Molecular and Biochemical Parasitology 99 (1999) 69 – 76

Phylogenetic position of the kinetoplastids, Cryptobia bullocki,

Cryptobia catostomi, and Cryptobia salmositica and monophyly
of the genus Trypanosoma inferred from small subunit
ribosomal RNA sequences

André-Denis G. Wright a,*, Sen Lia,1, Shujuan Feng a,1, Donald S. Martin b,2,
Denis H. Lynn a
a
Department of Zoology, Uni6ersity of Guelph, Guelph, Ont., N1G 2W1, Canada
b
Department of Pathobiology, Uni6ersity of Guelph, Guelph, Ont., N1G 2W1, Canada

Received 22 September 1998; received in revised form 10 December 1998; accepted 14 December 1998

Abstract

Phylogenetic relationships within the kinetoplastid flagellates were inferred from comparisons of small-subunit
ribosomal RNA gene sequences. These included three new gene sequences from Cryptobia bullocki, (2091 bp),
Cryptobia catostomi (2090 bp), and Cryptobia salmositica (2091 bp). Trees produced using maximum parsimony and
distance-matrix methods (least squares and neighbor-joining) demonstrated with strong bootstrap support, that the
kinetoplastids are a monophyletic group divided into two major lineages consistent with the two suborders,
Trypanosomatina and Bodonina. Within the trypanosomatid clade, the genus Trypanosoma is a monophyletic group
that divides into two groups, the salivarian trypanosomes and the stercorarian trypanosomes. Dimastigella and
Rhynchobodo, currently classified in the Bodonina, are basal to the trypanosomatid-bodonid clade, suggesting that the
suborder Bodonina is paraphyletic. Further, Trypanoplasma borreli grouped within the Cryptobia clade, and was more
closely related to C. salmositica than to either C. bullocki or C. catostomi. This new molecular evidence, coupled with
morphological similarities of the two genera, again calls into question the validity of the genus Trypanoplasma.

Abbre6iations: LS, least squares method; MP, maximum parsimony; NJ, neighbor-joining method; PAUP, phylogenetic analysis
using parsimony; PCR, polymerase chain reaction; PHYLIP, phylogenetic inference package; SSrRNA, small subunit ribosomal
RNA.
* Corresponding author. Present address: CSIRO Animal Production, Private Bag, PO Wembley, WA 6014, Australia. Tel.:
+ 61-89-3336417; fax: + 61-89-3878991; e-mail: [Link]@[Link].
1
Present address: Department of Molecular Biology and Genetics, College of Biological Science, University of Guelph, Ont. N1G
2W1, Canada.
2
Present address: Ontario Ministry of Health Laboratory Services Branch, 81 Resources Rd., Etobicoke, Ont. M9P 3T1, Canada.

Note: Nucleotide sequence data reported in this paper are available in the GenBank™, EMBL, and DDJB databases under the
accession numbers: AF080224 (C. bullocki ), AF080226 (C. catostomi ), and AF080225 (C. salmositica).

0166-6851/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 6 - 6 8 5 1 ( 9 8 ) 0 0 1 8 4 - 4
70 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76

© 1999 Elsevier Science B.V. All rights reserved.

