosystem
- a promising possibility for the very next future as there become more and more marine
cyanobacterial genomes sequenced.

4.1.2 Circadian rhythm: clock proteins and other periodosome compo-

nents
Cyanobacteria are the only prokaryotes which possess circadian oscillations. These 24-
hour rhythms are generated basically by three proteins in the cyanobacterial cell: KaiA,
KaiB and KaiC (Nakajima et al., 2005), culminating in the formation of a high-molecular-
weight complex during the night - the periodosome (Bell-Pedersen et al., 2005). Among
the known components of a cyanobacterial clock, two genes appear to be deleted from
marine strains. The circadian input kinase, CikA, is missing in all strains, so that only
one input pathway might remain, mediated by a recently identified protein, LpdA, sensing
the redox state of the cell (Ivleva et al., 2005).
Additionally, the gene encoding the core clock protein KaiA can not be annotated for Pro-
chlorococcus strains sequenced so far, though it is an indispensable part of the kai operon
in related marine Synechococcus strains. Only for some genomes of low-light-adapted Pro-
chlorococcus strains, a truncated kaiA coding region can be detected within the kaiB-rpl21
spacer region, indicating an evolutionary intermediate between the three- and the two-gene
form of the kai operon. That is in itself intriguing, because it directly contradicts specu-
lations on the Prochlorococcus kai gene operon (and clock) as the primitive evolutionary
ancestor to the more complex situation found in Synechococcus elongatus (Dvornyk et al.,
2003).
It has been shown recently, that the genomes of Prochlorococcus cells underwent an inten-
sive evolutionary reduction process (Dufresne et al., 2005). Here, the genome reduction
process can even be demonstrated by this example: the step by step deletion of the kaiA
gene. Thus, a reduced clock and a simplified circadian mechanism might be unsurpris-
ing for an even so minimised organism like Prochlorococcus. The question remains if in
Prochlorococcus cells an autonomous pacemaker can exist, ticking without the basic KaiA
protein component? For the model organism Synechococcus PCC 7942 all three kai genes
are essential for circadian rhythmicity, and an inactivation of any of them abolishes it
(Ishiura et al., 1998). On the other hand, studying mutations of sigma factors revealed
that PkaiA and PkaiB are on different regulatory circuits and that the kaiA expression
can be dramatically altered without changing the fundamental timing mechanism (Nair
et al., 2002).
Based on the data presented in this thesis, the transcriptional regulation of the kai genes
in marine strains differs from the model: A kaiBC promoter could not be found for marine
Synechococcus WH 8102. Furthermore, in Prochlorococcus the expression of kaiB seems
to be coupled to that of the ribosomal gene rpl21. One might assume that the reduction of
the clock components and the different timing of kai gene expression as compared to Syne-
chococcus PCC 7942 might result in a simplified circadian clock for marine cyanobacteria.
86 4. DISCUSSION

Further studies of the marine circadian system including its essential protein components
are needed to understand all the underlying regulatory mechanisms.

4.1.3 Transcriptional signals of cyanophage P-SSP7 infecting Prochlo-

rococcus Med4
The cyanophage P-SSP7 possesses several T7-like genes including the essential RNA poly-
merase (Sullivan et al., 2005). Additionally, a gene expression order from left to right
(Lindell, personal communication) implicates a transcriptional strategy as described for
the T7 group of phages. Intriguingly, in a computational study not a single promoter
site similar to the well-known and -conserved phage promoters was detected (Chen and
Schneider, 2005), which might indicate differences of the transcriptional mechanism for
the P-SSP7 genome. An analysis of phage mRNAs on the basis of new experimental data
was thought to discover sites of transcription initiation and therefore recognise promoters
to solve the question for the unknown transcription signals of RNA polymerase in P-SSP7.

Predictions of promoter and termination sites as well as the experimental analysis of the
mRNA 5’ ends revealed transcriptional features of P-SSP7, which are partly similar but
also different to the T7 model. Transcription of P-SSP7 from left to right demands tran-
scription of the first phage genes (including RNAP) from bacterial promoter sites located
near the left end of P-SSP7 DNA. A Med4-like promoter region was predicted upstream of
gene 1, although it could not be verified experimentally. The initial transcription by the
host polymerase might be stopped at the predicted termination sites between gp12 and
gene 3 or upstream of gp1, the gene for the RNA polymerase, which is not in agreement
with T7, where the expression of the phage RNA polymerase is essential for the following
steps of infection.
On the other hand, a bacterial-like promoter was found upstream of the gene for the phage
RNAP, which might indicate the possibility that this important phage gene is transcribed
independently of the other P-SSP7 genes by the host RNA polymerase. Consequently, the
phage RNAP protein would possibly be present for the transcription of P-SSP7 DNA very
early during Med4 infection. Nevertheless, the time order of gene transcription in P-SSP7
remains unclear.
Another similarity to a T7-like transcription was detected by the prediction of a termina-
tion site between the capsid protein (gp10) and the tail tubular protein gene (gp11). As a
promoter for the latter gene is absent, gp11 has to be transcribed by read-through of the
terminator, which is part of the transcriptional strategy in T7 and ensures production of
large amounts of the major capsid protein, gp10 (Dunn and Studier, 1983).

RNA samples of P-SSP7 infected Prochlorococcus Med4 cells were used in RACE exper-
iments in order to analyse the transcriptional units expressed during the infection cycle.
Various 5’ ends were identified for cyanophage mRNAs. Among them, one group with
signals of maximum strength and stability for the genes 1, 3 and 29 (gp10, capsid protein)
revealed a motif conserved between these three sites, which was shown to possess more
similarity to an RNase cleavage site than to a promoter site for the phage polymerase.
Additionally, inverted repeats are located next to this motif, which might serve as recog-
4.2. NEW FAMILIES OF SMALL RNAS IN CYANOBACTERIA 87

nition sites for RNase III - a processing mechanism of phage mRNAs described for the T7
infection as well (Dunn and Studier, 1983).
The fact, that the motif was also conserved in two environmental sequence samples (Venter
et al., 2004) upstream of orthologous gp10 genes, supports the idea of an essential element
for the P-SSP7 phage transcription.
Additional entries of the motif appeared within the host genome next to putative 3’ cleav-
age sites of tRNA and tmRNA. Thus, the identified motif might possess a biological
function for both, Prochlorococcus Med4 and P-SSP7, in representing a potential cleavage
site for their transcription products mediated by an RNase.

The phage encoded genes, hli and psbA, did not reveal a detectable promoter signal, sup-
porting the suggestion of co-transcription with the essential phage capsid genes (Lindell
et al., 2005).
Two host genes of Prochlorococcus Med4, PMM0684 and PMM0819, induced during infec-
tion (Lindell, unpublished) possess perfect bacterial promoters, verified by RACE signals,
which might indicate a response of the host to its infecting phage.

It still remains questionable if the well studied transcriptional mechanism for coliphage
T7 could be assigned to the marine phage P-SSP7 one to one. The observation from
T7, that promoter and cleavage sites often occur next to each other, might had compli-
cated the experimental detection of in vivo transcription initiation sites. Thus, further
experiments like an in vitro transcription system for P-SSP7 RNA polymerase would be
helpful to analyse its promoter recognition sites. Additionally, future studies of RNA pro-
cessing mechanisms in the host, Prochlorococcus Med4, as well as for its infecting phage,
P-SSP7, would provide important information for both members of the marine community.

4.2 New families of small RNAs in cyanobacteria

For several eubacteria, especially for E. coli as well as for Vibrio cholerae, non-coding
RNAs are described as essential regulatory factors mediating rapid responses to environ-
mental changes. The wide variety of underlying regulatory mechanisms are still poorly
understood. Compared to the immense and growing number of predicted small RNAs,
only a few well investigated functional interactions are described for RNA antisense bind-
ing to mRNAs, manipulation of proteins or direct sensing of metabolites (riboswitches).
For free-living marine phototrophs such as the cyanobacterial strains investigated in this
study, regulatory circuits involving small RNAs were expected as well. However, except
for RNase P RNA, scRNA and tmRNA, three RNAs easily to identify, nearly nothing
was explored about ncRNA genes in marine as well as other cyanobacteria. Based on a
biochemical protocol, a single ncRNA, 6Sa, was identified previously in unicellular fresh-
water cyanobacteria, Synechococcus PCC 6301 and Synechocystis PCC 6803 (Watanabe
et al., 1997). In addition, mapping of transcriptional units within the gas vesicle operon
of Calothrix, a filamentous and heterocysts forming algae, identified a single antisense
transcript (Csiszar et al., 1987). Recently, a cis antisense RNA was found in the nitrogen-
fixing cyanobacterium Anabaena sp. PCC 7120 interfering with furA transcript translation
(Hernandez et al., 2006).
88 4. DISCUSSION

It was already shown for promoter and phylogenetic footprinting analyses that the genomes
of Prochlorococcus SS120, MIT 9313, Med4 and Synechococcus WH 8102 provide a unique
dataset for comparative studies. These small, even smallest cyanobacterial genomes differ
by several hundred genes from each other, although most of the operons and gene clusters
present in more than a single genome are co-linear (Dufresne et al., 2003; Palenik et al.,
2003; Rocap et al., 2003). In a first genome-wide and systematical screen for ncRNAs in
marine cyanobacteria 17 ncRNAs were detected in Med4, SS120, MIT 9313 and WH8102.
With a focus on Prochlorococcus Med4 the presence of new non-coding RNAs in the group
of marine unicellular cyanobacteria was analysed experimentally.

4.2.1 Computational screening in marine genomes

Genes encoding functional RNAs are notoriously difficult to predict, because of their
low conservation in primary sequence. Therefore, in cooperation with Philip Kensche a
comparative computational approach based on sequence and structure conservation was
implemented for the cyanobacterial genome set (Kensche, 2004), including an algorithm
that was recently introduced for the identification of eukaryotic ncRNAs (Washietl and
Hofacker, 2004). In view of the daily growing number of microbial genome sequences,
comparative approaches will become more and more possible and required as well.
Only highest scoring candidates of these predictions were analysed further, detecting sev-
eral previously unknown ncRNAs as well as other elements that function at the RNA level.
By comparison to literature and experimental data, the predicted list of high-scoring can-
didates contained a very low rate of true negatives. This indicates that the employed
method is very efficient in finding microbial ncRNAs and other RNA elements. Therefore,
it can be suggested to exploit this algorithm to further related genome sets predicting
ncRNA for cyanobacteria or other microbes as well.

One has to keep in mind, that the 17 ncRNAs detected here in Med4, SS120, MIT 9313
and WH8102 are only a part of the total ncRNA population present in these species.
That could be caused by the restriction to intergenic regions. Thus, ncRNAs were missed
that reside within annotated regions including the whole class of antisense RNAs, encoded
complementary to their target. Additionally, incorrect annotations reduced the number of
predicted sequences, like in the case of yfr7, which is located in a region complementary
to a reading frame in two of the genomes investigated here.

The presence of new non-coding RNAs in the group of marine unicellular cyanobacteria
focussed on Prochlorococcus Med4 may present only the tip of the iceberg. Several more
ncRNA candidate genes were predicted in the two relatively larger genomes of WH 8102
and MIT 9313 but still await experimental testing. An overview of the candidate regions
identified by this screen is given in the Appendix, Table 3 and a summary of the experi-
mentally confirmed new ncRNAs is presented in Table 3.5. In addition to small ncRNAs
genes, the computational results indicated the presence of conserved secondary structure
elements belonging to untranslated upstream regions of several ribosomal protein operons.
Thus, autogenous control mechanisms for the expression of ribosomal genes similar to en-
terobacteria (Lindahl and Zengel, 1986; Zengel and Lindahl, 1994) may exist in marine
4.2. NEW FAMILIES OF SMALL RNAS IN CYANOBACTERIA 89

cyanobacteria as well. The four genomes analysed here represent only a tiny part of the
unmeasured and diverse oceanic gene pool, and the existence of numerous ncRNA genes
in newly sequenced genomes as well as in environmental samples (Venter et al., 2004) can
be predicted.

4.2.2 Non-coding RNAs of Prochlorococcus Med4 and their orthologs

The analysis of four strains of Prochlorococcus and Synechococcus revealed an interesting
set of structural RNA elements. Especially the ncRNAs found in Med4 and SS120 may be
of considerable importance, as these strains underwent a strong genome reduction includ-
ing deletion of the hfq gene. The larger genomes of WH 8102 and MIT 9313 contain an
hfq gene, whose product has been shown to be essential for the activity of small regulatory
RNAs in enterobacteria (Valentin-Hansen et al., 2004). It is likely that, together with hfq,
several ncRNA genes have been deleted during the evolution of the Prochlorococcus group
towards a minimal genome. Thus, the ncRNAs still remaining in an organism like Med4
must have been subject to strong positive selection and may act independently of Hfq.
On the other hand, Hfq binding is not essential for the functionality of every ncRNA, as
only 30 % of investigated ncRNAs in E. coli have been shown to be bound by Hfq (Zhang
et al., 2003).

Evidence of function of the ncRNAs described here may arise from the comparison of ex-
pression patterns, structures as well as genomic location, and from the presence or absence
of a given ncRNA gene in the different strains:
(1) The Yfr1 RNA might be dispensable for growth at greater depths, because its gene is
clearly absent from the ultra low-light-adapted SS120 but present at the identical genome
location (upstream trxA) in the other three marine strains. Surprisingly, it also appeared to
be conserved in freshwater cyanobacteria, Synechococcus PCC 7942, Thermosynechococcus
elongatus and Synechocystis PCC 6803, harbouring a highly similar secondary structure
with two terminal loops separated by an unpaired region that contains a CA dinucleotide
repeat. These highly conserved features might represent essential functional elements of
Yfr1.
(2) Yfr2 through Yfr5 have several entries in Med4 and WH 8102, which might be assumed
for other strains as well, and are in length and the degree of mutual identity similar to
four ncRNAs implicated in quorum sensing in Vibrio species (Lenz et al., 2004).
(3) The longer RNA Yfr6 seems to be restricted to the smallest marine genomes, as it
was only found in Med4 and SS120. Its very short half-life and a conserved short reading
frame represent unique properties compared to the other Med4 ncRNAs analysed experi-
mentally.
(4) The marine Yfr7 RNA was shown to represent an orthologue of 6Sa RNA (ssaA gene
product) of freshwater and other cyanobacteria as well as for the 6S RNA known from
γ-proteobacteria; described in detail in the next section.

Consequently, the ncRNAs identified here may constitute important regulatory or struc-
tural components not only of a free-living marine cyanobacterium but also for the whole
cyanobacterial radiation. Up to now, the detailed regulatory mechanisms of the detected
ncRNAs are unclear, and a genetic manipulation system, especially for such short genes,
90 4. DISCUSSION

is still a challenging task for marine strains. Thus, the ncRNAs which possess ortho-
logues over a wider phylogenetic distance might be functionally analysed directly in these
cyanobacteria for which well-established genetic tools exist, as the freshwater strains Syne-
chococcus PCC 7942 or Synechocystis PCC 6803. Therefore, promising candidates are Yfr1
and Yfr7/6Sa, which are likely present in all cyanobacteria.
Another possibility would be the transfer of the ncRNA as well as the putative target(s)
to an appropriate host or model organism to study the mode of ncRNA interaction, of-
ten an antisense mechanism. For unicellular marine cyanobacteria, Synechococcus WH
7803 might become such a model: There is a genetic manipulation system described (Bra-
hamsha, 1996) and its genome analysis has almost completely been finished (Partensky,
2006). Additionally, whole genome microarrays for WH 7803 are in progress (SynChips
project of the European network Marine Genomics, PI: W. Hess), which will allow high-
through-put experiments for putative ncRNA-deficient strains in the future.

4.2.3 Ubiquitous presence of 6S RNA in cyanobacteria and clues about

its function
In an experimental and detailed computational analysis the marine Yfr7 RNA was iden-
tified as the orthologue of 6Sa RNA (ssaA gene product), described and partly annotated
already for several freshwater and other cyanobacteria. Additionally, a recently published
study of Barrick et al. (2005) detected 6Sa computationally in all cyanobacterial strains
representing the orthologue of 6S RNA known from γ-proteobacteria. Unfortunately, ge-
nomic annotations of 6Sa RNA (ssaA) in cyanobacteria are often on the incorrect strand.
The transcription of the ssaA gene from the complement strand as annotated was demon-
strated for Thermosynechococcus elongatus, Anabaena PCC 7120 and Gloeobacter violaceus
PCC 7421 in this study; for Synechococcus PCC 6301 and Synechocystis PCC 6803 the
annotated genome location of ssaA was shown to be correct.

