See discussions, stats, and author profiles for this publication at: [Link]

net/publication/237197098

The Pines

Chapter · January 2007

DOI: 10.1007/978-3-540-34541-1_2 · Source: OAI

CITATIONS READS
18 2,594

19 authors, including:

Christophe Plomion David Chagne

French National Institute for Agriculture, Food, and Environment (INRAE) Plant & Food Research
362 PUBLICATIONS 11,691 CITATIONS 259 PUBLICATIONS 7,532 CITATIONS

SEE PROFILE SEE PROFILE

David Pot Sukul Kumar

Cirad - La recherche agronomique pour le développement Banaras Hindu University
150 PUBLICATIONS 3,122 CITATIONS 20 PUBLICATIONS 300 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Cotton repetitive DNA View project

GENETIC IMPROVEMENT OF ALNUS SP View project

All content following this page was uploaded by Charles Dana Nelson on 05 March 2014.

The user has requested enhancement of the downloaded file.

 USDA Forest Service – National Agroforestry Center

USDA Forest Service / UNL Faculty

Publications
University of Nebraska - Lincoln Year 

Pines
C. Plomion, INRA, UMR BIOGECO, 69 route d’Arcachon, 33610
Cestas, France
D. Chagné, HortResearch, Private Bag 11030, Palmerston North,
New Zealand
D. Pot, Coffee Genomic Team, CIRAD, UMR PIA 1096, 34398,
Montpellier cedex 5, France
S. Kumar, Ensis-Genetics, Ensis, Private Bag 3020, Rotorua, New
Zealand
P. L. Wilcox, Cellwall Biotechnology Centre, Scion, Private Bag
3020, Rotorua, New Zealand
R. D. Burdon, Ensis-Genetics, Ensis, Private Bag 3020, Rotorua,
New Zealand
D. Prat, Université Claude Bernard - Lyon 1, EA 3731 Génome et
Evolution des Plantes Supérieures, Bât F.A. Forel, 6 rue Raphaël
Dubois, 69622, Villeurbanne Cedex, France
D. G. Peterson, Department of Plant and Soil Sciences, 117 Dorman
Hall, Box 9555, Mississippi State University, Mississippi State, MS
39762, USA
J. Paiva, INRA, UMR BIOGECO, 69 route d’Arcachon, 33610 Ces-
tas, France
P. Chaumeil, INRA, UMR BIOGECO, 69 route d’Arcachon, 33610
Cestas, France
G. G. Vendramin, Plant Genetics Institute, National Research Coun-
cil, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Florence,
Italy
F. Sebastiani, Department of Agricultural Biotechnology, University
of Florence, Via della Lastruccia 14, 50019 Sesto Fiorentino, Flo-
rence, Italy
C. D. Nelson, USDA - Forest Service, Southern Research Station,
Southern Institute of Forest Genetics, 23332 Mississippi 67, Saucier,
MS 39574, USA
C. S. Echt, USDA - Forest Service, Southern Research Station,
Southern Institute of Forest Genetics, 23332 Mississippi 67, Saucier,
MS 39574, USA
O. Savolainen, Department of Biology, PL 3000, 90014, University
of Oulu, Finland
T. L. Kubisiak, USDA - Forest Service, Southern Research Sta-
tion, Southern Institute of Forest Genetics, 23332 Mississippi 67,
Saucier, MS 39574, USA
M. T. Cervera, CIFOR - INIA, Genética y Ecofisiologı́a Forestal
& Unidad Mixta de Genómica y Ecofisiologı́a Forestal INIA-UPM,
Departamento de Sistemas y Recursos Forestales, Carretera de La
Coruña km 7, 28040 Madrid, Spain
N. de Marı́a, CIFOR - INIA, Genética y Ecofisiologı́a Forestal &
Unidad Mixta de Genómica y Ecofisiologı́a Forestal INIA-UPM, De-
partamento de Sistemas y Recursos Forestales, Carretera de La Coruña
km 7, 28040 Madrid, Spain
M. N. Islam-Faridi, USDA-Forest Service, Southern Research Sta-
tion, Southern Institute of Forest Genetics, Forest Tree Molecular
Cytogenetics Laboratory, Texas A&M University, College Station,
TX 77843, USA

This paper is posted at DigitalCommons@University of Nebraska - Lincoln.