Keywords: Bodonina; Evolution; Flagellates; Protozoa; Parasites; Trypanoplasma

1. Introduction transmission by leeches) and Cryptobia for non-

haematozoic species (direct transmission and pres-
Flagellated protozoa belonging to the order ence of a contractile vacuole). However,
Kinetoplastida are a unique group of microorgan- Cryptobia salmositica and Cryptobia bullocki are
isms, distinguishable from other flagellates by the both haematozoic parasites that have been shown
presence of a DNA-containing compartment of to have an ectoparasitic phase [5,12]. Moreover,
their mitochondrion called a kinetoplast [1,2]. electron micrographs indicate the presence of a
Traditionally, the order Kinetoplastida has been contractile vacuole in the haematozoic phase of C.
divided into two suborders, the Trypanosomatina salmositica [13]. Thus, it appears the genus Try-
and the Bodonina. The Trypanosomatina consists panoplasma may not be valid and Crawley may
of the single family Trypanosomatidae, whose have been correct when he synonymized the two
members are obligate parasites and the causative genera.
agents of such human diseases as sleeping sick- For this report, we sequenced in both directions
ness, Leishmaniasis, and Chagas’ disease. The the complete SSrRNA genes from two pathogenic
Bodonina consists of two families, the Bodonidae species of Cryptobia, C. bullocki and C. salmosit-
and the Cryptobiidae, whose members are free- ica, and a nonpathogenic species, Cryptobia
living species, as well as obligate and facultative catostomi. These three new sequences, along with
parasites. previously published sequences of Tr. borreli and
Within the family Cryptobiidae, there are 52 30 other kinetoplastids, were used in molecular
species of Cryptobia [3 – 5] that infect the body phylogenetic analyses to determine (1) if the kine-
surface, gills, vascular system, or the digestive toplastids form a monophyletic group; (2) if the
tract of many species of freshwater and marine division of the kinetoplastids into two suborders
teleosts [4,6]. Moreover, at least five species of is supported by molecular data; (3) if the recently
these parasites are known to cause disease and are proposed monophyly of the genus Trypanosoma
responsible for killing commercially important by Lukes' et al. [14] is corroborated by our inde-
fish species along the west coast of North America pendent study; and (4) if the genus Cryptobia is a
[7 – 10]. These flagellates are elongate with a broad monophyletic group closely associated with Try-
anterior end and pointed posterior end, with two panoplasma, or if Trypanoplasma groups within
flagella arising from the anterior end; the anterior the Cryptobia clade.
flagellum is free while the second flagellum is
attached to the body and extends posteriorly as a
free flagellum. Their oval kinetoplast is prominent 2. Materials and methods
and is usually anterior to or beside the round
nucleus. However, because of morphological simi- 2.1. Source of samples and culture conditions
larities between Cryptobia and another fish para-
site, Trypanoplasma, controversy has existed over Cryptobia species used in this study were ob-
the very close relationships between these genera tained from the culture collection of Dr P.T.K.
for nearly 100 years, when Crawley [11] syn- Woo (Department of Zoology, University of
onymized Trypanoplasma Laveran and Mesnil, Guelph). Originally, virulent C. salmositica was
1901 with Cryptobia Leidy, 1846. Still, some isolated in 1976 from the leech vector (Piscicola
workers (e.g. Lom [3]) retained the genus Try- salmositica) found on spawning coho salmon
panoplasma for haematozoic species (indirect (Oncorhynchus kisutch Walbaum) in streams on
 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76 71