E. coli 6S RNA binds to the σ 70 polymerase holoenzyme to globally regulate gene expres-
sion in response to the shift from exponential growth to stationary phase. The inhibition
of transcription occurs by direct competition for promoter DNA binding to Eσ 70 , the
house-keeping form of RNA polymerase. Thereby, 6S RNA is mimicking the structure
of a DNA template in an open promoter complex (Barrick et al., 2005; Trotochaud and
Wassarman, 2005). The highly stable secondary structure of 6S RNA essential for this
mechanism, containing a single-stranded central bulge within a highly double-stranded
molecule (Trotochaud and Wassarman, 2005), was also predicted for the Yfr7/6Sa RNA
identified here. Thus, at least a structural ortholog of γ-proteobacterial 6S RNA is con-
served within the whole cyanobacterial lineage.
In E. coli the 6S RNA is mainly active under stress conditions, like nutrient limitation,
and an accumulation can be observed during entry into stationary phase. To get insights
into the situation in cyanobacterial strains as well, the growth-phase dependent expres-
sion of two transcripts of the Med4 6S RNA homolog have been identified in this study.
One transcript of the Med4 6S RNA (220 nt) reaches maximal accumulation at high cell
densities. The second, 332 nt long and processed transcript is peaking earlier in Med4,
probably during entry into stationary phase. The existence of two differentially expressed
4.2. NEW FAMILIES OF SMALL RNAS IN CYANOBACTERIA 91

6S RNAs, 6Sa and 6Sb, was recently shown for Bacillus subtilis, where 6Sb is the or-
tholog to E. coli 6S, and 6Sa has functionally diverged (Barrick et al., 2005). Although
both transcripts of Med4 (and other high-light-adapted strains) originate from only one
genomic locus, the different growth-phase dependent expression of Yfr7/6S transcripts in
Med4 may also indicate a fine tuning of the transcriptional response to nutrient limitation.

Additionally, a secondary effect of 6S RNA in the activation of σ S -dependent promoter

transcription was observed in E. coli cells (Trotochaud and Wassarman, 2004) mediated
by the general stress response factor σ S , which is the only group 2 σ factor in E. coli.
In contrast, cyanobacteria likely possess a more complex regulatory network of group 2 σ
factors. They harbour multiple group 2 σ factors, so that a single orthologous gene can
not be assigned to the enterobacterial stationary phase σ factor, σ S . It was suggested
that these multiple group 2 σ factors are differentially active during the day, transmit-
ting the circadian rhythm generated by the cyanobacterial clock (Nair et al., 2002; Ditty
et al., 2003). Interestingly, a time-shifted, rhythmic expression pattern exists between
both Med4 Yfr7/6S transcripts, which was demonstrated using RNA samples from syn-
chronised cultures.
Therefore, it was of great interest to find out if the cyanobacterial Yfr7/6S RNA is also
functionally related to the γ-proteobacterial one. A common experimental approach to
achieve more information about the gene of interest is the construction of its knock-out
strain. As there was a transformation system already tested for marine Synechococcus,
the yfr7/ssaA gene of WH 8102 was targeted for a single-cross-over insertion. Although
the insertion was found to be successful, it was not possible to isolate Yfr7/6S-deficient
cells, up to now. But the widespread occurrence of this ncRNA opens additional exciting
opportunities to test the function of 6S directly in cyanobacteria like Synechocystis PCC
6803, which are easier and faster to manipulate.

4.2.4 A novel antisense RNA to isiA mRNA in Synechocystis PCC 6803

In cyanobacteria, the iron stress induced gene A, isiA, encodes a chlorophyll binding pro-
tein homologous to the photosystem II (PSII) protein PsbC (CP43). High attention was
paid to IsiA (CP43’) when it was detected to form a giant alternative antennae ring around
the photosystem I (PSI) trimer under iron limitation (Bibby et al., 2001a; Boekema et al.,
2001). However, the precise role and regulation of isiA gene expression has remained enig-
matic as it is also induced under high light, salt and oxidative stress (Vinnemeier et al.,
1998; Jeanjean et al., 2003; Yousef et al., 2003; Havaux et al., 2005). It was suggested that
IsiA may serve as a stress-inducible antennae in the dissipation of excess light energy or
as storage for excess chlorophyll a (Cadoret et al., 2004; Ihalainen et al., 2005).

Here, the transcription of an antisense RNA, IsrR (Iron stress repressed RNA) from the
isiA complement strand in the freshwater cyanobacterium Synechocystis PCC 6803 was
demonstrated, that is inversely regulated to the accumulation of its host gene’s mRNA.
Thus, at least four different transcripts were shown to originate from the isiAB genomic
locus: The unprocessed isiAB precursor transcript, the isiA monocistronic transcript,
IsrR with maximum accumulation under iron replete and low-light conditions, when isiA
mRNA is missing, and the 5’ UTR transcript which appears to be co-accumulated with
92 4. DISCUSSION

isiA mRNA.
For each newly identified non-coding RNA it often appears challenging to find out to which
targets it may act. In this example, the reciprocal relation between the amounts IsrR and
isiA mRNA suggests that they could react to each other targeting both for a degrada-
tion mechanism. An antisense RNA that is degraded together with its target makes a
perfect reversible switch to respond to environmental changes. Indeed, also in the case of
trans-encoded ncRNAs with imperfect complementarity, a growing number of examples
indicate that they can target specific mRNAs for degradation by RNase III. Examples are
the IstR RNAs in E. coli (Vogel et al., 2004) and the spa mRNA in Staphylococcus aureus
(Huntzinger et al., 2005). Thus, the formation of ncRNA-protein complexes containing
RNase III, and in some cases RNase E (Morita et al., 2005), could be a general way by
which mRNAs are destabilised by small RNAs in bacteria in a similar way as in higher
organisms (Morita et al., 2005).

Depending on the respective stress, different levels of isiA induction were observed: A re-
sponse to high light or iron depletion began after about 17 hours with increasing amounts
of isiA mRNA until at least 48 hours, whereas a very rapid but short burst of expression
occurred as a response to H2 O2 treatment. This underscores the well decided answer of
the Synechocystis cell to the chosen conditions and might refer to the underlying mecha-
nism of isiA induction: The triggering factor for a highly induced isiA expression could
be a redox stress over a specified period, which orders the cell to arrange the PSI-IsiA
supercomplexes. Interesting future experiments might be the variation of the amount and
period for the stress trigger, high light or H2 O2 .

Together, these results suggested a stringent regulatory mechanism acting on the expres-
sion of isiA at the RNA level, at least. The coupled degradation of IsrR/isiA might
constitute a perfectly timed reversible switch to respond to environmental changes in iron
concentrations, redox conditions and possibly other stresses (Duehring et al., 2006).

The costly expression of giant amounts of isiA mRNA under stress conditions and the
later high level of IsiA protein arranging into a large alternative antennae ring might
indicate a reason why the cell utilises this special antisense mechanism as an additional
level of regulation for isiA and likely other genes, whose expression is only advantageous
for the cell under defined conditions. Moreover, the latest experiments in Synechocystis
(performed by Ulf Duehring (Duehring et al., 2006)) provided further evidence that the
novel IsrR-mediated regulatory mechanism affects not only the isiA mRNA abundance
but also the assembly of IsiA-PSI supercomplexes as shown in Figure 4.1.

IsrR is the first non-coding RNA regulating a component involved in oxygenic photosyn-
thesis. The fact, that it is located over its full length complementary to a protein-coding
region encoding its likely target, might open a new view on bacterial regulatory RNAs,
which have been thought so far to be transcribed mainly from intergenic regions. Thus,
a future focus to protein-coding regions, especially for genes underlying stringent but un-
known regulatory mechanisms, can be suggested for the bacterial genomes. Further on, the
designed opening of a well-defined expression window for a mRNA target by a cis-acting
RNA is more alike mechanisms reported for eukaryotes. Again, a regulatory feature might
4.3. CONCLUSION AND OUTLOOK 93

[-] -
wi B F +

rR ] Fe

-
M rR -

Fe
Is Fe
Fe
ld e-

u: [ +
M ype
∆i ype

Is
t
A
ld

u:
Chlorophyll-containing

si
wi
complexes

PSI-IsiA supercomplexes

PSI-trimeric complexes

PSII-dimer

PSI-monomeric complexes
PSII-monomer

Figure 4.1: Influence of IsrR on the accumulation of trimeric PSI and PSI-IsiA supercomplexes (Duehring
et al., 2006). Membrane protein complexes isolated from thylakoids of Synechocystis PCC 6803 wild type,
an isiAB deletion mutant, the IsrR overexpressor (Mu:IsrR[+]) mutant, and cells with reduced amounts
of IsrR (Mu:IsrR[-]). Thylakoid membranes were isolated before (Fe+) and after 2 days of iron starvation
(Fe-) and separated by Blue-Native-PAGE. Coomassie-stained sections of the Blue-Native-gel are shown.
Chlorophyll containing protein complexes are indicated on the right. PS, photosystem.

not be invented the first time for a higher organism and is likely existing for prokaryotes
as well.

4.3 Conclusion and outlook

Computational prediction algorithms are often limited in detecting biological relevant sig-
nals (Hu et al., 2005). Thus, a trend of recent motif discovery algorithms is to incorporate
additional information to improve the prediction accuracy. For this purpose, comparative
studies are often implemented to identify conservations between genomes, which are as-
sumed to represent essential elements of the group of organisms investigated. In this study
about marine cyanobacterial genomes, it was successfully demonstrated that the compar-
ison of four related genomes revealed so far unknown motifs for transcription factors and
even new genes for non-coding RNAs, which might have never been detected within a
single genome.
The global importance of the marine ecosystem including the major group of oxygen-
producing Synechococcus/Prochlorococcus becomes indisputable these days. For these
organisms, a giant gene pool exists in the ocean, which is mixed up by viruses and possi-
bly other mechanisms of lateral gene transfer still to be discovered. Only a tiny fraction
of the marine biodiversity is being cultivated in the laboratory, and from an even smaller
number genome sequences are available for studies like this.
One has to keep in mind the fact, that one genome sequence resembles only a snapshot for a
single organism. In reality, this organism is part of a larger community which is constantly
changing to adapt to variations in its environment. Thus, a suggestive approach might
be analysing many related bacteria in parallel - the microbial "pan-genome" consisting of
94 4. DISCUSSION

a core genome shared by all members plus a dispensable genome consisting of partially
shared and strain-specific genes (Tettelin et al., 2005) to better understand evolutionary
processes. Thereby, RNA genes as well as transcription factors constitute very informative
sequences to connect the genome information to the conditions set by the respective en-
vironmental constraints. Additionally, phylogenetic analysis of both, DNA elements and
the proteins that recognise them, might provide new conclusions about their evolutionary
history as demonstrated for LexA (Mazon et al., 2004a) and T7 RNA polymerase (Chen
and Schneider, 2005).
Environmental sequence data are now available (Venter et al., 2004) and ongoing genome
projects are providing a fast-growing number of sequences, although unique genes will con-
tinue to be identified even after sequencing hundreds of genomes (Tettelin et al., 2005).
Thus, new ways need to be explored to gain deeper insights into complex networks such
as the community of marine cyanobacteria and their phages.
Bibliography

Altschul, S., Gish, W., Miller, W., Myers, E., and Lipman, D. (1990). Basic local alignment
search tool. J Mol Biol, 215(3):403–10.

Altuvia, S., Weinstein-Fischer, D., Zhang, A., Postow, L., and Storz, G. (1997). A small,
stable RNA induced by oxidative stress: role as a pleiotropic regulator and antimutator.
Cell, 90(1):43–53.

Altuvia, S., Zhang, A., Argaman, L., Tiwari, A., and Storz, G. (1998). The Escherichia coli
OxyS regulatory RNA represses fhlA translation by blocking ribosome binding. EMBO
J, 17(20):6069–75.

Argaman, L. and Altuvia, S. (2000). fhlA repression by OxyS RNA: kissing complex
formation at two sites results in a stable antisense-target RNA complex. J Mol Biol,
300(5):1101–12.

Argaman, L., Hershberg, R., Vogel, J., Bejerano, G., Wagner, E., Margalit, H., and Al-
tuvia, S. (2001). Novel small RNA-encoding genes in the intergenic regions of Escherichia
coli. Curr Biol, 11(12):941–50.

Asato, Y. (2003). Toward an understanding of cell growth and the cell division cycle of
unicellular photoautotrophic cyanobacteria. Cell Mol Life Sci, 60(4):663–87.

Asayama, M., Imamura, S., Yoshihara, S., Miyazaki, A., Yoshida, N., Sazuka, T., Kaneko,
T., Ohara, O., Tabata, S., Osanai, T., Tanaka, K., Takahashi, H., and Shirai, M. (2004).
SigC, the group 2 sigma factor of RNA polymerase, contributes to the late-stage gene
expression and nitrogen promoter recognition in the cyanobacterium Synechocystis sp.
strain PCC 6803. Biosci Biotechnol Biochem, 68(3):477–87.

Axmann, I., Kensche, P., Vogel, J., Kohl, S., Herzel, H., and Hess, W. (2005). Identification
of cyanobacterial non-coding RNAs by comparative genome analysis. Genome Biol,
6(9):R73.

Axmark, D., Larsson, A., and Widenius, M. (2006). MySQL AB: The world’s most popular
open source database. [Link]

Barrick, J., Sudarsan, N., Weinberg, Z., Ruzzo, W., and Breaker, R. (2005). 6S RNA is a
widespread regulator of eubacterial RNA polymerase that resembles an open promoter.
RNA, 11(5):774–84.
Barry, G., Squires, C., and Squires, C. (1980). Attenuation and processing of RNA from
the rplJL–rpoBC transcription unit of Escherichia coli. Proc Natl Acad Sci U S A,
77(6):3331–5.

Bell-Pedersen, D., Cassone, V., Earnest, D., Golden, S., Hardin, P., Thomas, T., and
Zoran, M. (2005). Circadian rhythms from multiple oscillators: lessons from diverse
organisms. Nat Rev Genet, 6(7):544–56.

Bensing, B., Meyer, B., and Dunny, G. (1996). Sensitive detection of bacterial transcrip-
tion initiation sites and differentiation from RNA processing sites in the pheromone-
induced plasmid transfer system of Enterococcus faecalis. Proc Natl Acad Sci U S A,
93(15):7794–9.

Berriman, M. and Rutherford, K. (2003). Viewing and annotating sequence data with
Artemis. Brief Bioinform, 4(2):124–32.

Bertilsson, S., Berglund, O., Karl, D., and Chisholm, S. (2003). Elemental composition of
marine prochlorococcus and synechococcus: Implications for the ecological stoichiome-
try of the sea. Limnology and Oceanography, 48:1721–31.

Bibby, T., Nield, J., and Barber, J. (2001a). Iron deficiency induces the formation of an
antenna ring around trimeric photosystem I in cyanobacteria. Nature, 412(6848):743–5.

Bibby, T., Nield, J., Partensky, F., and Barber, J. (2001b). Oxyphotobacteria. Antenna
ring around photosystem I. Nature, 413(6856):590.

Birge, E. (2000). Bacterial and Bacteriophage Genetics (4th edn). Springer Verlag, New
York.

Birkey, S., Liu, W., Zhang, X., Duggan, M., and Hulett, F. (1998). Pho signal transduc-
tion network reveals direct transcriptional regulation of one two-component system by
another two-component regulator: Bacillus subtilis PhoP directly regulates production
of ResD. Mol Microbiol, 30(5):943–53.

Blanco, A., Sola, M., Gomis-Ruth, F., and Coll, M. (2002). Tandem DNA recognition
by PhoB, a two-component signal transduction transcriptional activator. Structure,
10(5):701–13.

Blattner, F., Plunkett, 3rd, G., Bloch, C., Perna, N., Burland, V., Riley, M., Collado-
Vides, J., Glasner, J., Rode, C., Mayhew, G., Gregor, J., Davis, N., Kirkpatrick, H.,
Goeden, M., Rose, D., Mau, B., and Shao, Y. (1997). The complete genome sequence
of Escherichia coli K-12. Science, 277(5331):1453–74.

Boekema, E., Hifney, A., Yakushevska, A., Piotrowski, M., Keegstra, W., Berry, S., Michel,
K., Pistorius, E., and Kruip, J. (2001). A giant chlorophyll-protein complex induced by
iron deficiency in cyanobacteria. Nature, 412(6848):745–8.