Vancouver Island, Canada and identified by Woo 2.3. Sequence a6ailability and phylogenetic analysis
[15]. The attenuated strain of C. salmositica was
developed from the virulent strain by Woo and Li The nucleotide sequences in this paper are avail-
[16] and cloned and cultured at 10°C in a medium able from the GenBank/EMBL databases under
consisting of 75% (v/v) minimal essential medium the following accession numbers: Blastocrithidia
(MEM), 25% (v/v) fetal bovine serum, and 0.025 culicis U05679, Bodo caudatus X53910, Chromulina
M HEPES buffer (pH 7.2 – 7.4) and maintained chromophila M87332, Crithidia fasciculata X03450,
since 1988. Cryptobia bullocki was isolated from Crithidia oncopelti L29264, Dimastigella trypani-
the blood of a southern flounder (Paralichthys formis X76494, Endotrypanum monterogeii
lethostigma) at Gloucester Point, VA, USA by Dr X53911, Euglena gracilis M12677, Herpetomonas
E.M. Burreson (College of William and Mary). muscarum U01015, Leishmania amazonensis
Cells of C. bullocki were cultured as above and X53912, Leishmania brasiliensis, Leishmania dono-
shown to be infective to fish in 1986 [4]. Cryptobia 6ani U42465, Leishmania tarentolae M84225, Lep-
catostomi was isolated from the blood of a tomonas sp. X53914, Mallomonas papillosa
naturally infected white sucker (Catostomus M55285, Nannochloropsis salina M87328, Phy-
commersoni ) caught in May 1993 from tomonas serpens U39577, Phytomonas sp. L35077,
Grindstone Creek, Burlington, Ontario and Rhynchobodo sp. U67183, Synura spinosa M87336,
maintained in culture and by blood inoculation in Tr. borreli L14840, Trypanosoma a6ium U39578, T.
laboratory raised suckers [17]. boissoni U39580, T. brucei M12676, T. carassii
L14841, T. congolense U22317, T. cruzi M31432, T.
2.2. Nuclear DNA extraction and sequencing e6ansi U75507, T. grayi AJ223565, T. lewisi
AJ223566, T. rotatorium U39583, T. scelopori
Approximately 1.0×108 cells of cultured para- U67182, T. simiae U22320, T. therezieni AJ223571,
sites were pelleted by centrifugation at 2.6× 103 × T. triglae U39584, T. 6i6ax U22316.
g for 20 min at 4°C and washed three times with Sequences were globally aligned using the Dedi-
Ringer’s solution. DNA extraction followed the cated Comparative Sequence Editor (DCSE) pro-
protocol of Lynn and Sogin [18] with the following gram [28]. The sequence alignment was further
exception, the lysate was laid on the top of 20% refined by eye. Hypervariable regions that could
(w/v) sucrose solution in NET buffer (0.1 M NaCl, not be unambiguously aligned were omitted from
0.1 M EDTA, and 0.01 M Tris – HCl, pH 8.0) and the matrix since the alignment for these regions
centrifuged at 5.4×104 ×g for 60 min at 4°C to remains arbitrary [29,30].
separate kinetoplast DNA from nuclear DNA. Four stramenopiles (Chromulina, Mallomonas,
Universal SSrRNA primers [19] were used in a Nannochloropsis, and Synura) were used as the
PCR amplification using a Perkin-Elmer 9600 ther- outgroups for the kinetoplastids [31,32] and
mal cycler following the protocol of Wright et al. PHYLIP (version 3.51C) [33] was used to calcu-
[20]. Cloning was performed using the Sure Clone late the sequence similarity and evolutionary dis-
Ligation Kit (Pharmacia, Baie D’Urfé, Québec) tances between pairs of nucleotide sequences
and the Flexi-Prep 100 Kit (Qiagen Inc., using the Kimura two-parameter model [34]. A
Chatsworth, CA) was used to harvest and purify distance-matrix tree was then constructed using
plasmid DNA from confirmed clones for the Fitch and Margoliash [35] least squares (LS)
sequencing. method and the neighbor-joining (NJ) method
The SSrRNA gene was sequenced in both direc- [36]. Sequence data for the maximum parsimony
tions using an ABI Prism 377 Automated DNA (MP) analysis were further reduced from 2374
Sequencer (Applied Biosystems Inc, Foster City, positions to 1106 phylogenetically informative
CA) using dye-terminator and Taq FS with five sites and the Heuristic Search of PAUP [37] was
forward and five reverse internal universal 18S used to find the most parsimonious tree. Both
primers [21] and the M13 vector forward and parsimony and distance data were bootstrap re-
reverse primers. sampled [38] 1000 times.
72 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76

3. Results monophyletic clade (100% MP, LS, and NJ), con-

sisting of Cryptobia and Trypanoplasma consistent
The lengths of the complete SSrRNA gene were with the placement of these two genera within the
as follows: C. bullocki, 2091 bp; C. catostomi, Cryptobiidae (Figs. 1 and 2). However, C.
2,090 bp; and C. salmositica 2091 bp.

3.1. Molecular phylogenies

Bootstrap data of parsimony (Fig. 1) and dis-

tance-matrix trees (Fig. 2) strongly support the
kinetoplastids as a monophyletic group 100%
(MP, LS, and NJ) of the time. They are divided
into two main lineages, consistent with their divi-
sion into the suborders Trypanosomatina (100%
MP, 100% LS, and 98% NJ) and Bodonina (74%
MP, 92% LS, and 74% NJ). However, the
bodonids Dimastigella trypaniformis and Rhyn-
chobodo did not group with the other bodonids,
but appeared basal to the split between the try-
panosomatids and bodonids.
Within the larger Trypanosomatina lineage,
bootstrap data strongly support the trypanosomes
as a monophyletic clade (100% MP, 91% LS, and
68% NJ) that is divided into two main groups, the
salivarian trypanosome clade (T. brucei, T. congo-
lense, T. e6ansi, T. simiae, and T. 6i6ax) and the
non-salivarian trypanosome clade (T. a6ium, T.
boissoni, T. carassii, T. cruzi, T. grayi, T. lewisi, T.
rotatorium, T. scelopori, T. therezieni, and T.
triglae) (Figs. 1 and 2). The trypanosomes are
also the sister group (100% LS and 98% NJ) to
the monophyletic clade of non-trypanosome gen-
era (Blastocrithidia, Crithidia, Endotrypanum,
Herpetomonas, Leishmania, Leptomonas, and Phy-
tomonas). However, the monophyly of the non-
trypanosome genera and the sister group to the
trypanosome clade was unresolved in the parsi-
mony analysis (see Fig. 1). Within this clade of
non-trypanosome genera, both Leishmania (100%
MP, LS and NJ) and Phytomonas (100% MP, LS,
and NJ) were strongly supported as monophyletic Fig. 1. A maximum parsimony tree of the kinetoplastids
groups. However, the two species of Crithidia did produced by the heuristic search of PAUP and inferred from
not form a monophyletic group, but paired with complete SSrRNA gene sequences using a bootstrap resam-
other trypanosomatids within the clade (see Figs. pling of the data set. The numbers at the forks represent the
1 and 2). percentage of times the group occurred out of 1000 trees.
Branches with less than 50% bootstrap support were collapsed.
Within the Bodonina lineage, Bodo No significance is placed on the lengths of the branches
(Bodonidae) was the sister group (74% MP, 92% connecting the species. The consistency and retention indices
LS, and 74% NJ) to a strongly supported were 0.55 and 0.70, respectively.
 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76 73

Fig. 2. Phylogenetic relationships of the kinetoplastids derived from SSrRNA evolutionary distances produced by the Kimura
two-parameter [34] correction model and constructed using the Fitch and Margoliash [35] LS method and the NJ method [36]. The
consensus trees for 1000 bootstrap resamplings of the data set were computed independently and the bootstrap resamplings of
SSrRNA gene sequences are included as a percentage at the forks. The LS bootstrap value is followed by the NJ bootstrap value.
However, when NJ bootstrap support is identical to LS bootstrap data, only one value is shown. Bootstrap values less than 50%
are indicated by an asterisk (*). Evolutionary distance is represented by the horizontal component separating species in the figure.
The scale bar corresponds to 5 changes per 100 positions.

salmositica was more closely related to Tr. borreli 4. Discussion

(66% MP, 96% LS, and 91% NJ) than to either C.
bullocki or C. catostomi, thereby making the Bootstrap data from maximum parsimony,
genus Cryptobia paraphyletic. There are only 19 neighbor-joining, and the least squares methods
bp differences between C. salmositica and Tr. strongly support the kinetoplastids and the try-
borreli, and less than 1% genetic distance amongst panosomatids as monophyletic assemblages (Figs.
these four species. 1 and 2). However, Dimastigella and Rhyncho-
74 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76