Brahamsha, B. (1996). A genetic manipulation system for oceanic cyanobacteria of the

genus Synechococcus. Appl Environ Microbiol, 62(5):1747–51.
Bruyant, F., Babin, M., Genty, B., Prasil, O., Behrenfeld, M., Claustre, H., Bricaud, A.,
Holtzendorff, J., Koblizek, M., Garczareck, L., and Partensky, F. (2005). Diel variations
in the photosynthetic parameters of prochlorococcus strain pcc 9511: combined effects
of light and cell cycle. Limnology and Oceanography, 50:850–63.
Bryant, D. (2003). The beauty in small things revealed. Proc Natl Acad Sci U S A,
100(17):9647–9.
Bulyk, M. (2003). Computational prediction of transcription-factor binding site locations.
Genome Biol, 5(1):201.
Cadoret, J., Demouliere, R., Lavaud, J., van Gorkom, H., Houmard, J., and Etienne, A.
(2004). Dissipation of excess energy triggered by blue light in cyanobacteria with CP43’
(isiA). Biochim Biophys Acta, 1659(1):100–4.
Cao, Q. and Kibbe, W. (2006). Oligonucleotide Properties Calculator.
[Link]
Chen, S., Lesnik, E., Hall, T., Sampath, R., Griffey, R., Ecker, D., and Blyn, L. (2002).
A bioinformatics based approach to discover small RNA genes in the Escherichia coli
genome. Biosystems, 65(2-3):157–77.
Chen, Z. and Schneider, T. (2005). Information theory based T7-like promoter models:
classification of bacteriophages and differential evolution of promoters and their poly-
merases. Nucleic Acids Res, 33(19):6172–87.
Chisholm, P. (2006). CHISHOLM LAB: Protocols, natural seawater-based PRO99
medium. [Link]
Chisholm, S., Frankel, S., Goericke, R., Olson, R., Palenik, B., Waterbury, J. B., West-
Johnsrud, L., and Zettler, E. (1992). Prochlorococcus marinus nov. gen. nov. sp.: an
oxyphototrophic marine prokaryote containing divinyl chlorophyll a and b. Archives of
Microbiology, 157:297–300.
Chisholm, S., Olson, R., Zettler, E., Goericke, R., Waterbury, J., and Welschmeyer, N.
(1988). A novel free-living prochlorophyte abundant in the oceanic euphotic zone. Na-
ture, 334:340–3.
Cliften, P., Hillier, L., Fulton, L., Graves, T., Miner, T., Gish, W., Waterston, R., and
Johnston, M. (2001). Surveying Saccharomyces genomes to identify functional elements
by comparative DNA sequence analysis. Genome Res, 11(7):1175–86.
Crooks, G., Hon, G., Chandonia, J., and Brenner, S. (2004). WebLogo: a sequence logo
generator. Genome Res, 14(6):1188–90.
Crooks, G., Hon, G., Chandonia, J., and Brenner, S. (2006). WebLogo.
[Link]/cgi-bin/seqlogo/[Link].
Csiszar, K., Houmard, J., Damerval, T., and Tandeau de Marsac, N. (1987). Transcrip-
tional analysis of the cyanobacterial gvpABC operon in differentiated cells: occurrence
of an antisense RNA complementary to three overlapping transcripts. Gene, 60(1):29–
37.
Curtis, S. (1987). Genes encoding the beta and epsilon subunits of the proton-translocating
ATPase from Anabaena sp. strain PCC 7120. J Bacteriol, 169(1):80–6.

di Bernardo, D., Down, T., and Hubbard, T. (2003). ddbRNA: detection of conserved
secondary structures in multiple alignments. Bioinformatics, 19(13):1606–11.

Ditty, J., Williams, S., and Golden, S. (2003). A cyanobacterial circadian timing mecha-
nism. Annu Rev Genet, 37:513–43.

Domain, F., Houot, L., Chauvat, F., and Cassier-Chauvat, C. (2004). Function and
regulation of the cyanobacterial genes lexA, recA and ruvB: LexA is critical to the
survival of cells facing inorganic carbon starvation. Mol Microbiol, 53(1):65–80.

Douglas, S. (1998). Plastid evolution: origins, diversity, trends. Curr Opin Genet Dev,
8(6):655–61.

Dubchak, I. and Frazer, K. (2003). Multi-species sequence comparison: the next frontier
in genome annotation. Genome Biol, 4(12):122.

Duehring, U., Axmann, I., Hess, W., and Wilde, A. (2006). An internal antisense rna
regulates expression of the photosynthesis gene isia. submitted.

Dufresne, A., Garczarek, L., and Partensky, F. (2005). Accelerated evolution associated
with genome reduction in a free-living prokaryote. Genome Biol, 6(2):R14.

Dufresne, A., Salanoubat, M., Partensky, F., Artiguenave, F., Axmann, I., Barbe, V.,
Duprat, S., Galperin, M., Koonin, E., Le Gall, F., Makarova, K., Ostrowski, M., Oztas,
S., Robert, C., Rogozin, I., Scanlan, D., Tandeau de Marsac, N., Weissenbach, J.,
Wincker, P., Wolf, Y., and Hess, W. (2003). Genome sequence of the cyanobacterium
Prochlorococcus marinus SS120, a nearly minimal oxyphototrophic genome. Proc Natl
Acad Sci U S A, 100(17):10020–5.

Dunn, J. and Studier, F. (1973a). T7 early RNAs and Escherichia coli ribosomal RNAs
are cut from large precursor RNAs in vivo by ribonuclease 3. Proc Natl Acad Sci U S
A, 70(12):3296–3300.

Dunn, J. and Studier, F. (1973b). T7 early RNAs are generated by site-specific cleavages.
Proc Natl Acad Sci U S A, 70(5):1559–63.

Dunn, J. and Studier, F. (1975). Effect of RNAase III, cleavage on translation of bacte-
riophage T7 messenger RNAs. J Mol Biol, 99(3):487–99.

Dunn, J. and Studier, F. (1983). Complete nucleotide sequence of bacteriophage T7 DNA

and the locations of T7 genetic elements. J Mol Biol, 166(4):477–535.

Dvornyk, V., Vinogradova, O., and Nevo, E. (2003). Origin and evolution of circadian
clock genes in prokaryotes. Proc Natl Acad Sci U S A, 100(5):2495–500.

Ellson, J. and North, S. (2006). Graphviz - Graph Visualization Software.

[Link]
Ermolaeva, M., Khalak, H., White, O., Smith, H., and Salzberg, S. (2000). Prediction of
transcription terminators in bacterial genomes. J Mol Biol, 301(1):27–33.

Falkowski, P. (2002). The ocean’s invisible forest. Sci Am, 287(2):54–61.

Field, C., Behrenfeld, M., Randerson, J., and Falkowski, P. (1998). Primary production of
the biosphere: integrating terrestrial and oceanic components. Science, 281(5374):237–
40.

Franche, C. and Damerval, T. (1988). Tests on nif probes and dna hybridizations. Methods
Enzymol., 167:803–8.

Fraser, C. (2006). tigr fams: tigr protein families. [Link]

Frazer, K., Elnitski, L., Church, D., Dubchak, I., and Hardison, R. (2003). Cross-species
sequence comparisons: a review of methods and available resources. Genome Res,
13(1):1–12.

Frith, M., Hansen, U., Spouge, J., and Weng, Z. (2004). Finding functional sequence
elements by multiple local alignment. Nucleic Acids Res, 32(1):189–200.

Fuhrman, J. and Capone, D. (2001). Nifty nanoplankton. Nature, 412(6847):593–4.

Fuller, N., Marie, D., Partensky, F., Vaulot, D., Post, A., and Scanlan, D. (2003). Clade-
specific 16S ribosomal DNA oligonucleotides reveal the predominance of a single marine
Synechococcus clade throughout a stratified water column in the Red Sea. Appl Environ
Microbiol, 69(5):2430–43.

Garcia-Fernandez, J., Hess, W., Houmard, J., and Partensky, F. (1998). Expression of the
psbA gene in the marine oxyphotobacteria Prochlorococcus spp. Arch Biochem Biophys,
359(1):17–23.

Garczarek, L., Partensky, F., Irlbacher, H., Holtzendorff, J., Babin, M., Mary, I., Thomas,
J., and Hess, W. (2001). Differential expression of antenna and core genes in Prochloro-
coccus PCC 9511 (Oxyphotobacteria) grown under a modulated light-dark cycle. Env-
iron Microbiol, 3(3):168–75.

Gaudin, C., Zhou, X., Williams, K., and Felden, B. (2002). Two-piece tmRNA in cyanobac-
teria and its structural analysis. Nucleic Acids Res, 30(9):2018–24.

Gelfand, M., Koonin, E., and Mironov, A. (2000). Prediction of transcription regulatory
sites in Archaea by a comparative genomic approach. Nucleic Acids Res, 28(3):695–705.

Gerhart, E., Wagner, H., and Brantl, S. (1998). Kissing and RNA stability in antisense
control of plasmid replication. Trends Biochem Sci, 23(12):451–4.

Golden, S., Brusslan, J., and Haselkorn, R. (1986). Expression of a family of psbA genes
encoding a photosystem II polypeptide in the cyanobacterium Anacystis nidulans R2.
EMBO J, 5(11):2789–98.

Golden, S. and Canales, S. (2003). Cyanobacterial circadian clocks–timing is everything.

Nat Rev Microbiol, 1(3):191–9.
Golden, S., Ishiura, M., Johnson, C., and Kondo, T. (1997). CYANOBACTERIAL CIR-
CADIAN RHYTHMS. Annu Rev Plant Physiol Plant Mol Biol, 48:327–354.

Gottesman, S. (1984). Bacterial regulation: global regulatory networks. Annu Rev Genet,
18:415–41.

Gottesman, S. (2004). The small RNA regulators of Escherichia coli: roles and mecha-
nisms*. Annu Rev Microbiol, 58:303–28.

Gottesman, S. (2005). Micros for microbes: non-coding regulatory RNAs in bacteria.

Trends Genet, 21(7):399–404.

Gottgens, B., Gilbert, J., Barton, L., Grafham, D., Rogers, J., Bentley, D., and Green,
A. (2001). Long-range comparison of human and mouse SCL loci: localized regions
of sensitivity to restriction endonucleases correspond precisely with peaks of conserved
noncoding sequences. Genome Res, 11(1):87–97.

Griffiths-Jones, S., Moxon, S., Marshall, M., Khanna, A., Eddy, S., and Bateman, A.
(2005). Rfam: annotating non-coding RNAs in complete genomes. Nucleic Acids Res,
33(Database issue):D121–4.

Grundy, F. and Henkin, T. (2003). The T box and S box transcription termination control
systems. Front Biosci, 8:d20–31.

Gutekunst, K., Phunpruch, S., Schwarz, C., Schuchardt, S., Schulz-Friedrich, R., and
Appel, J. (2005). LexA regulates the bidirectional hydrogenase in the cyanobacterium
Synechocystis sp. PCC 6803 as a transcription activator. Mol Microbiol, 58(3):810–23.

Hagstrom, A., Pommier, T., Rohwer, F., Simu, K., Stolte, W., Svensson, D., and Zweifel,
U. (2002). Use of 16S ribosomal DNA for delineation of marine bacterioplankton species.
Appl Environ Microbiol, 68(7):3628–33.

Havaux, M., Guedeney, G., Hagemann, M., Yeremenko, N., Matthijs, H., and Jeanjean,
R. (2005). The chlorophyll-binding protein IsiA is inducible by high light and protects
the cyanobacterium Synechocystis PCC6803 from photooxidative stress. FEBS Lett,
579(11):2289–93.

Hawley, D. and McClure, W. (1983). Compilation and analysis of Escherichia coli promoter
DNA sequences. Nucleic Acids Res, 11(8):2237–55.

Hernandez, J., Muro-Pastor, A., Flores, E., Bes, M., Peleato, M., and Fillat, M. (2006).
Identification of a furA cis antisense RNA in the cyanobacterium Anabaena sp. PCC
7120. J Mol Biol, 355(3):325–34.

Herrero, A., Muro-Pastor, A., and Flores, E. (2001). Nitrogen control in cyanobacteria. J
Bacteriol, 183(2):411–25.

Hess, W. (2004). Genome analysis of marine photosynthetic microbes and their global
role. Curr Opin Biotechnol, 15(3):191–8.

Hess, W. (2006). Cyanolab: HESS lab - research on cyanobacteria.

[Link]
Hess, W., Rocap, G., Ting, C., Larimer, F., Stilwagen, S., Lamerdin, J., and Chisholm, S.
(2001). The photosynthetic apparatus of Prochlorococcus: Insights through comparative
genomics. Photosynth Res, 70(1):53–71.

Hu, J., Li, B., and Kihara, D. (2005). Limitations and potentials of current motif discovery
algorithms. Nucleic Acids Res, 33(15):4899–913.

Huntzinger, E., Boisset, S., Saveanu, C., Benito, Y., Geissmann, T., Namane, A., Lina, G.,
Etienne, J., Ehresmann, B., Ehresmann, C., Jacquier, A., Vandenesch, F., and Romby,
P. (2005). Staphylococcus aureus RNAIII and the endoribonuclease III coordinately
regulate spa gene expression. EMBO J, 24(4):824–35.

Huttenhofer, A., Schattner, P., and Polacek, N. (2005). Non-coding RNAs: hope or hype?
Trends Genet, 21(5):289–97.

Ihalainen, J., D’Haene, S., Yeremenko, N., van Roon, H., Arteni, A., Boekema, E., van
Grondelle, R., Matthijs, H., and Dekker, J. (2005). Aggregates of the chlorophyll-binding
protein IsiA (CP43’) dissipate energy in cyanobacteria. Biochemistry, 44(32):10846–53.

Imamura, S., Asayama, M., and Shirai, M. (2004). In vitro transcription analysis by
reconstituted cyanobacterial RNA polymerase: roles of group 1 and 2 sigma factors and
a core subunit, RpoC2. Genes Cells, 9(12):1175–87.

Imamura, S., Asayama, M., Takahashi, H., Tanaka, K., Takahashi, H., and Shirai, M.
(2003a). Antagonistic dark/light-induced SigB/SigD, group 2 sigma factors, expression
through redox potential and their roles in cyanobacteria. FEBS Lett, 554(3):357–62.

Imamura, S., Yoshihara, S., Nakano, S., Shiozaki, N., Yamada, A., Tanaka, K., Takahashi,
H., Asayama, M., and Shirai, M. (2003b). Purification, characterization, and gene
expression of all sigma factors of RNA polymerase in a cyanobacterium. J Mol Biol,
325(5):857–72.

Imburgio, D., Rong, M., Ma, K., and McAllister, W. (2000). Studies of promoter recog-
nition and start site selection by T7 RNA polymerase using a comprehensive collection
of promoter variants. Biochemistry, 39(34):10419–30.

Ishihama, A. (2000). Functional modulation of Escherichia coli RNA polymerase. Annu

Rev Microbiol, 54:499–518.

Ishiura, M., Kutsuna, S., Aoki, S., Iwasaki, H., Andersson, C., Tanabe, A., Golden, S.,
Johnson, C., and Kondo, T. (1998). Expression of a gene cluster kaiABC as a circadian
feedback process in cyanobacteria. Science, 281(5382):1519–23.

Ivleva, N., Bramlett, M., Lindahl, P., and Golden, S. (2005). LdpA: a component of the
circadian clock senses redox state of the cell. EMBO J, 24(6):1202–10.

Iwasaki, H. and Kondo, T. (2004). Circadian timing mechanism in the prokaryotic clock
system of cyanobacteria. J Biol Rhythms, 19(5):436–44.

Jacquet, S., Partensky, F., Marie, D., Casotti, R., and Vaulot, D. (2001). Cell cycle
regulation by light in Prochlorococcus strains. Appl Environ Microbiol, 67(2):782–90.
Jeanjean, R., Zuther, E., Yeremenko, N., Havaux, M., Matthijs, H., and Hagemann, M.
(2003). A photosystem 1 psaFJ-null mutant of the cyanobacterium Synechocystis PCC
6803 expresses the isiAB operon under iron replete conditions. FEBS Lett, 549(1-3):52–
6.

Johansson, J. and Cossart, P. (2003). RNA-mediated control of virulence gene expression

in bacterial pathogens. Trends Microbiol, 11(6):280–5.