bodo, currently classified in the Bodonina, are cently published by Noyes [43]. However, in the
basal to the trypanosomatid-bodonid clade, sug- study by Lukes' et al. [14], the positions of T. cruzi
gesting that the suborder Bodonina is para- and T. a6ium are less certain; they formed an
phyletic [14,39,40]. In two recently published unresolved polychotomy in the parsimony tree
studies by Lukes' et al. [14] and Hollar et al. [39], and weakly shared a branch with the salivarian
Dimastigella and Rhynchobodo are depicted as the trypanosomes in the maximum likelihood tree.
sister group to the trypanosomatids with Tr. bor- The deep relationships within the clade of non-
reli and B. caudatus appearing basal to the try- trypanosome genera are only resolved in the least
panosomatid-bodonid clade. It is important to squares and neighbor-joining trees, but not in the
note that in their studies, there were no non-kine- parsimony tree. However, within the non-try-
toplastids, but only the bodonids, used as the panosome clade, Leishmania and the plant patho-
outgroup to the trypanosomatids. Thus, the ‘root’ gen Phytomonas formed monophyletic groups
of their tree could have been within the bodonid that are strongly supported by parsimony and
‘clade’. In contrast, we included more bodonids distance-matrix bootstrap data. Further, the
than they did and used stramenopiles as the out- genus Crithidia appears to be paraphyletic as the
group taxa so that we could test the monophyly two species of Crithidia (C. oncopelti and C. fasci-
of the kinetoplastids and the clades within it. Our culata) do not group together. Crithidia oncopelti,
results demonstrate paraphyly of the Bodonina, once placed in the genus Strigomonas, groups with
as currently conceived. Blastocrithidia [2,14,23,25–27,39,43,44], whereas
Within the larger trypanosomatid clade the try- C. fasciculata groups within the Leishmania, Lep-
panosomes are also a monophyletic group. Our tomonas and Endotrypanum clade [14,22–
monophyly of the trypanosomes corroborates the 26,39,43,44]. Because it was discovered using
recent study by Lukes' et al. [14], in which they ultrastructural and isoenzyme data that another
suggested that the paraphyly of the trypanosomes species of Crithidia, Crithidia roitmani, was
in previous studies may have been an artifact of misidentified and actually belongs to the genus
‘long branch attraction’ [41 – 43]. The long branch Herpetomonas [45], we believe that either C.
length of Euglena in our analysis did not have an oncopelti or C. fasciculata could also have been
affect on the branching order as the same tree misidentified. Although we cannot rule this possi-
topology was obtained after its removal. In our bility out, other studies involving additional spe-
analysis, we used the same 11 species of try- cies of Crithidia (C. deanei and C. desouzai )
panosomes as Lukes' et al. [14], but included T. indicate that the genus is polyphyletic [23,39,46].
e6ansi, T. lewisi, T. grayi, and T. therezieni. More- Recently, Hollar et al. [39] showed that bacterial
over, we used more non-trypanosome species and endosymbiont-containing members of the poly-
more bodonid species, and our sequences were phyletic Crithidia and Herpetomonas form a well
aligned independently of their alignment. In our defined clade, whereas the endosymbiont-free
maximum parsimony analysis, our study indicated members of these genera are found elsewhere in
higher bootstrap support (i.e. data were bootstrap the tree. They suggested that a taxonomic revision
resampled 1000 times) for our tree of 39 taxa than is needed in order to depict natural groupings
Lukes' et al. [14] reported for their tree of 20 taxa within the trypanosomatids [46,47].
which was bootstrap resampled only 100 times. Within the Bodonina, the Trypanoplasma–
We also employed two distance-matrix methods, Cryptobia clade is strongly supported (92% LS,
least squares and neighbor-joining, which also and 74% MP and NJ) as the sister group to B.
strongly supported the monophyly of the group. caudatus. In contrast, Blom et al. [40] indicated
The phylogenetic position of the stercorarian try- that another species of Bodo, B. saltans, is more
panosomes (T. cruzi and T. a6ium) differs slightly closely related to the trypanosomatids than to
between the two studies. In our analysis, T. cruzi either Tr. borreli or Cryptobia helicis.
and T. a6ium have a closer affinity to the leech- All trees indicate that Trypanoplasma and
transmitted trypanosomes, similar to the tree re- Cryptobia form a clade 100% of the time, with C.
 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76 75

salmositica more closely related to Tr. borreli References

than to either C. catostomi or C. bullocki. This
debate over the validity of the genus Try- [1] Vickerman K, Preston TM. The diversity of the kineto-
plastid flagellates. In: Lumsden WHR, Evans DA, edi-
panoplasma has continued for nearly 100 years
tors. Biology of the kinetoplastida. London: Academic
[5,12,48,49] since Crawley [11] synonymized the Press, 1976:35 – 130.
two genera. For example, recently Lukes' et al. [2] Fernandes AP, Nelson K, Beverley SM. Evolution of
[48] provided evidence that the pan-kDNA nuclear ribosomal RNAs in kinetoplastid protozoa: per-
structure in C. helicis differs from the kDNA spectives on the age and origins of parasitism. Proc Natl
Acad Sci USA 1993;90:11608– 12.
structure in Tr. borreli and argued for the
[3] Lom J. Biology of trypanosomes and trypanoplasms of
generic status of these two flagellates. However, fish. In: Lumsden WHR, Evans DA, editors. Biology of
as C. helicis is a parasite of garden snails, per- the kinetoplastids, vol. II. London: Academic Press,
haps it would have been better to compare a 1979:269 – 337.
Cryptobia species that parasitizes fish with Tr. [4] Woo PTK. Cryptobia and Cryptobiosis in fishes. Adv
Parasitol 1987;26:199 – 237.
borreli. Meanwhile, Blom et al. [40] have
[5] Woo PTK, Poynton SL. Diplomonadida, Kinetoplastida
demonstrated that trypanosomatids, including B. and Amoebida (Phylum Sarcomastigophora). In: Woo
saltans, display limited editing of the transcript PTK, editor. Fish diseases and disorders. Wallingford,
for cytochrome c oxidase subunit 2, which is UK: CAB International, 1995:27 – 96.
not edited in Tr. borreli and C. helicis. [6] Woo PTK. Flagellate parasites of fishes. In: Kreier J-T,
editor. Parasitic protozoa. London: Academic Press,
Thus, neither the tree topology, nor the num-
1994:1 – 80.
ber of nucleotide differences, nor the type of [7] Becker CD, Katz M. Infections of the haemoflagellate
RNA editing supports a generic distinctiveness Cryptobia salmositica Katz 1951 in freshwater teleosts of
for Trypanoplasma. However, Woo [6] divided the Pacific coast. Trans Am Fish Soc 1965;94:327 – 33.
the genus Cryptobia into two subgenera, the [8] Katz M, Woodey JC, Becker CD, Woo PTK, Adams JR.
Records of Cryptobia salmositica from sockeye salmon
subgenus Trypanoplasma for those haematozoic
from the Fraser River drainage and from the State of
species having an indirect life cycle and the sub- Washington. J Fish Res Board Can 1966;23:1965 – 6.
genus Cryptobia for those non-haematozoic spe- [9] Bower SM, Margolis L. Detection of infection and sus-
cies having one host. Following Woo’s [6] ceptibility of different Pacific salmon stocks
characterization of the subgenera, the four (Oncorhynchus spp.) to the haemoflagellate Cryptobia
salmositica. J Parasitol 1984;70:273 – 8.
haematozoic species in our Trypanoplasma –
[10] Bower SM, Margolis L. Distribution of Cryptobia
Cryptobia clade would all be assigned to the salmositica, a haemoflagellate of fishes in British Colum-
subgenus Trypanoplasma. Therefore, research on bia and the seasonal pattern of infection in a coastal river.
non-haematozoic Cryptobia species (e.g. C. heli- Can J Zool 1984;62:2512 – 8.
cis) is required to test whether there is a second [11] Crawley H. Studies on blood parasites, II. The priority of
Cryptobia Leidy 1846 over Trypanoplasma Laveran and
clade within the genus corresponding to Woo’s
Mesnil 1901. Bull US Bur Anim Ind 1909;119:16 – 20.
[6] subgenus Cryptobia. If monophyly of these [12] Woo PTK, Wehnert SD. Direct transmission of a
subgenera is not supported by molecular charac- haemoflagellate Cryptobia salmositica Katz, 1951 (Kineto-
teristics, then we would recommend abandoning plastida, Bodonina) between rainbow trout under labora-
the use of the subgenera, and would strongly tory conditions. J Protozool 1983;39:334 – 7.
[13] Patterson WB, Woo PTK. Electron microscopic observa-
support Crawley’s [11] decision to submerge
tions of the blood stream form of Cryptobia salmositica
Trypanoplasma. Katz 1951 (Kinetoplastida, Bodonina). J Protozool
1983;39:431 – 7.
[14] Lukes' J, Jirku M, Dolezel D, Kral’ová I, Hollar L,
Acknowledgements Maslov D. Analysis of ribosomal RNA genes suggests
that trypanosomes are monophyletic. J Mol Evol
1997;44:521 – 7.
This work was funded by a Natural Sciences [15] Woo PTK. The division process of Cryptobia salmositica
and Engineering Research Council of Canada in experimentally infected rainbow trout (Salmo gaird-
research grant awarded to Dr D.H. Lynn. neri ). Can J Zool 1978;56:1514 – 8.
76 A.-D.G. Wright et al. / Molecular and Biochemical Parasitology 99 (1999) 69–76