Johansson, J., Mandin, P., Renzoni, A., Chiaruttini, C., Springer, M., and Cossart, P.
(2002). An RNA thermosensor controls expression of virulence genes in Listeria mono-
cytogenes. Cell, 110(5):551–61.

Johnson, C. (2004). Precise circadian clocks in prokaryotic cyanobacteria. Curr Issues

Mol Biol, 6(2):103–10.

Johnson, P. and Sieburth, J. (1979). Chroococcoid cyanobacteria in the sea - ubiquitous

and diverse phototropic biomass. Limnology and Oceanography, 24:928–35.

Johnson, Z., Zinser, E., Coe, A., McNulty, N., Woodward, E., and Chisholm, S. (2006).
Niche partitioning among prochlorococcus ecotypes along ocean-scale environmental
gradients. Science, in press.

Kaneko, T., Nakamura, Y., Wolk, C., Kuritz, T., Sasamoto, S., Watanabe, A., Iriguchi,
M., Ishikawa, A., Kawashima, K., Kimura, T., Kishida, Y., Kohara, M., Matsumoto,
M., Matsuno, A., Muraki, A., Nakazaki, N., Shimpo, S., Sugimoto, M., Takazawa, M.,
Yamada, M., Yasuda, M., and Tabata, S. (2001). Complete genomic sequence of the
filamentous nitrogen-fixing cyanobacterium Anabaena sp. strain PCC 7120. DNA Res,
8(5):205–13; 227–53.

Kaneko, T. and Tabata, S. (1997). Complete genome structure of the unicellular cyanobac-
terium Synechocystis sp. PCC6803. Plant Cell Physiol, 38(11):1171–6.

Katayama, M., Tsinoremas, N., Kondo, T., and Golden, S. (1999). cpmA, a gene involved
in an output pathway of the cyanobacterial circadian system. J Bacteriol, 181(11):3516–
24.

Kawano, M., Reynolds, A., Miranda-Rios, J., and Storz, G. (2005). Detection of 5’- and 3’-
UTR-derived small RNAs and cis-encoded antisense RNAs in Escherichia coli. Nucleic
Acids Res, 33(3):1040–50.

Keiler, K., Shapiro, L., and Williams, K. (2000). tmRNAs that encode proteolysis-
inducing tags are found in all known bacterial genomes: A two-piece tmRNA functions
in Caulobacter. Proc Natl Acad Sci U S A, 97(14):7778–83.

Kensche, P. (2004). Regulatory rna elements - a motif search in cyanobacterial genomes.

Master’s thesis, Humboldt University Berlin.

Kensche, P. (2005). Complete results for non-coding rna screening of cyanobacteria.

[Link]
Kielbasa, S., Gonze, D., and Herzel, H. (2005). Measuring similarities between transcrip-
tion factor binding sites. BMC Bioinformatics, 6:237.

Kielbasa, S., Korbel, J., Beule, D., Schuchhardt, J., and Herzel, H. (2001). Combin-
ing frequency and positional information to predict transcription factor binding sites.
Bioinformatics, 17(11):1019–26.

Kolb, A., Busby, S., Buc, H., Garges, S., and Adhya, S. (1993). Transcriptional regulation
by cAMP and its receptor protein. Annu Rev Biochem, 62:749–95.

Kondo, T., Mori, T., Lebedeva, N., Aoki, S., Ishiura, M., and Golden, S. (1997). Circadian
rhythms in rapidly dividing cyanobacteria. Science, 275(5297):224–7.

Korbel, J., Jensen, L., von Mering, C., and Bork, P. (2004). Analysis of genomic context:
prediction of functional associations from conserved bidirectionally transcribed gene
pairs. Nat Biotechnol, 22(7):911–7.

Kramer, R. and Rosenberg, M. (1976). The isolation and characterization of bacteriophage

T7 messenger RNA fragments containing an RNase III cleavage site. Nucleic Acids Res,
3(10):2411–26.

Kramer, R., Rosenberg, M., and Steitz, J. (1974). Nucleotide sequences of the 5’ and 3’
termini of bacteriophage T7 early messenger RNAs synthesized in vivo: evidence for
sequence specificity in RNA processing. J Mol Biol, 89(4):767–76.

Kucho, K., Okamoto, K., Tsuchiya, Y., Nomura, S., Nango, M., Kanehisa, M., and Ishiura,
M. (2005). Global analysis of circadian expression in the cyanobacterium Synechocystis
sp. strain PCC 6803. J Bacteriol, 187(6):2190–9.

Kutsuna, S., Nakahira, Y., Katayama, M., Ishiura, M., and Kondo, T. (2005). Transcrip-
tional regulation of the circadian clock operon kaiBC by upstream regions in cyanobac-
teria. Mol Microbiol, 57(5):1474–84.

La Roche, J., van der Staay, G., Partensky, F., Ducret, A., Aebersold, R., Li, R., Golden,
S., Hiller, R., Wrench, P., Larkum, A., and Green, B. (1996). Independent evolution of
the prochlorophyte and green plant chlorophyll a/b light-harvesting proteins. Proc Natl
Acad Sci U S A, 93(26):15244–8.

Lenz, D., Mok, K., Lilley, B., Kulkarni, R., Wingreen, N., and Bassler, B. (2004). The
small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio
harveyi and Vibrio cholerae. Cell, 118(1):69–82.

Lewin, B. (2000). genes VII. Oxford University Press, Oxford University Press Inc., New
York.

Lewis, B., Burge, C., and Bartel, D. (2005). Conserved seed pairing, often flanked by
adenosines, indicates that thousands of human genes are microRNA targets. Cell,
120(1):15–20.

Lindahl, L. and Zengel, J. (1986). Ribosomal genes in Escherichia coli. Annu Rev Genet,
20:297–326.
Lindell, D., Jaffe, J., Johnson, Z., Church, G., and Chisholm, S. (2005). Photosynthesis
genes in marine viruses yield proteins during host infection. Nature.

Lindell, D., Sullivan, M., Johnson, Z., Tolonen, A., Rohwer, F., and Chisholm, S. (2004).
Transfer of photosynthesis genes to and from Prochlorococcus viruses. Proc Natl Acad
Sci U S A, 101(30):11013–8.

Logemann, J., Schell, J., and Willmitzer, L. (1987). Improved method for the isolation of
RNA from plant tissues. Anal Biochem, 163(1):16–20.

Loots, G., Locksley, R., Blankespoor, C., Wang, Z., Miller, W., Rubin, E., and Frazer,
K. (2000). Identification of a coordinate regulator of interleukins 4, 13, and 5 by cross-
species sequence comparisons. Science, 288(5463):136–40.

Lopez-Maury, L., Florencio, F., and Reyes, J. (2003). Arsenic sensing and resis-
tance system in the cyanobacterium Synechocystis sp. strain PCC 6803. J Bacteriol,
185(18):5363–71.

Madan Babu, M. and Teichmann, S. (2003). Evolution of transcription factors and the
gene regulatory network in Escherichia coli. Nucleic Acids Res, 31(4):1234–44.

Mandal, M., Boese, B., Barrick, J., Winkler, W., and Breaker, R. (2003). Riboswitches
control fundamental biochemical pathways in Bacillus subtilis and other bacteria. Cell,
113(5):577–86.

Mandal, M. and Breaker, R. (2004). Gene regulation by riboswitches. Nat Rev Mol Cell
Biol, 5(6):451–63.

Mann, N. (2003). Phages of the marine cyanobacterial picophytoplankton. FEMS Micro-

biol Rev, 27(1):17–34.

Mann, N., Cook, A., Millard, A., Bailey, S., and Clokie, M. (2003). Marine ecosystems:
bacterial photosynthesis genes in a virus. Nature, 424(6950):741.

Martin, W., Rujan, T., Richly, E., Hansen, A., Cornelsen, S., Lins, T., Leister, D.,
Stoebe, B., Hasegawa, M., and Penny, D. (2002). Evolutionary analysis of Arabidop-
sis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of
cyanobacterial genes in the nucleus. Proc Natl Acad Sci U S A, 99(19):12246–51.

Martinez-Antonio, A. and Collado-Vides, J. (2003). Identifying global regulators in tran-

scriptional regulatory networks in bacteria. Curr Opin Microbiol, 6(5):482–9.

Mazon, G., Erill, I., Campoy, S., Cortes, P., Forano, E., and Barbe, J. (2004a). Re-
construction of the evolutionary history of the LexA-binding sequence. Microbiology,
150(Pt 11):3783–95.

Mazon, G., Lucena, J., Campoy, S., Fernandez de Henestrosa, A., Candau, P., and Barbe,
J. (2004b). LexA-binding sequences in Gram-positive and cyanobacteria are closely
related. Mol Genet Genomics, 271(1):40–9.

McDaniel, L., Houchin, L., Williamson, S., and Paul, J. (2002). Lysogeny in marine
Synechococcus. Nature, 415(6871):496.
McGuire, A., Hughes, J., and Church, G. (2000). Conservation of DNA regulatory motifs
and discovery of new motifs in microbial genomes. Genome Res, 10(6):744–57.

Millard, A., Clokie, M., Shub, D., and Mann, N. (2004). Genetic organization of the
psbAD region in phages infecting marine Synechococcus strains. Proc Natl Acad Sci U
S A, 101(30):11007–12.

Molineux, I. (2001). No syringes please, ejection of phage T7 DNA from the virion is
enzyme driven. Mol Microbiol, 40(1):1–8.

Moller, T., Franch, T., Hojrup, P., Keene, D., Bachinger, H., Brennan, R., and Valentin-
Hansen, P. (2002). Hfq: a bacterial Sm-like protein that mediates RNA-RNA interac-
tion. Mol Cell, 9(1):23–30.

Moore, L., Rocap, G., and Chisholm, S. (1998). Physiology and molecular phylogeny of
coexisting Prochlorococcus ecotypes. Nature, 393(6684):464–7.

Morden, C., Delwiche, C., Kuhsel, M., and Palmer, J. (1992). Gene phylogenies and the
endosymbiotic origin of plastids. Biosystems, 28(1-3):75–90.

Moreira, D., Le Guyader, H., and Philippe, H. (2000). The origin of red algae and the
evolution of chloroplasts. Nature, 405(6782):69–72.

Morita, T., Maki, K., and Aiba, H. (2005). RNase E-based ribonucleoprotein complexes:
mechanical basis of mRNA destabilization mediated by bacterial noncoding RNAs.
Genes Dev, 19(18):2176–86.

Mount, D. W. (2001). Bioinformatics: sequences and genome analysis. Cold Spring Harbor
Laboratority Press, Cold Spring Harbor, New York.

Nahvi, A., Barrick, J., and Breaker, R. (2004). Coenzyme B12 riboswitches are widespread
genetic control elements in prokaryotes. Nucleic Acids Res, 32(1):143–50.

Nair, U., Ditty, J., Min, H., and Golden, S. (2002). Roles for sigma factors in global
circadian regulation of the cyanobacterial genome. J Bacteriol, 184(13):3530–8.

Nakajima, M., Imai, K., Ito, H., Nishiwaki, T., Murayama, Y., Iwasaki, H., Oyama, T.,
and Kondo, T. (2005). Reconstitution of circadian oscillation of cyanobacterial KaiC
phosphorylation in vitro. Science, 308(5720):414–5.

Nakamura, Y., Kaneko, T., Sato, S., Ikeuchi, M., Katoh, H., Sasamoto, S., Watanabe,
A., Iriguchi, M., Kawashima, K., Kimura, T., Kishida, Y., Kiyokawa, C., Kohara, M.,
Matsumoto, M., Matsuno, A., Nakazaki, N., Shimpo, S., Sugimoto, M., Takeuchi, C.,
Yamada, M., and Tabata, S. (2002). Complete genome structure of the thermophilic
cyanobacterium Thermosynechococcus elongatus BP-1. DNA Res, 9(4):123–30.

Onizuka, T., Akiyama, H., Endo, S., Kanai, S., Hirano, M., Tanaka, S., and Miyasaka,
H. (2002). CO(2) response element and corresponding trans-acting factor of the pro-
moter for ribulose-1,5-bisphosphate carboxylase/oxygenase genes in Synechococcus sp.
PCC7002 found by an improved electrophoretic mobility shift assay. Plant Cell Physiol,
43(6):660–7.
Ortmann, A., Lawrence, J., and Suttle, C. (2002). Lysogeny and lytic viral production
during a bloom of the cyanobacterium Synechococcus spp. Microb Ecol, 43(2):225–31.

Overbeek, R., Begley, T., Butler, R., Choudhuri, J., Chuang, H., Cohoon, M., de Crecy-
Lagard, V., Diaz, N., Disz, T., Edwards, R., Fonstein, M., Frank, E., Gerdes, S., Glass,
E., Goesmann, A., Hanson, A., Iwata-Reuyl, D., Jensen, R., Jamshidi, N., Krause, L.,
Kubal, M., Larsen, N., Linke, B., McHardy, A., Meyer, F., Neuweger, H., Olsen, G.,
Olson, R., Osterman, A., Portnoy, V., Pusch, G., Rodionov, D., Ruckert, C., Steiner,
J., Stevens, R., Thiele, I., Vassieva, O., Ye, Y., Zagnitko, O., and Vonstein, V. (2005).
The subsystems approach to genome annotation and its use in the project to annotate
1000 genomes. Nucleic Acids Res, 33(17):5691–702.

Palenik, B., Brahamsha, B., Larimer, F., Land, M., Hauser, L., Chain, P., Lamerdin, J.,
Regala, W., Allen, E., McCarren, J., Paulsen, I., Dufresne, A., Partensky, F., Webb,
E., and Waterbury, J. (2003). The genome of a motile marine Synechococcus. Nature,
424(6952):1037–42.

Palinska, K., Jahns, T., Rippka, R., and Tandeau De Marsac, N. (2000). Prochlorococcus
marinus strain PCC 9511, a picoplanktonic cyanobacterium, synthesizes the smallest
urease. Microbiology, 146 Pt 12:3099–107.

Partensky, F. (2006). Synechococcus sp. wh7803 et rcc307, ubiquitous marine cyanobacte-

ria. [Link]

Partensky, F., Hess, W., and Vaulot, D. (1999). Prochlorococcus, a marine photosynthetic
prokaryote of global significance. Microbiol Mol Biol Rev, 63(1):106–27.

Paul, J., Williamson, S., Long, A., Authement, R., John, D., Segall, A., Rohwer, F., An-
drolewicz, M., and Patterson, S. (2005). Complete genome sequence of phiHSIC, a pseu-
dotemperate marine phage of Listonella pelagia. Appl Environ Microbiol, 71(6):3311–20.

Pennacchio, L. and Rubin, E. (2001). Genomic strategies to identify mammalian regulatory

sequences. Nat Rev Genet, 2(2):100–9.

Perez-Rueda, E. and Collado-Vides, J. (2000). The repertoire of DNA-binding transcrip-

tional regulators in Escherichia coli K-12. Nucleic Acids Res, 28(8):1838–47.

Pichard, S. and Paul, J. (1991). Detection of Gene Expression in Genetically Engineered

Microorganisms and Natural Phytoplankton Populations in the Marine Environment by
mRNA Analysis. Appl Environ Microbiol, 57(6):1721–1727.

Proctor, L. and Fuhrman, J. (1990). Viral mortality of marine bacteria and cyanobacteria.
Nature (London), 343:60–2.

Regnier, P. and Grunberg-Manago, M. (1989). Cleavage by RNase III in the transcripts

of the met Y-nus-A-infB operon of Escherichia coli releases the tRNA and initiates the
decay of the downstream mRNA. J Mol Biol, 210(2):293–302.

Repoila, F., Majdalani, N., and Gottesman, S. (2003). Small non-coding RNAs, co-
ordinators of adaptation processes in Escherichia coli: the RpoS paradigm. Mol Micro-
biol, 48(4):855–61.
Reyes, J., Muro-Pastor, M., and Florencio, F. (1997). Transcription of glutamine syn-
thetase genes (glnA and glnN) from the cyanobacterium Synechocystis sp. strain
PCC 6803 is differently regulated in response to nitrogen availability. J Bacteriol,
179(8):2678–89.

Ripp, S. and Miller, R. (1997). The role of pseudolysogeny in bacteriophage-host interac-

tions in a natural freshwater environment. Microbiology, 143:2065–70.

Rippka, R., Deruelles, J., Waterbury, J., Herdman, M., and Stanier, R. (1979). Generic
assignments, strain histories and properties of pure cultures of cyanobacteria. J. Gen.
Microbiol., 111:1–61.