[16] Woo PTK, Li S. In vitro attenuation of Cryptobia [31] Watrous LE, Wheeler QD. The out-group comparison
salmositica and its use as a live vaccine against cryptobio- method of character analysis. Syst Zool 1981;30:1-11.
sis in Oncorhynchus mykiss. J Parasitol 1990;76:752–5. [32] Maddison WP, Donoghue MJ, Maddison DR. Outgroup
[17] Li S, Woo PTK. Effects of temperature and white sucker analysis and parsimony. Syst Zool 1984;33:83 – 103.
(Catostomus commersoni ) serum supplement on the in [33] Felsenstein J. PHYLIP (Phylogeny Inference Package)
vitro multiplication of Cryptobia catostomi in cell-free Documentation Files: Version 3.51 c. Seattle, WA: De-
culture medium. Parasitol Res 1996;82:276–8. partment of Genetics, University of Washington, 1993.
[18] Lynn DH, Sogin ML. Assessment of the phylogenetic [34] Kimura M. A simple method of estimating evolutionary
relationships among ciliated protists using partial riboso- rates of base substitutions through comparative studies of
mal RNA sequences derived from reverse transcripts. nucleotide sequences. J Mol Evol 1980;16:111 – 20.
BioSystems 1988;21:249–54. [35] Fitch WK, Margoliash E. Construction of phylogenetic
[19] Medlin L, Elwood HJ, Stickel S, Sogin ML. The charac- trees. Science 1967;155:279 – 84.
terization of enzymatically amplified eukaryotic 16S-like [36] Saitou N, Nei M. The neighbor-joining method: a new
rRNA-coding regions. Gene 1988;71:491–9. method for reconstructing phylogenetic trees. Mol Biol
[20] Wright A-DG, Dehority BA, Lynn DH. Phylogeny of the Evol 1987;4:406 – 25.
rumen ciliates Entodinium, Epidinium and Polyplastron [37] Swofford DL. PAUP (Phylogenetic analysis using parsi-
(Litostomatea: Entodiniomorphida) inferred from small mony) Version 3.1. Distributed by the author. Cham-
subunit ribosomal RNA sequences. J Euk Microbiol paign: Illinois Natural History Survey, 1993.
1997;44:61 – 7. [38] Felsenstein J. Confidence limits on phylogenies: An ap-
[21] Elwood HJ, Olsen GJ, Sogin ML. The small subunit proach using the bootstrap. Evolution 1985;39:783 – 91.
ribosomal RNA gene sequences from the hypotrichous [39] Hollar L, Lukes' J, Maslov DA. Monophyly of endosym-
ciliates Oxytricha no6a and Stylonychia pustulata. Mol biont containing Trypanosomatids: Phylogeny versus tax-
Biol Evol 1985;2:399–410. onomy. J Euk Microbiol 1998;45:293 – 7.
[22] Maslov DA, Avila HA, Lake JA, Simpson L. Evolution [40] Blom D, de Haan A, van den Berg M, Sloof P, Jirku M,
of RNA editing in kinetoplastid protozoa. Nature Lukes' J, Benne R. RNA editing in the free-living bodonid
1994;368:345 – 8 [published erratum appears in Nature Bodo saltans. Nucl Acids Res 1998;26:1205 – 13.
368:694]. [41] Felsenstein J. Cases in which parsimony or compatibility