Rivas, E. and Eddy, S. (2001). Noncoding RNA gene detection using comparative sequence
analysis. BMC Bioinformatics, 2:8.

Rivas, E., Klein, R., Jones, T., and Eddy, S. (2001). Computational identification of
noncoding RNAs in E. coli by comparative genomics. Curr Biol, 11(17):1369–73.

Robertson, H., Dickson, E., and Dunn, J. (1977). A nucleotide sequence from a ribonucle-
ase III processing site in bacteriophage T7 RNA. Proc Natl Acad Sci U S A, 74(3):822–6.

Rocap, G., Larimer, F., Lamerdin, J., Malfatti, S., Chain, P., Ahlgren, N., Arellano, A.,
Coleman, M., Hauser, L., Hess, W., Johnson, Z., Land, M., Lindell, D., Post, A., Regala,
W., Shah, M., Shaw, S., Steglich, C., Sullivan, M., Ting, C., Tolonen, A., Webb, E.,
Zinser, E., and Chisholm, S. (2003). Genome divergence in two Prochlorococcus ecotypes
reflects oceanic niche differentiation. Nature, 424(6952):1042–7.

Rojo, F. (1999). Repression of transcription initiation in bacteria. J Bacteriol,

181(10):2987–91.

Rosenberg, M. and Kramer, R. (1977). Nucleotide sequence surrounding a ribonuclease

III processing site in bacteriophage T7 RNA. Proc Natl Acad Sci U S A, 74(3):984–8.

Rosenberg, M., Kramer, R., and Steitz, J. (1974). T7 early messenger RNAs are the direct
products of ribonuclease III cleavage. J Mol Biol, 89(4):777–82.

Rujan, T. and Martin, W. (2001). How many genes in Arabidopsis come from cyanobac-
teria? An estimate from 386 protein phylogenies. Trends Genet, 17(3):113–20.

Salvador, M., Klein, U., and Bogorad, L. (1998). Endogenous fluctuations of DNA topology
in the chloroplast of Chlamydomonas reinhardtii. Mol Cell Biol, 18(12):7235–42.

Sambrook, J. and Russell, D. (2001). Molecular Cloning: A Laboratory Manual, Third

Edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Schneider, T. and Stephens, R. (1990). Sequence logos: a new way to display consensus
sequences. Nucleic Acids Res, 18(20):6097–100.

Schopf, J. (1993). Microfossils of the Early Archean Apex chert: new evidence of the
antiquity of life. Science, 260:640–6.
Schyns, G., Jia, L., Coursin, T., Tandeau de Marsac, N., and Houmard, J. (1998). Pro-
moter recognition by a cyanobacterial RNA polymerase: in vitro studies with the
Calothrix sp. PCC 7601 transcriptional factors RcaA and RcaD. Plant Mol Biol,
36(5):649–59.

Shen-Orr, S., Milo, R., Mangan, S., and Alon, U. (2002). Network motifs in the transcrip-
tional regulation network of Escherichia coli. Nat Genet, 31(1):64–8.

Sledjeski, D., Gupta, A., and Gottesman, S. (1996). The small RNA, DsrA, is essential
for the low temperature expression of RpoS during exponential growth in Escherichia
coli. EMBO J, 15(15):3993–4000.

Steglich, C. (2003). Biochemical, spectroscopic and moleculargenetic characterisation of

novel phycoerythrin species from Prochlorococcus sp. PhD thesis, Humboldt University
Berlin.

Su, Z., Dam, P., Chen, X., Olman, V., Jiang, T., Palenik, B., and Xu, Y. (2003). Com-
putational inference of regulatory pathways in microbes: an application to phosphorus
assimilation pathways in Synechococcus sp. WH8102. Genome Inform Ser Workshop
Genome Inform, 14:3–13.

Su, Z., Olman, V., Mao, F., and Xu, Y. (2005). Comparative genomics analysis of NtcA
regulons in cyanobacteria: regulation of nitrogen assimilation and its coupling to pho-
tosynthesis. Nucleic Acids Res, 33(16):5156–71.

Sullivan, M., Coleman, M., Weigele, P., Rohwer, F., and Chisholm, S. (2005). Three
Prochlorococcus Cyanophage Genomes: Signature Features and Ecological Interpreta-
tions. PLoS Biol, 3(5):e144.

Sullivan, M., Waterbury, J., and Chisholm, S. (2003). Cyanophages infecting the oceanic
cyanobacterium Prochlorococcus. Nature, 424(6952):1047–51.

Sun, G., Birkey, S., and Hulett, F. (1996). Three two-component signal-transduction
systems interact for Pho regulation in Bacillus subtilis. Mol Microbiol, 19(5):941–8.

Suttle, C., Chan, A., and Cottrell, M. (1990). Infection of viruses by phytoplankton and
reduction of primary productivity. Nature (London), 347:467–9.

Suzuki, S., Ferjani, A., Suzuki, I., and Murata, N. (2004). The SphS-SphR two compo-
nent system is the exclusive sensor for the induction of gene expression in response to
phosphate limitation in synechocystis. J Biol Chem, 279(13):13234–40.

Tatusov, R., Fedorova, N., Jackson, J., Jacobs, A., Kiryutin, B., Koonin, E., Krylov,
D., Mazumder, R., Mekhedov, S., Nikolskaya, A., Rao, B., Smirnov, S., Sverdlov, A.,
Vasudevan, S., Wolf, Y., Yin, J., and Natale, D. (2003). The COG database: an updated
version includes eukaryotes. BMC Bioinformatics, 4:41.

Tatusov, R., Koonin, E., and Lipman, D. (1997). A genomic perspective on protein
families. Science, 278(5338):631–7.
Tegmark, K., Morfeldt, E., and Arvidson, S. (1998). Regulation of agr-dependent virulence
genes in Staphylococcus aureus by RNAIII from coagulase-negative staphylococci. J
Bacteriol, 180(12):3181–6.

Tettelin, H., Masignani, V., Cieslewicz, M., Donati, C., Medini, D., Ward, N., Angiuoli,
S., Crabtree, J., Jones, A., Durkin, A., Deboy, R., Davidsen, T., Mora, M., Scarselli, M.,
Margarit y Ros, I., Peterson, J., Hauser, C., Sundaram, J., Nelson, W., Madupu, R.,
Brinkac, L., Dodson, R., Rosovitz, M., Sullivan, S., Daugherty, S., Haft, D., Selengut,
J., Gwinn, M., Zhou, L., Zafar, N., Khouri, H., Radune, D., Dimitrov, G., Watkins, K.,
O’Connor, K., Smith, S., Utterback, T., White, O., Rubens, C., Grandi, G., Madoff,
L., Kasper, D., Telford, J., Wessels, M., Rappuoli, R., and Fraser, C. (2005). Genome
analysis of multiple pathogenic isolates of Streptococcus agalactiae: implications for the
microbial "pan-genome". Proc Natl Acad Sci U S A, 102(39):13950–5.

Thiel, T. (1988). Phosphate transport and arsenate resistance in the cyanobacterium

Anabaena variabilis. J Bacteriol, 170(3):1143–7.

Thompson, J., Higgins, D., and Gibson, T. (1994). CLUSTAL W: improving the sensitivity
of progressive multiple sequence alignment through sequence weighting, position-specific
gap penalties and weight matrix choice. Nucleic Acids Res, 22(22):4673–80.

Ting, C., Rocap, G., King, J., and Chisholm, S. (2002). Cyanobacterial photosynthesis in
the oceans: the origins and significance of divergent light-harvesting strategies. Trends
Microbiol, 10(3):134–42.

Tjaden, B., Saxena, R., Stolyar, S., Haynor, D., Kolker, E., and Rosenow, C. (2002). Tran-
scriptome analysis of Escherichia coli using high-density oligonucleotide probe arrays.
Nucleic Acids Res, 30(17):3732–8.

Tomitani, A., Okada, K., Miyashita, H., Matthijs, H., Ohno, T., and Tanaka, A. (1999).
Chlorophyll b and phycobilins in the common ancestor of cyanobacteria and chloroplasts.
Nature, 400(6740):159–62.

Trotochaud, A. and Wassarman, K. (2004). 6S RNA function enhances long-term cell

survival. J Bacteriol, 186(15):4978–85.

Trotochaud, A. and Wassarman, K. (2005). A highly conserved 6S RNA structure is

required for regulation of transcription. Nat Struct Mol Biol, 12(4):313–9.

Urbach, E., Robertson, D., and Chisholm, S. (1992). Multiple evolutionary origins of
prochlorophytes within the cyanobacterial radiation. Nature, 355(6357):267–70.

Ureta-Vidal, A., Ettwiller, L., and Birney, E. (2003). Comparative genomics: genome-wide
analysis in metazoan eukaryotes. Nat Rev Genet, 4(4):251–62.

Valentin-Hansen, P., Eriksen, M., and Udesen, C. (2004). The bacterial Sm-like protein
Hfq: a key player in RNA transactions. Mol Microbiol, 51(6):1525–33.

Vazquez-Bermudez, M., Flores, E., and Herrero, A. (2002). Analysis of binding sites
for the nitrogen-control transcription factor NtcA in the promoters of Synechococcus
nitrogen-regulated genes. Biochim Biophys Acta, 1578(1-3):95–8.
Venter, J., Remington, K., Heidelberg, J., Halpern, A., Rusch, D., Eisen, J., Wu, D.,
Paulsen, I., Nelson, K., Nelson, W., Fouts, D., Levy, S., Knap, A., Lomas, M., Nealson,
K., White, O., Peterson, J., Hoffman, J., Parsons, R., Baden-Tillson, H., Pfannkoch,
C., Rogers, Y., and Smith, H. (2004). Environmental genome shotgun sequencing of the
Sargasso Sea. Science, 304(5667):66–74.

Verhamme, D., Arents, J., Postma, P., Crielaard, W., and Hellingwerf, K. (2002). Inves-
tigation of in vivo cross-talk between key two-component systems of Escherichia coli.
Microbiology, 148(Pt 1):69–78.

Vinnemeier, J., Kunert, A., and Hagemann, M. (1998). Transcriptional analysis of the
isiAB operon in salt-stressed cells of the cyanobacterium Synechocystis sp. PCC 6803.
FEMS Microbiol Lett, 169(2):323–30.

Vogel, J., Argaman, L., Wagner, E., and Altuvia, S. (2004). The small RNA IstR inhibits
synthesis of an SOS-induced toxic peptide. Curr Biol, 14(24):2271–6.

Vogel, J., Axmann, I., Herzel, H., and Hess, W. (2003a). Experimental and computational
analysis of transcriptional start sites in the cyanobacterium Prochlorococcus MED4.
Nucleic Acids Res, 31(11):2890–9.

Vogel, J., Bartels, V., Tang, T., Churakov, G., Slagter-Jager, J., Huttenhofer, A., and
Wagner, E. (2003b). RNomics in Escherichia coli detects new sRNA species and indicates
parallel transcriptional output in bacteria. Nucleic Acids Res, 31(22):6435–43.

Vogel, J. and Sharma, C. (2005). How to find small non-coding RNAs in bacteria. Biol
Chem, 386(12):1219–38.

Wagner, E. and Simons, R. (1994). Antisense RNA control in bacteria, phages, and
plasmids. Annu Rev Microbiol, 48:713–42.

Walker, G. (1984). Mutagenesis and inducible responses to deoxyribonucleic acid damage

in Escherichia coli. Microbiol Rev, 48(1):60–93.

Washietl, S. and Hofacker, I. (2004). Consensus folding of aligned sequences as a new

measure for the detection of functional RNAs by comparative genomics. J Mol Biol,
342(1):19–30.

Washietl, S., Hofacker, I., and Stadler, P. (2005). Fast and reliable prediction of noncoding
RNAs. Proc Natl Acad Sci U S A, 102(7):2454–9.

Wassarman, K., Repoila, F., Rosenow, C., Storz, G., and Gottesman, S. (2001). Identifi-
cation of novel small RNAs using comparative genomics and microarrays. Genes Dev,
15(13):1637–51.

Wassarman, K. and Storz, G. (2000). 6S RNA regulates E. coli RNA polymerase activity.
Cell, 101(6):613–23.

Wasserman, W. and Sandelin, A. (2004). Applied bioinformatics for the identification of

regulatory elements. Nat Rev Genet, 5(4):276–87.
Watanabe, T., Sugiura, M., and Sugita, M. (1997). A novel small stable RNA, 6Sa RNA,
from the cyanobacterium Synechococcus sp. strain PCC6301. FEBS Lett, 416(3):302–6.

Waterbury, J., Watson, S., Guillard, R., and Brand, L. (1979). Widespread occurrence of
a unicellular, marine, planktonic, cyanobacterium. Nature, 277(5694):293–4.

Waterbury, J., Watson, S., Valois, F., and Franks, D. (1986). Biologial and ecological
characterization of the marine unicellular cyanobacterium synechococcus. T. Platt and
W. Li (eds.), Photosynthetic Picoplankton. Can. J. Fish. Aquat. Sci. Bull., 214:71–120.

Waterbury, J. and Willey, J. (1988). Isolation and growth of marine planktonic cyanobac-
teria. Methods in Enzymology, 167:100–5.

Webb, R., Reddy, K., and Sherman, L. (1990). Regulation and sequence of the Syne-
chococcus sp. strain PCC 7942 groESL operon, encoding a cyanobacterial chaperonin.
J Bacteriol, 172(9):5079–88.

Weilbacher, T., Suzuki, K., Dubey, A., Wang, X., Gudapaty, S., Morozov, I., Baker, C.,
Georgellis, D., Babitzke, P., and Romeo, T. (2003). A novel sRNA component of the
carbon storage regulatory system of Escherichia coli. Mol Microbiol, 48(3):657–70.

West, N. and Scanlan, D. (1999). Niche-partitioning of Prochlorococcus populations in a

stratified water column in the eastern North Atlantic Ocean. Appl Environ Microbiol,
65(6):2585–91.

Wheeler, D., Barrett, T., Benson, D., Bryant, S., Canese, K., Church, D., DiCuccio,
M., Edgar, R., Federhen, S., Helmberg, W., Kenton, D., Khovayko, O., Lipman, D.,
Madden, T., Maglott, D., Ostell, J., Pontius, J., Pruitt, K., Schuler, G., Schriml, L.,
Sequeira, E., Sherry, S., Sirotkin, K., Starchenko, G., Suzek, T., Tatusov, R., Tatusova,
T., Wagner, L., and Yaschenko, E. (2005). Database resources of the National Center
for Biotechnology Information. Nucleic Acids Res, 33(Database issue):D39–45.

Wilderman, P., Sowa, N., FitzGerald, D., FitzGerald, P., Gottesman, S., Ochsner, U.,
and Vasil, M. (2004). Identification of tandem duplicate regulatory small RNAs in
Pseudomonas aeruginosa involved in iron homeostasis. Proc Natl Acad Sci U S A,
101(26):9792–7.

Williams, K. (2002). Integration sites for genetic elements in prokaryotic tRNA and
tmRNA genes: sublocation preference of integrase subfamilies. Nucleic Acids Res,
30(4):866–75.

Wilson, W., Carr, N., and Mann, N. (1996). The effect of phosphate status on the kinet-
ics of cyanophage infection of the oceanic cyanobacterium synechococcus sp. wh7803.
Journal of Phycology, 32:506–16.

Winkler, W., Nahvi, A., and Breaker, R. (2002). Thiamine derivatives bind messenger
RNAs directly to regulate bacterial gene expression. Nature, 419(6910):952–6.

Xie, W., Jager, K., and Potts, M. (1989). Cyanobacterial RNA polymerase genes rpoC1
and rpoC2 correspond to rpoC of Escherichia coli. J Bacteriol, 171(4):1967–73.
Xu, Y., Mori, T., and Johnson, C. (2003). Cyanobacterial circadian clockwork: roles of
KaiA, KaiB and the kaiBC promoter in regulating KaiC. EMBO J, 22(9):2117–26.

Yan, B., Methe, B., Lovley, D., and Krushkal, J. (2004). Computational prediction of
conserved operons and phylogenetic footprinting of transcription regulatory elements in
the metal-reducing bacterial family Geobacteraceae. J Theor Biol, 230(1):133–44.

Young, M. and Kay, S. (2001). Time zones: a comparative genetics of circadian clocks.
Nat Rev Genet, 2(9):702–15.

Yousef, N., Pistorius, E., and Michel, K. (2003). Comparative analysis of idiA and isiA
transcription under iron starvation and oxidative stress in Synechococcus elongatus PCC
7942 wild-type and selected mutants. Arch Microbiol, 180(6):471–83.

Zago, M., Dennis, P., and Omer, A. (2005). The expanding world of small RNAs in the
hyperthermophilic archaeon Sulfolobus solfataricus. Mol Microbiol, 55(6):1812–28.

Zengel, J. and Lindahl, L. (1994). Diverse mechanisms for regulating ribosomal protein
synthesis in Escherichia coli. Prog Nucleic Acid Res Mol Biol, 47:331–70.

Zhang, A., Altuvia, S., Tiwari, A., Argaman, L., Hengge-Aronis, R., and Storz, G. (1998).
The OxyS regulatory RNA represses rpoS translation and binds the Hfq (HF-I) protein.
EMBO J, 17(20):6061–8.

Zhang, A., Wassarman, K., Ortega, J., Steven, A., and Storz, G. (2002). The Sm-like Hfq
protein increases OxyS RNA interaction with target mRNAs. Mol Cell, 9(1):11–22.

Zhang, A., Wassarman, K., Rosenow, C., Tjaden, B., Storz, G., and Gottesman, S. (2003).
Global analysis of small RNA and mRNA targets of Hfq. Mol Microbiol, 50(4):1111–24.

Zhang, K. and Nicholson, A. (1997). Regulation of ribonuclease III processing by double-

helical sequence antideterminants. Proc Natl Acad Sci U S A, 94(25):13437–41.

Zhang, X. and Studier, F. (1997). Mechanism of inhibition of bacteriophage T7 RNA

polymerase by T7 lysozyme. J Mol Biol, 269(1):10–27.

Zhang, X. and Studier, F. (2004). Multiple roles of T7 RNA polymerase and T7 lysozyme
during bacteriophage T7 infection. J Mol Biol, 340(4):707–30.

Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction.
Nucleic Acids Res, 31(13):3406–15.
Appendix

matrix 1 2 ArsR 3

4 5 6

7 8 9 10

11 LexA 12 NtcA 13

14 15 16 17

20 21
18 19

Figure 2: Sequence logos of the best conserved elements obtained by phylogenetic footprinting analysis
of Prochlorococcus Med4, MIT 9313, SS120 and Synechococcus WH 8102. Based on these 21 clusters,
position-specific weight matrices (PSWM) were constructed to search for candidate regulatory elements
in upstream regions of all four genomes. Three predicted clusters bear analogy to already known sites for
certain cyanobacteria: ArsR (matrix 2), LexA (matrix 11) and NtcA (matrix 12). The focus on spaced
and palindromic DNA motifs revealed two additional best clusters (matrix 8 and 18).
 Med4 MIT 9313 SS120 WH 8102 function
PMM1697 PMT2246 Pro1863 SYNW2496 Type II alternative RNA polymerase sigma factor
PMM1637 PMT0118 Pro1799 SYNW0091 cyanobacteria-specific GntR-like HTH domain containing
transcriptional regulator
PMM1629 PMT0127 Pro1791 SYNW0102 Type II alternative RNA polymerase sigma factor
PMM1619 PMT0146 Pro1780 SYNW0126 Two-component response regulator
PMM1494 PMT1528 Pro1649 SYNW0597 Translation initiation factor 2
PMM1490 PMT1512 Pro1645 SYNW0608 Trypsin-like serine protease
PMM1463 PMT1481 Pro1616 SYNW0462 Nitrogen regulatory protein PII
PMM1444 PMT1459 Pro1597 SYNW0504 Membrane associated GTPase
PMM1341 PMT1417 Pro1422 SYNW0551 Sensory protein (HAMP domain e: 9.5e-17)
PMM1289 PMT0346 Pro1363 SYNW1621 Type II alternative RNA polymerase sigma factor
PMM1281 PMT0355 Pro1355 SYNW1613 GTPase, Era homolog
PMM1262 PMT0380 Pro1336 SYNW1582 SOS-response transcriptional repressor LexA
PMM1138 PMT1155 Pro1232 SYNW0704 Predicted GTPase, probable translation factor
PMM1113 PMT1097 Pro1083 SYNW0904 Two-component response regulator
PMM1030 PMT2208 Pro1502 SYNW2401 Fe2+/Zn2+ uptake regulation protein
PMM0916 PMT0597 Pro1033 SYNW1078 Membrane associated GTPase
PMM0762 PMT0575 Pro0834 SYNW1109 Predicted membrane GTPase
PMM0714 PMT1001 Pro1578 SYNW0798 Transcriptional regulator ArsR-like
PMM0637 PMT0858 Pro0794 SYNW0865 Fe2+/Zn2+ uptake regulation protein
PMM0577 PMT0456 Pro0580 SYNW1509 Type II alternative RNA polymerase sigma factor
PMM0573 PMT0464 Pro0575 SYNW1518 AbrB family transciptional regulator
PMM0515 PMT1252 Pro0515 SYNW1763 Phosphoglycerate mutase
PMM0496 PMT1272 Pro0495 SYNW1783 Putative principal RNA polymerase sigma factor
(sigA,rpoD)
PMM0482 PMT0908 Pro0481 SYNW1909 Membrane protease subunits
PMM0443 PMT1344 Pro0439 SYNW1864 AbrB family transciptional regulator
PMM0391 PMT0203 Pro0388 SYNW0657 Predicted GTPase
PMM0262 PMT1852 Pro0294 SYNW0254 DNA-binding protein, stimulates sugar fermentation
PMM0246 PMT1831 Pro0277 SYNW0275 Global nitrogen regulator NtcA
PMM0189 PMT2074 Pro0214 SYNW2322 Predicted GTPase
PMM0169 PMT2043 Pro0194 SYNW2289 Two-component response regulator
PMM0155 PMT2015 Pro0179 SYNW2269 Uncharacterized conserved protein (CYTH domain e:
7.6e-05)
PMM0154 PMT2014 Pro0177 SYNW2268 Transcriptional regulator, contains HTH motif (luxR fam-
ily e: 3.2e-21)
PMM0152 PMT2012 Pro0175 SYNW2266 Nucleoside-diphosphate-sugar transferase
PMM0147 PMT2007 Pro0170 SYNW2260 RuBisCO operon transcriptional regulator
PMM0134 PMT1994 Pro0156 SYNW2246 Two-component response regulator (cztR_silR_copR:
heavy metal response e: 1e-61)
PMM0128 PMT1988 Pro0150 SYNW2236 Two-component response regulator
PMM0090 PMT1636 Pro0106 SYNW2176 Periplasmic trypsin-like serine protease
PMM0080 PMT1623 Pro0094 SYNW2163 Predicted transcriptional regulator
PMM0077 PMT1618 Pro0091 SYNW2158 ATPase