[23] Du Y, Chang K-P. Phylogenetic heterogeneity of three methods will be positively misleading. Syst Zool
Crithidia spp. vs. Crithidia fasciculata. Mol Biochem Par- 1978;27:401 – 10.
asitol 1994;66:171 – 4. [42] Hendy MD, Penny D. A framework for quantitative
[24] Landweber LF, Gilbert W. Phylogenetic analysis of RNA study of evolutionary trees. Syst Zool 1989;38:297 – 309.
editing: a primitive genetic phenomenon. Proc Natl Acad [43] Noyes H. Can Trypanosoma trees be trusted? Parasitol
Sci USA 1994;91:918–21. Today 1998;14:49 – 50.
[25] Maslov DA, Lukes' J, Jirku M, Simpson L. Phylogeny of [44] Du Y, Maslov DA, Chang K-P. Monophyletic origin of
trypanosomes as inferred from the small and large sub- b-division proteobacterial endosymbionts and their co-
unit rRNAs: implications for the evolution of parasitism evolution with insect trypanosomatid protozoa Blast-
in the trypanosomatid protozoa. Mol Biochem Parasitol ocrithidia culicis and Crithidia spp. Proc Natl Acad Sci
1996;75:197 – 206. USA 1994;91:8437 – 41.
[26] Marche S, Roth C, Philippe H, Dollet M, Baltz T. [45] Silva PMFE, Solé-Cava AM, Soares MJ, Motta MCM,
Characterization and detection of plant trypanosomatids Fiorini JE, de Souza W. Herpetomonas roitami (Fiorini et
by sequence analysis of the small subunit ribosomal RNA al., 1989) N. comb.: Endosymbiont in the cytoplasm. J
gene. Mol Biochem Parasitol 1995;71:15–26. Euk Microbiol 1991;38:489 – 94.
[27] Hernandez R, Rios P, Valdes AM, Pinero D. Primary [46] Clark CG. Riboprinting: A tool for the study of genetic
structure of Trypanosoma cruzi small subunit ribosomal diversity in microorganisms. J Euk Microbiol
RNA coding region: Comparison with other trypanoso- 1997;44:277 – 83.
matids. Mol Biochem Parasitol 1990;41:207–12. [47] De Souza MA, Corte-Real S. Postnuclear kinetoplast in
[28] de Rijk P, de Wachter R. DCSE an interactive tool for choanomastigotes of Crithidia deanei Carvalho,1973. Pro-
sequence alignment and secondary structure research. posal of a new genus. Rev Inst Med Trop S Paulo
Comp Appl Biosci 1993;9:735–40. 1991;33:S8.
[29] Swofford DL, Olsen GJ. Phylogeny reconstruction. In: [48] Lukes' J, Jirku M, Aviyakulov N, Benada O. Pankineto-
Hillis DM, Moritz C, editors. Molecular systematics. plast DNA structure in a primitive bodonid flagellate,
Sunderland: Sinauer Associates, 1990:411–501. Cryptobia helicis. EMBO J 1998;17:838 – 46.
[30] Chapman RL, Buchheim MA. Ribosomal RNA gene [49] Bower SM, Margolis L. Direct transmission of the
sequences: analysis and significance in the phylogeny and haemoflagellate Cryptobia salmositica among Pacific
taxonomy of green algae. Crit Rev Plant Sci 1991;10:343– salmon (Oncorhynchus spp.). Can J Zool 1983;61:1242 –
68. 50.