Table 1: A core set of 35 regulatory factor and 5 σ factor genes conserved between Prochlorococcus
Med4, MIT 9313, SS120 and Synechococcus WH 8102 was revealed by 4-way BLASTp comparison with a
cutoff E-value of 10−5 and comparison to The SEED database (Overbeek et al., 2005), TIGRFAMs/Pfams
"Regulatory functions" (Fraser, 2006). Putative functions were assigned from the SS120 genome annotation
(and from TIGRFAM hits to SS120).
downstream gene position score hit sequence
1. motif; matrix 2 (cutoff score: 15) ArsR-like
PMM0010 49 17.71367 ATAAATCAATAATTTTTTAT
PMM0214 71 15.59813 ATTTTTTAACAAATTTTGAT
PMM0215 176 16.56511 ATCAAAATTTGTTAAAAAAT
PMM0335 293 15.14369 TTAAATTTATTTTAATCTGA
PMM0359 216 17.32203 ATAAAAAATTTCTGATTTAT
PMM0414 24 16.06529 ATAGATTAAGTAATTTTTAA
PMM0465 28 15.35478 CTATAATAAGAAAAATTGAT
PMM0482 247 16.65583 AGAAATCGAGAAATTTTGTT
PMM0506 54 15.74686 TTCAATAATTTTGAATATAT
PMM0519 24 15.53282 ATAAAAATATATTGTTATGT
PMM0619 65 17.48895 ATACATTATGATTTTTTGAA
PMM0622 83 18.6638 TTCAAAATTTTTTAATTTAA
PMM0623 35 18.39738 TTAAATTAAAAAATTTTGAA
PMM0705 49 15.88768 TTACAACAAATAAATTTGAT
PMM0709 143 17.45089 TTAAAAATATGTCAATCAAT
PMM0710 235 17.95111 ATTGATTGACATATTTTTAA
PMM0713 49 27.79706 ATACATCAAAATATTTTGAT
PMM0713 45 27.79287 ATCAAAATATTTTGATGTAT
PMM0714 31 27.69488 ATACATCAAAATATTTTGAT
PMM0714 27 27.69031 ATCAAAATATTTTGATGTAT
PMM0717 137 19.09974 ATAAATTAAAAAATATTTAA
PMM0718 64 18.21427 TTAAATATTTTTTAATTTAT
PMM0726 123 15.0388 TTAAATTAAGAAATTTGTAT
PMM0947 58 16.95098 ATTAATCAAAATTACTTGAA
PMM0975 250 15.05821 ATAAATTAAATTTGATGAAT
PMM1077 62 19.38194 ATATAAATATCTTAATGTGT
PMM1078 64 19.37619 ACACATTAAGATATTTATAT
PMM1118 252 15.16239 ATACCTGAAAAAATTTTTTT
PMM1155 94 15.41322 ATTTTTTAAGATATTTTTAT
PMM1197 156 17.5173 NTAAATCAAAATTAATTTAT
medRNA_44 24 16.58527 ATCAAAAATTGCTAATATAT
PMM1310 41 16.66728 ATATATTAGCAATTTTTGAT
PMM1404 277 16.39646 TTATCTTAAAAAATTTTTAA
PMM1413 59 15.92726 ATCAATTATTTTTGATTGAT
PMM1440 77 15.73466 AGCAAGAATTTTTGATCAAA
PMM1511 72 15.38256 ATCAAAAATGCTTTAGATAT
PMM1512 142 15.81907 ATATCTAAAGCATTTTTGAT
PMM1588 125 18.52823 ATAAAATTTTTTTGAGAAAT
PMM1707 291 17.16661 TTAAAAAAATCTTTATTTAT
PMM1708 20 15.24595 ATAAATAAAGATTTTTTTAA
PMT0282 292 18.04943 TGCTCAATTTCTTGATGTAT
PMT0283 236 17.97373 ATACATCAAGAAATTGAGCA
PMT0304 273 20.0033 ATACCTCGAGCAATTTTCAT
PMT0305 26 19.55711 ATGAAAATTGCTCGAGGTAT
PMT0993 94 24.75818 ATATATCAAGATTTCTTGAT
PMT0993 90 21.98635 ATCAAGATTTCTTGATGGAT
PMT1000 28 29.12228 ATACATCAAGATATTTTGAT
PMT1000 24 18.5389 ATCAAGATATTTTGATTGAT
PMT1001 98 18.83245 ATCAATCAAAATATCTTGAT
PMT1001 94 29.41211 ATCAAAATATCTTGATGTAT
PMT1045 138 16.77599 ATCAAGATATTCTGATAAAT
PMT1435 112 15.73631 ATACGTCAATAGATTTTGAT
Pro0259 129 20.8185 ATAGAGCAACAAAATTTGAT
Pro0260 142 20.83304 ATCAAATTTTGTTGCTCTAT
Pro0263 40 21.583 TTCAATATATTTTGATGAAT
Pro0264 38 21.58681 ATTCATCAAAATATATTGAA
Pro0378 41 15.42825 ATAAAATAAAAAAACATGAT
Pro0510 293 15.61229 ATAAATTATTCTTGTTAAGT
Pro1002 162 15.09725 ATAAATCCAGAGATTTTCTT
Pro1379 76 15.47173 ATATCACAAGCAAAAGTGAT
Pro1494 152 16.89803 AAAAAAATATTTTAATTTGT
Pro1559 29 15.89025 ATAGATTAAACAATCTTTTT
Pro1572 79 16.47627 ATACATAAAGAGTATTTTAT
Pro1573 95 16.51243 ATAAAATACTCTTTATGTAT
Pro1574 57 15.76892 TTATATCAAGTATTATTGAT
Pro1575 72 26.76199 ATACATCAAGAAAACTTGAT
Pro1577 55 18.7366 ATATATCAATAATTCTTGAT
Pro1577 51 24.30857 ATCAATAATTCTTGATATAT
Pro1578 88 24.5432 ATATATCAAGAATTATTGAT
Pro1578 84 18.92793 ATCAAGAATTATTGATATAT
Pro1642 74 15.39444 TTTGAACAAGATATTTTTTT
Pro1643 79 15.45753 AAAAAAATATCTTGTTCAAA
Pro1667 252 15.07919 AGAAATCAAGCTTTTTAGAT
Pro1847 110 16.17289 ATTAAACAAGAATATTTGTT
Pro1847 103 16.19721 AAGAATATTTGTTGATCTAA
ss1RNA_44 239 16.96463 TTAGATCAACAAATATTCTT
ss1RNA_44 232 16.98786 AACAAATATTCTTGTTTAAT
SYNW0798 46 23.51881 ATCAAATTATGTTGATCTAA
SYNW0799 37 23.47682 TTAGATCAACATAATTTGAT
wh8RNA_40 163 15.6728 ATAGATAGAGAAATCTAGAT
wh8RNA_53 163 15.6728 ATAGATAGAGAAATCTAGAT
2. motif; matrix 8 (cutoff score: 10)
PMM0194 33 10.8534 GGTTCATACGTACAACTTGT
PMM0195 212 11.11291 ACAAGTTGTACGTATGAACC
PMM0344 146 12.45796 ATAAATTTTGGTTATCAGTA
PMM0548 273 12.41028 TACTGATAAGTCAAGCATTT
PMM0548 221 21.16172 CATTGATAAGTTAAACTTAT
PMM0548 216 27.54579 ATAAGTTAAACTTATCACCC
PMM0549 145 27.32375 GGGTGATAAGTTTAACTTAT
PMM0549 140 20.74804 ATAAGTTTAACTTATCAATG
PMM0549 88 11.84022 AAATGCTTGACTTATCAGTA
PMM1352 131 12.25501 ATAAGTTTTTCTAATGTAAC
PMM1378 39 16.12585 CTAAGCTATGCTAATCAAAG
PMM1485 40 10.43569 TGGTAATTAGCAAATCTTAC
PMT0400 221 10.04086 GACTCATCAATATGACTTAT
PMT0423 216 14.99887 AAAAGCTTGGCTAATCATTG
PMT0544 169 12.43199 ATAAGACTTACTAATCCATG
PMT0847 37 12.03551 TAGTGATAACTTCGACTAAT
PMT0870 49 10.00709 ATCAATCAGACTTATCATTC
PMT1206 130 30.34871 GGGTGATTAGCCGAGCTTAT
PMT1207 226 30.29095 ATAAGCTCGGCTAATCACCC
PMT1250 23 12.16054 GATTTATTAGAGTGGCTTAT
PMT1387 230 12.99374 GGCTGATTAGCTGGAATTTT
PMT1429 146 20.31483 ATAAGCCAGAGTTATCAGTA
PMT2077 270 11.5046 CTCTGATTAATCTAGCTTGT
Pro0056 251 13.15664 GATTCATAAGAAAAATTTAT
Pro0057 208 13.03005 ATAAATTTTTCTTATGAATC
Pro0146 21 12.22789 CGGTGAAAAACTAATCTTAT
Pro0296 71 11.10765 TGGCGTTAAGCCAAGCTTTT
Pro0431 69 10.8438 TAGTCATAACTAAGGCATAT
Pro0432 139 11.0982 ATATGCCTTAGTTATGACTA
Pro0549 252 21.06009 CGCTGATAAGCTGTGCTTAT
Pro0549 247 29.62511 ATAAGCTGTGCTTATCACCC
Pro0550 132 29.59745 GGGTGATAAGCACAGCTTAT
Pro0550 127 21.0197 ATAAGCACAGCTTATCAGCG
Pro0715 92 11.81753 ATAATCCAGGCTTATCTTTA
Pro0718 28 10.32374 GTTTGATTAATATGGCTTAA
Pro1146 58 13.87713 ATAAGCTCTTGGTATCCCCC
Pro1434 159 21.33937 CATTGATAAGTCAATCTTAT
Pro1443 130 10.0577 GCGTAATAAAACAAACTTAT
Pro1492 245 12.43014 TAGTAATTTGTATAGCTTAT
Pro1527 174 19.2668 GACAGATCAGCCTGGCTTAT
Pro1527 169 10.41246 ATCAGCCTGGCTTATGCATG
Pro1528 121 10.47057 CATGCATAAGCCAGGCTGAT
Pro1528 116 19.3178 ATAAGCCAGGCTGATCTGTC
Pro1579 157 10.19982 NTGAGATAAATAGTGCTTAT
SYNW0403 47 14.68371 AGAAGTTGGTCTAATCTGCC
SYNW0404 49 14.667 GGCAGATTAGACCAACTTCT
SYNW0535 150 12.09009 AAGCAATAAGCCAGGCTTAT
SYNW0535 145 25.45269 ATAAGCCAGGCTTATCAGTG
SYNW0610 51 10.27696 TGGTGATCAGTCCGGGCTAT
SYNW0611 64 10.20396 ATAGCCCGGACTGATCACCA
SYNW0983 150 14.29245 AGCTGATAAGCGGAGCCTAT
SYNW1179 119 10.67813 GGGTTATAAGCAAAAAATAT
SYNW1719 129 31.11245 GGGTGATAAGCCGGGCTTAT
SYNW1719 124 17.86779 ATAAGCCGGGCTTATCAGGT
SYNW1720 246 17.79714 ACCTGATAAGCCCGGCTTAT
SYNW1720 241 30.94115 ATAAGCCCGGCTTATCACCC
SYNW2224 295 10.77148 ATAAGCTCAGGATCTCAACA
3. motif; matrix 11 (cutoff score: 10) LexA
PMM0273 22 13.01708 AAATATACATGGACTATTAG
PMM0334 268 10.01359 CTAGTCGTACTGATGTCCTT
PMM0334 36 21.64735 TTTATAGTATATTTGTACTA
PMM0334 32 19.23182 TAGTATATTTGTACTATCAA
PMM0337 290 20.5413 CTAATAATACATCTGTACTA
PMM0337 286 19.82232 TAATACATCTGTACTAATAA
PMM0337 266 12.27045 TAGTACAAAAATACCGTTGC
PMM0383 196 11.87095 TTCATAATACAAATATACTT
PMM0779 58 15.15134 TTAAGAGTACTTTTTTATTA
PMM0819 27 17.56949 TTAGTAGTACACATGTATTA
PMM0819 23 17.09537 TAGTACACATGTATTACTAA
PMM0936 60 17.83723 TATATAGTATATATGTACTA
PMM0936 56 23.47661 TAGTATATATGTACTATTAA
PMM1134 87 14.10629 TAGTAAATGAGTACTTTAAC
PMM1134 28 14.74932 CTAATAGTATAAATGTACCA
PMM1134 24 21.2969 TAGTATAAATGTACCATTAA
PMM1388 123 14.36985 GTATTAGTATTGATGTATTG
PMM1389 25 13.90304 CAATACATCAATACTAATAC
PMM1427 27 14.49208 ATATTAGTACACTTGTACTA
PMM1427 23 13.12711 TAGTACACTTGTACTAACTA
PMM1562 100 15.85654 GCTAAAGTACACATGTACTA
PMM1562 96 23.38556 AAGTACACATGTACTATGAA
PMT0090 41 12.07384 CAGTACAGCTGTAGCTGTAA
PMT0118 160 11.56402 CAGTATGCAAATTCTCTAAA
PMT0221 49 10.90696 CACTACTTACGTACTACAAC
PMT0222 113 11.09087 GTTGTAGTACGTAAGTAGTG
PMT0302 169 20.85696 CAATACACCTGTACTAGTCC
PMT0303 22 20.92371 GGACTAGTACAGGTGTATTG
PMT0635 25 19.57433 CTGCTGGAACGGGTGTACTA
PMT0637 38 13.3055 TAGTACAAAGATACTAGTGC
PMT0638 57 13.26128 GCACTAGTATCTTTGTACTA
PMT0834 34 10.71573 GCTGTAGAACATGAGTACTA
PMT0841 34 15.01391 TTGATAGAACATAAGTACTG
PMT1009 31 20.32052 GGGCTAGTACAGGTGTACTT
mitRNA_37 22 18.19112 GTGATGGTTCTGCTGTACTA
PMT1398 133 18.20555 TAGTACAGCAGAACCATCAC
PMT1661 88 11.01718 TAATTAGTTCATATGTACTA
PMT1661 84 20.0101 TAGTTCATATGTACTATTCC
PMT1727 150 25.96999 GTTAGGGTACGTCTGTACTA
PMT2115 23 17.68274 GAGCTAGTACAGGTGTACTT
PMT2116 282 14.68024 NGTCAAGTACACCTGTACTA
PMT2116 278 17.69666 AAGTACACCTGTACTAGCTC
PMT2209 274 10.09333 CAGTTCATCCATTCTCTCAA
Pro0031 35 13.26159 NNNNGAGTACAAATGTATTA
Pro0623 114 22.90965 GTAATAGTATAAATGTACTA
Pro0623 110 13.86032 TAGTATAAATGTACTACTGC
Pro0624 84 13.77464 GCAGTAGTACATTTATACTA
Pro0624 80 22.65136 TAGTACATTTATACTATTAC
Pro0685 33 12.48103 AGACTAGTATACTTGTACTA
Pro0685 29 18.67888 TAGTATACTTGTACTACTAC
Pro0686 179 19.28547 GTAGTAGTACAAGTATACTA
Pro0686 175 13.00231 TAGTACAAGTATACTAGTCT
Pro0703 249 14.60367 TAACTAGTACTTAAGTACTA
Pro0703 245 13.77225 TAGTACTTAAGTACTAGCCG
Pro0704 38 13.50878 CGGCTAGTACTTAAGTACTA
Pro0704 34 13.72053 TAGTACTTAAGTACTAGTTA
Pro0709 52 10.03042 TTAATGATATAAGTTTACTA
Pro0710 121 10.47232 TAGTAAACTTATATCATTAA
Pro0760 37 25.8413 GTTATGGTACATTTGTACTA
Pro0760 33 10.96035 TGGTACATTTGTACTATCTG
Pro0914 184 12.59865 TTTATAATATAATTGTATTA
Pro0915 23 11.11338 TAATACAATTATATTATAAA
Pro1231 97 11.73546 TTTCTAGAATATCTATATTG
Pro1247 38 11.73832 TAGTTTGTTTGTACTATTAG
Pro1336 48 10.7095 CTTTAGGTACATATGTATTG
Pro1484 131 11.05655 TCTATAGTACATAGTTACTA
Pro1484 127 12.01076 TAGTACATAGTTACTATAGA
Pro1485 48 11.60802 TCTATAGTAACTATGTACTA
Pro1485 44 10.67737 TAGTAACTATGTACTATAGA
Pro1716 108 26.65132 GTTAGCGTACGTGTGTACTA
Pro1839 27 11.30375 TAGTAGCGCTGTTCTCTTAA
Pro1840 119 11.46455 TTAAGAGAACAGCGCTACTA
SYNW0037 20 14.18618 GAACTGATACAGGTGTACTG
SYNW1044 40 20.90725 CAATACATCAGTACTATTAA
SYNW1045 71 21.07099 TTAATAGTACTGATGTATTG
SYNW1138 31 22.37538 GTACTAGTACTGGTGTATTG
SYNW1140 32 24.14263 GTACTAGTACTTGCGTACTA
SYNW1140 28 10.41996 TAGTACTTGCGTACTAGCGG
SYNW1405 21 12.84859 GAACCAGTATGGCTGTACTA
SYNW1466 20 11.12337 GCTTCAGTACTGATGTACTA
SYNW1467 39 11.76793 NNNNTAGTACATCAGTACTG
SYNW1467 35 11.17171 TAGTACATCAGTACTGAAGC
SYNW1660 192 11.04585 TAATACACCTGTACCAACTG
SYNW2062 83 23.34863 GTTAGCGTACGCCTGTACTA
SYNW2104 80 12.71489 TGGTACAGATATATTAGTAC
SYNW2105 198 13.15952 GTACTAATATATCTGTACCA
SYNW2106 33 15.97836 TGTTTAGAACAGGTGTACTA
SYNW2106 29 15.69584 TAGAACAGGTGTACTACTAC
4. motif; matrix 12 (cutoff score: 12) NtcA
PMM0120 56 13.85507 ATAATTACAATTGATACAAA
PMM0121 107 14.08338 TTTGTATCAATTGTAATTAT
PMM0245 159 26.91616 CTTGTATCAACAGTAACATT
PMM0246 65 26.58152 AATGTTACTGTTGATACAAG
PMM0378 128 12.93099 AACCTCTTTTTTGATACAAA
PMM0404 211 13.27921 GTGGTTACTTTTGATATCAA
PMM0417 181 16.56816 AATGTAGCAATAGCTACTTT
PMM0919 167 23.79687 TATGTATCAAAGGTAACTTT
PMM0920 67 23.27283 AAAGTTACCTTTGATACATA
PMM0970 63 21.58554 AATGTTACCTATGCTACAAA
PMM1118 131 12.97309 TATGTATCAAATATGTCTTT
PMM1164 98 12.30187 TATGTATTAATTGATACAAG
PMM1461 132 24.79925 ATTGTTATCATTGATACAAA
PMM1462 67 24.38745 TTTGTATCAATGATAACAAT
PMT0509 26 16.7513 ATGGTTGGCGTTGATACCTG
PMT0580 255 12.72427 GATGTGCTTTCTGCTACATC
PMT0601 81 23.84146 AAGGTACCTGTTGCTACAAA
PMT1023 41 12.13741 TAATTAGCAATGGTCACTTT
PMT1061 119 13.60201 ATTGTTACTATTGAAACCAC
PMT1213 147 13.57789 ATTGGGATCTTTGATAAAAA
PMT1244 268 13.1531 TCTGTAGCAACAAAAACCGA
PMT1276 141 16.31724 GATGTGGCTTTTGCTACCTG
PMT1277 28 16.46187 CAGGTAGCAAAAGCCACATC
PMT1480 47 15.52831 TCGGTAACAACAGCCACAAC
PMT1579 172 12.45182 TATGGATGAATAAGCACAGT
PMT1758 72 14.00264 AATGTGCCTGATGCTCCCTG
PMT1831 138 19.22064 ACCGTCACCATTGCTACATG
PMT1853 100 14.32158 TTCGTATCAACATGAACATA
PMT1979 101 13.61964 ATTGAAACCATTGATACATC
PMT1980 278 14.1815 GATGTATCAATGGTTTCAAT
PMT2229 97 12.04587 TTTGTATCATTCACTACAGA
PMT2246 176 19.46608 AAAGTACTCTTTGCTACCTA
Pro0080 120 12.18497 AATGTTTTCAATAATACAAA
Pro0088 157 13.48113 TAAGTACCCTTTTATACCAA
Pro0276 20 22.82149 CTTGTATCAAAAATGACTTT
Pro0277 65 23.04317 AAAGTCATTTTTGATACAAG
Pro0369 279 12.95705 TTGTGATCAAAGATCACATT
Pro0415 99 14.90975 TTTGTAGCGATTGCTACAAA
Pro0416 220 15.23282 TTTGTAGCAATCGCTACAAA
Pro0525 71 12.40095 NTGGTCTCAATGGTTACAGC
Pro0807 230 17.61154 TTGGTATAAATACTTACTTT
Pro0874 270 12.60331 TATGTATAAAAGATAAGAAT
Pro0884 142 12.56437 TTCGTATCATTCGTCACTAT
Pro1037 187 24.75488 TTTGTATCAATAAGTACTTT
Pro1038 71 24.04508 AAAGTACTTATTGATACAAA
Pro1209 280 12.66658 TCTGTATTAACAGATACATA
Pro1262 189 16.20907 TATGTAAGAAAAATCACTTT
Pro1270 61 13.10047 TATGTCTCATCAGTAGCATT
Pro1276 72 12.03402 CAAGTATCAAAAATTACAAG
Pro1466 29 16.23659 TTTGTATCAAATGTCACATG
Pro1471 44 20.03587 CTGGTATCAACACGTACATC
Pro1614 146 19.67509 TTTGTAATTGATGATACATA
Pro1615 67 19.15405 TATGTATCATCAATTACAAA
Pro1620 42 13.45334 TATGAATCACCAATAACTAT
ss1RNA_7 51 12.70874 AGCGTTAGCGTTGATATAAG
SYNW0314 80 12.40029 GATGTAATTGCTGCTCCAGG
SYNW0382 36 16.35403 TAGGTAGCAACGGTTACTGG
SYNW0388 76 15.61574 AACGTCACTCTTGATACAAT
SYNW0389 79 15.69087 ATTGTATCAAGAGTGACGTT
SYNW1073 77 25.35258 AATGTGCGCGTTGATACAAA
SYNW1074 164 25.44243 TTTGTATCAACGCGCACATT
SYNW1351 129 14.62308 TCTGTATCAGCAGCTACCAA
SYNW1352 116 14.63848 TTGGTAGCTGCTGATACAGA
SYNW1357 53 12.39165 TTCGTATCAGGACGAACAAT
SYNW1358 90 12.39574 ATTGTTCGTCCTGATACGAA
SYNW1557 217 14.28763 ACTGGCGCTGTTGTTACCAA
SYNW1564 291 12.15362 ATAGACACTATTGATAGCAA
wh8RNA_24 85 14.98478 CTGGTTTCAACGGTCACAGA
SYNW1633 93 14.97918 TCTGTGACCGTTGAAACCAG
SYNW1951 59 20.23161 TTGGTAGCAGCACTCACAGT
SYNW2255 52 15.65645 CATGTGTCAACGATGACGGT
SYNW2442 69 14.25096 TCGGTTCCGGTTGATACCAA
SYNW2450 150 12.35086 TTGGTATCAAGATGAACAAC
5. motif; matrix 18 (cutoff score: 10)
PMM0075 93 12.09851 GTAGTCAGAAGATCCTGACG
PMM0075 91 29.45427 AGTCAGAAGATCCTGACGAC
PMM1133 245 10.76102 GTCGCCTGGAACTCCTGATT
PMT0117 54 10.03596 GTTGACAGGGTCGGCCGACC
PMT0170 193 11.48746 GTAGTCATGTTGCTCTGTCT
PMT0193 44 10.67474 GTCAGCAGGAGCATCTGACA
PMT0409 37 13.23403 TTGCAGAAGATGTTGACCAC
PMT0552 176 11.2638 TGTCAGTAGATTTTGATTAA
PMT0553 167 11.21573 TTAATCAAAATCTACTGACA
PMT0618 140 10.82812 GGTCAGTTGGTGCTGTTGCT
PMT0742 237 10.35142 GATAATTTGATCCTGATGAC
PMT0863 96 10.48249 AGTCAACAGGTGCTAGCCAC
PMT0864 248 10.23127 GTGGCTAGCACCTGTTGACT
PMT1081 129 11.24873 AGCATCAGCATCAGCTCACC
PMT1144 35 11.66163 AGTCTGCAGACGCTGACCTG
PMT1220 146 12.41705 AGTCAGTGGATGGTGGGGAG
PMT1277 150 17.22112 NTGGACAAGATCTACTGCCC
PMT1294 21 10.01894 ATGGTCAGATTTTCATGACC
PMT1447 37 10.93535 AGTCAGAGCATGATCAACAC
PMT1565 217 10.83992 AGTCTGAGGACGCAGACACC
PMT1581 227 16.54442 GTGCTCAGGATCCTTTGAGT
PMT1608 97 12.90467 GCGGTCAGTAGATGCTGACC
PMT1608 95 30.54717 GGTCAGTAGATGCTGACCAC
PMT2030 170 16.55756 AGGTTCAGGATCTATTGCCC
mitRNA_25 248 16.53636 GGGCAATAGATCCTGAACCT
Pro0088 98 11.52647 GCAGTCAGTAGATCCTGACT
Pro0088 96 30.5446 AGTCAGTAGATCCTGACTAC
Pro0839 170 12.85559 AGCCAGAATAGCTTGACTAC
Pro0840 58 12.81418 GTAGTCAAGCTATTCTGGCT
SYNW0009 46 10.66969 CTGGTTAGGTTGGTCTGACT
SYNW0238 205 12.0862 GCTCGGTGGCTGCAGACGAC
SYNW0313 72 28.09206 AGGGTCAGGATCCACTGACC
SYNW0314 23 14.72306 CGGCAGTGGATGCTGAAGCT
SYNW0476 137 10.46759 GACGTCAGCAGCTCCGCACC
SYNW0477 32 11.10806 GGTGCGGAGCTGCTGACGTC
SYNW0532 70 13.06778 GTGATCGGCATTCACTGCCC
SYNW0533 52 13.20797 GGGCAGTGAATGCCGATCAC
SYNW0577 182 13.12466 GGTCAGTCCTTGCTGATCCC
SYNW0710 36 15.47353 GGAGTCAGCGTCCGCTGATT
SYNW0888 161 13.72338 GTTCACCAGAGGCTGACGAC
SYNW0947 53 10.07786 GCCGTCTGGATTGTCAGACT
SYNW1139 285 14.00371 GTGGATAGGGTCCACGGACT
SYNW1317 118 11.5402 GTTCAGAGGATCGTGATCGT
SYNW1430 235 10.76317 TTCCAGTGGATCCTGCTCAC
SYNW1441 51 10.43215 TGTCAGCACATGCAGACAAT
SYNW1470 60 11.38806 ATAGTCTTGATCCACCGATC
SYNW1562 53 10.8299 ATTGTCACGATCTGATCACC
SYNW1653 67 10.64802 ATCGTCAGGTGCGGCTGAGC
SYNW1654 29 10.70775 GCTCAGCCGCACCTGACGAT
SYNW1922 24 11.25092 AGTCAGGGGATTCAGAAGCT
SYNW1923 64 11.04157 AGCTTCTGAATCCCCTGACT
SYNW2107 140 10.06795 GTCGTGAGCAGCTCCTGGGC
SYNW2303 254 10.22033 GGTCACAAGAGCCATACCAG
SYNW2336 204 11.08036 AGGTTCAGGCTTTTCAGACC
SYNW2365 34 10.70607 AGTCCGCAGATGTTCACTCT
 SYNW2427 251 10.99942 CGATTCAGCATCGTCTGAGT
SYNW2491 33 11.50884 GTTCAGGAGTGCCGGACCCC
SYNW2492 45 11.45128 GGGGTCCGGCACTCCTGAAC

Table 2: List of predictions for putative TF binding sites in marine cyanobacteria, Prochlorococcus Med4,
MIT 9313, SS120 and Synechococcus WH 8102, calculated with the five best PSWMs (based on matrix
2 (ArsR-like), 8, 11 (LexA), 12 (NtcA) and 18 given in Tab. 2), which were obtained by phylogenetic
footprinting in these strains.
 Strain Align. Z Z rev Exp Comment
MED SS1 MIT WH8 length

3 - - - 201 -6.28 -12.94 + yfr2, yfr3, yfr4

- 1 1 1 345 -7.58 -10.18 n.t. rplCD operon leader, corresponds to E. coli S10
r-operon
1 1 1 2 756 -4.47 -9.9 n.t. rrn operon leader
2 - - - 1129 -8.15 -9.15 n.t. reciprocal coverage of 7.9%, artifact due to low-
complexity sequences
- - 1 1 161 -7.98 -4.90 n.t. highly similar sequences, putative ncRNAs
- - 1 1 229 -7.38 -7.32 n.t. rplJ operon leader, corresponds to E. coli β r-
operon
- 1 - - 122 -6.27 -5.54 + yfr2
- 1 1 1 152 -5.77 -5.61 n.t. putative Cobalamin riboswitch
- - 1 1 142 -5.29 -5.28 n.t. possible bidirectional terminator of the rplKAJL
operon
- 1 1 4 397 -4.38 -4.95 n.t. no conserved position, no significant BLASTN
hit to MED4
1 1 - - 146 -0.84 -4.92 + yfr7
2 2 1 4 697 -3.26 -4.59 + yfr6 in MED4 and SS120 and a subgroup of 5’
UTR regions to annotated genes and putative
unannotated genes in all 4 strains
- - 1 1 259 -3.7 -4.53 - rpoBC operon leader, corresponds to E. coli
attenuator separating the rpl genes from rpoBC in
the rplKAJLrpoBC gene cluster
1 - - - 153 -1.63 -4.28 - Located between genes for a two-component
sensor histidine kinase and a cons. hypoth. protein
- - 1 1 336 -4.24 -3.64 n.t. region upstream of the rbcLS cluster containing
conserved promoter
- - 1 1 106 -0.67 -4.00 n.t. rpl11 operon leader, corresponds to E. coli L11 r-
operon
- - 1 1 176 -3.42 -3.97 + yfr1
- 1 1 1 197 -3.93 -2.94 n.t. putative TPP riboswitch in front of thiC

Figure 3: List of high scoring clusters revealed by RNA prediction in marine cyanobacteria (Axmann
et al., 2005). RNA elements were predicted according to the scheme shown in Materials and Methods,
Figure 2.3. The total number of sequences in each cluster and the distribution within the four compared
genomes plus the total alignment length are given. The elements are ordered according to the lowest score
in either forward (Z) or reverse (Z rev) orientation (in bold letters). The lower the Z-score the higher
the support for structural conservation. Exp (experimental testing): +, tested positively by Northern
hybridisation; NT, not tested. The cluster identities (CLID) were also used in Results, Table 3.5. For
further details and exact positions of sequences see Results, Table 3.5 and Kensche (2005); Axmann et al.
(2005).
 c_PCC 7120 : -TGAG-TTGCACTGCGACGTTGGTGACTGCCAT--CAA-GTTTTTTGATCTATGTCT-TTAAGAAAAAGGGAACC--
[Link]. : -TGAG-TTGCACTACGACGTTGGTGACTGCCAT--CAA-GTTTTTTGATCTATGCCT-TTAATGACAAGGGAGTC--
c_Gloeobac. : ATGCTCTTGGCCTGCGACGTTGGTGAATGCCAA--TAA-AGTTATCGATCTATA-CT-TTGTGCGATAGGGAGCC--
PCC 6803 : ----GAAGTAACTACCGCGTTGGTGACTGCCAA--CAACGTTTTTTGATCTATG-CT-TGAAAAATAAGGGTGTCTC
c_Tsyn.elong : ----AGTTTAGCTGCGACGTTGGTGACTGCCAA--TATTGTTTTGTGATCTACACTTGCAAAACATTTGGGAATC--
PCC 7942 : ----GATCTGGCTGCGACGTTGGTGACTGCCAA--CAATGTTTTTTGATCTATACCT-TTCGAAGTTGGGCGCTCT-
PCC 6301 : ----GATCTGGCTGCGACGTTGGTGACTGCCAA--CAATGTTTTTTGATCTATACCT-TTCGAAGTTGGGCGCTCT-
Med4 : ----------ACTGGATCTTTGGTCAAGTTCTTCTTTATGTGCTCTGATCTGACTCTTTTTCGACTTGAGGAGAC--
MIT 9312 : ----------ACTGGATCTTTGGTTAAGTTCTTCTTTATGTGCTCTGATCTGACTTTCTTTCGACGTGAGGAGAC--
SS120 : -------AGAACTGTTTTGTTGGTGATGACCTT--TTAAGTGCTCTGATCTGACTCTTTTTCGACTTGAGGAGAC--
WH 8102 : ----------ACTGCTGTGTTGGTGACGGCCTTTTCTA-GTGCTCTGATCTGACTCTCTTTCGACTTGGGGAGAC--
WH 7803 : ----------ACTACCCCGTTGGTGACAGCCTTCTTCTAGTGCTCTGATCTGACTCTCTTTCGACTTGGGGCAAC--
MIT 9313 : ----------ACTGCTGTGTTGGTGACGGCCTTTTTCTAGTGCTCTGATCTGACTCTCTTTCGATTTGGGGAGGC--

c_PCC 7120 : -AAACGTCTCTCGTGGC-AGTATACCGGGCGTGTAC---------CCGGAAACTTTAAACGAGGAAAATCGGTACCC

[Link]. : -AACTGT-TCTCGTGGC-AGTATACCGGGCATGTAC---------CCGGAAACTTTAAACGAGGAATTTAGGTTCCC
c_Gloeobac. : -AAACGCGGTCGGCGGTGAGCAAGCCGGGCATCTTCGGATGAATCTCGGAGTCTCTAAAGCCGATTCACCGGTTCCC
PCC 6803 : TGTGCATTCTGCGTGGT-AGTAGACCGGGCTCGTAC---------CCGGAAACGA-AAACGAGAATTGATGACTCCC
c_Tsyn.elong : --AATACTGTGGGCGGC-AGTAAACTGGTCTTTGAG--------CCAGAAACTTCTAAGTCTGCAGTTAGTATTCCC
PCC 7942 : --CGCAGAGCCGACGGC-AGTAGACCGAGCTTGACT----------CGGAAACTT--AATGTTGGCCGCAAGCTCCC
PCC 6301 : --CGCAGAGCCGACGGC-AGTAGACCGAGCTTGACT----------CGGAAACTT--AATGTTGGCCGCAAGCTCCC
Med4 : ----TGTCGTGATGTAGACGTAAACCAATCCTGTAA---------TTGGAAACG---AAAGCCCACTTTGACTCCTC
MIT 9312 : ----TGTCGTGATGCGGTAGTAAACCGATCTTGTAA---------TCGGAAACG---AAAGCCCACTTTGAATCCTC
SS120 : ----TGTTGTTATCGTGAAGTAAACCAGTCATGCAA---------CTGGAAATG---GATCATTACTTTGACTCCTC
WH 8102 : ----TGTCGTGACGGTGCCGTAAACCGGCCCTGTAG---------CCGGAAACG---AATCCCCACTTTGTTTCCCC
WH 7803 : ----TGTCGTGATGGTGCCGTAGACCGGCCTCGTAG---------CCGGAAACG---AAGCCTCACTTTGTCGCCCC
MIT 9313 : ----TGTCGTGAAGGTGACGTAGACCGGACTTGCTT---------CCGGAAACG---AAGCCCCACTTTGCTTCCCC

c_PCC 7120 : ACCTGGTTTAACC-AGGTCATAAACTTAGGTAAA--ACGGCGTTGCGGTGAACTCAA--------------------

[Link]. : ACCTGGATAAACCCAGGTCATAAACTTAGGTAAA--ACGGCGTCGCGGTGAACTTAA--------------------
c_Gloeobac. : ACCTGGGTTACCAGAGGTCGCATACTGAGGTAAA—-ACGGCGTTGCGGGT--CTTCA--------------------
PCC 6803 : CCCTGG-TTACACCAGGTCATAAACTGAGGTAAA--ACGGCGTGGTGGTGCTTCTC---------------------
c_Tsyn.elong : CCCTAGTTTTTACTAGGTCATAAACTGAGGTAAA--ACGGCGTCGCGGTTAGACTC---------------------
PCC 7942 : CCCTGG-TTCTCCCAGGTCGAAATCTGAGGTAAA--ACGGCGTCGCGGTTAGATCTC--------------------
PCC 6301 : CCCTGG-TTCTCCCAGGTCGAAATCTGAGGTAAA--ACGGCGTCGCGGTTAGATCTC--------------------
Med4 : TTCTGGCTTTTCC-AGGTCGATGCATCAGAGAACTGGCGGGGATTCGGTGTTACCTATCAAACTGCATGCTAATACA
MIT 9312 : TTCTGGCTTTTCC-AGGTCGATGCAGCAGAGAACTGACGGGGATCCGGTAA--------------------------
SS120 : TTCTGGCTTATCC-AGGTCGAACGCTTAGGTCAC--ACGGCAAAATGGTTC--------------------------
WH 8102 : TTCTGGTTCTTCC-AGGTCGAC-CACTGGGGCTCGCACGGCACGGTGGTTC--------------------------
WH 7803 : TTCTGGTAACTCC-AGGTCGAT-CACTGGGGCTTGCACGGCTGGGTGGTTC--------------------------
MIT 9313 : TTCTGGTTCTTCC-AGGTCGAA-CACTGGGGCTCGCACGGCATGGCGGTTC--------------------------

Figure 4: 6Sa alignment of 13 sequences from diverse cyanobacteria from Anabaena PCC 7120, Nostoc
punctiforme, Synechocystis PCC 6803, Gloeobacter violaceus PCC 7421, Thermosynechococcus elongatus,
Synechococcus PCC 7942, Synechococcus PCC 6301, Prochlorococcus MIT 9313, Synechococcus WH 7803,
WH 8102, Prochlorococcus SS120, Med4 and MIT 9312.
Abbreviations

A, G, C, T, U bases of the DNA and RNA, respectively

aa amino acid
ATCC American Tissue Culture Collection
ATP adenosintriphosphate
bp base pairs
cDNA complementary DNA
Ci Curie
DCMU 3-(3,4-dichlorophenyl)-N-N’-dimethylurea
DMSO dimethyl sulfoxide
DNA deoxyribonucelic acid
E Einstein: 6 x 1023 photons
EDTA ethylendiaminetetraacetic acid
EtOH ethanol
g constant of gravitation
h hour
HSPs high-scoring segment pairs
IGR intergenic spacer region
kb kilobase pairs
l liter
L/D light-dark
M mole
Mb megabase pairs
min minutes
MIT Massachusetts Institute of Technology
ml milliliter
mRNA messenger RNA
µ micro (10−6 )
n nano (10−9 )
NaOAc sodium acetate
ncRNA non-coding RNA
nt nucleotide
ORF open reading frame
p pico (10−12 )
PCC Pasteur Culture Collection
PCR polymerase chain reaction
pH potentia hydrogenii
RNA ribonucleic acid
rRNA ribosomal RNA
RT reverse transcription
SDS sodiumdodecylsulfate
s seconds
TEMED N,N,N’,N’-Tetramethylethylendiamine
TF transcription factor
TFBS transcription factor binding site
TIS transcriptional initiation site
tRNA transfer RNA
UTR untranslated region
UV ultraviolet
vol/vol volume per total volume
vol volume
w/v weight per volume (g/100ml)
Yfr cyanobacterial functional RNA
Acknowledgement

Herrn Prof. Thomas Börner darf ich als erstes danken für die Möglichkeit, diese Arbeit am Institut
für Biologie der HU Berlin anzufertigen. Damit auch ein ganz besonderer Dank an alle Mitarbeiter
der Genetik (vor allem an Caro, Anne R., Reimo, Tobias, Kristina, Conny, Brita, Holger und Elke)
- die super Arbeitsatmosphäre und Eure große Hilfsbereitschaft haben viele Experimente erfolg-
reich verlaufen lassen und die weniger erfolgreichen schnell vergessen gemacht.
Ein ganz großes Dankeschön an Martin - The Lord of the Sequences und aller molekularbiologi-
schen Methoden.
Annegret und Ulf habt Dank für die perfekte Zusammenarbeit und die aufregenden Monate in 2005.

Einen herzlichen Dank ebenso an Jörg - immer bereit sein unerschöpfliches Wissen über RNAs zu
teilen.

Vielen, vielen Dank an Szymon, Philip, Nils und Lutz, die nie aufgegeben haben sich meiner
Computerfragen anzunehmen; gemeinsam konnten wir einige interessante Informationen aus den
ATGC’s entschlüsseln.

Many, many thanks to the ’MARGENES clique’, we had a nice time together discussing short and
complex questions about science and everyday live as well.

Julia, lieben Dank für die Zusammenarbeit und Unterstützung beim Probelesen so vieler Seiten.

I would like to thank Debbie Lindell for the friendly cooperation and the introduction into one of
the most fascinating life forms - viruses.

Ein großes Dankeschön an Wolfgang, der wahrscheinlich niemals müde wird neue Ideen zu disku-
tieren und damit Menschen zu begeistern, stets bereit eine email zu beanworten oder auch mal
sonntags zu telefonieren. So waren die letzten drei Jahre ausgefüllt mit spannenden Experimenten
(und langen Labortagen), aber auch unterbrochen von äußerst fröhlichen workshops und meetings
in den verschiedensten europäischen Ländern.

Natürlich meiner ganzen großen Familie vielen lieben Dank für all die schönen Wochenenden in
der Lausitz und im Rheinland zwischen den anstrengenden Arbeitstagen.

Lieben Dank vor allem Iris und Anne, Nina, Jörn und Jan für eure beständige Freundschaft trotz
zahllosen Nicht-Anrufen und Absagen.
Den liebsten Dank an Sascha - du bist immer für mich da, und ich bin wegen dir hier, ILD.
Selbständigkeitserklärung

Hiermit erkläre ich, die vorliegende Arbeit selbständig ohne fremde Hilfe verfaßt und nur
die angegebene Literatur und Hilfsmittel verwendet zu haben.

Berlin, den 8. März 2006

Ilka Axmann