[Link]
CHAPTER 2

2 Pines

C. Plomion1 , D. Chagné2 , D. Pot3 , S. Kumar4 , P. L. Wilcox5 , R. D. Burdon4 , D. Prat6 , D. G. Peterson7 , J. Paiva1 ,

P. Chaumeil1 , G. G. Vendramin8 , F. Sebastiani9 , C. D. Nelson10 , C. S. Echt10 , O. Savolainen11 , T. L. Kubisiak10 ,
M. T. Cervera12 , N. de María12 , and M. N. Islam-Faridi13
1 INRA, UMR BIOGECO, 69 route d’Arcachon, 33610 Cestas, France
e-mail: plomion@[Link]
2 HortResearch, Private Bag 11030, Palmerston North, New Zealand
3 Coffee Genomic Team, CIRAD, UMR PIA 1096, 34398, Montpellier cedex 5, France
4 Ensis-Genetics, Ensis, Private Bag 3020, Rotorua, New Zealand
5 Cellwall Biotechnology Centre, Scion, Private Bag 3020, Rotorua, New Zealand
6 Université Claude Bernard - Lyon 1, EA 3731 Génome et Evolution des Plantes Supérieures, Bât F.A. Forel, 6 rue Raphaël
Dubois, 69622, Villeurbanne Cedex, France
7 Department of Plant and Soil Sciences, 117 Dorman Hall, Box 9555, Mississippi State University, Mississippi State,
MS 39762, USA
8 Plant Genetics Institute, National Research Council, Via Madonna del Piano 10, 50019 Sesto Fiorentino, Florence, Italy
9 Department of Agricultural Biotechnology, University of Florence, Via della Lastruccia 14, 50019 Sesto Fiorentino,
Florence, Italy
10 USDA - Forest Service, Southern Research Station, Southern Institute of Forest Genetics, 23332 Mississippi 67, Saucier,
MS 39574, USA
11 Department of Biology, PL 3000, 90014, University of Oulu, Finland
12 CIFOR - INIA, Genética y Ecofisiología Forestal & Unidad Mixta de Genómica y Ecofisiología Forestal INIA-UPM,
Departamento de Sistemas y Recursos Forestales, Carretera de La Coruña km 7, 28040 Madrid, Spain
13 USDA-Forest Service, Southern Research Station, Southern Institute of Forest Genetics, Forest Tree Molecular Cytogenetics
Laboratory, Texas A&M University, College Station, TX 77843, USA

2.1 (e.g., P. ponderosa, P. sylvestris). Pines are adapted to

Introduction a wide range of ecological conditions: from tropical
(e.g., P. merkusii, P. kesiya, P. tropicalis), temperate
2.1.1 (e.g., P. pungens, P. thunbergii), and subalpine (e.g.,
History of the Genus P. albicaulis, P. cembra) to boreal (e.g., P. pumila) cli-
mates (Richardson and Rundel 1998, Burdon 2002).
Origin and Distribution They can grow in quite pure stands or in mixed forest
Pinus is the most important genus within the Family with other conifers or broadleaved trees. Some species
Pinaceae and also within the gymnosperms by the are especially adapted to forest fires, e.g., P. banksiana,
number of species (109 species recognized by Farjon in which fire is virtually essential for cone opening and
2001) and by its contribution to forest ecosystems. seed dispersal. They can grow in arid conditions, on
All pine species are evergreen trees or shrubs. They alluvial plain soils, on sandy soils, on rocky soils, or
are widely distributed in the northern hemisphere, on marsh soils. Trees of some species can have a very
from tropical areas to northern areas in America and long life as in P. longaeva (more than 3,000 years).
Eurasia. Their natural range reaches the equator only
in Southeast Asia. In Africa, natural occurrences are Botanical Descriptions
confined to the Mediterranean basin. Pines grow at The genus is distinguished from other members of
various elevations from sea level (not usual in tropical the Pinaceae family by its needlelike secondary leaves,
areas) to highlands. Two main regions of diversity are borne commonly in fascicles of 1 to 8 on dwarf shoots,
recorded, the most important one in Central America with a fascicle sheath of bud scales. The leaves of pines
(43 species found in Mexico) and a secondary one in are of four types encompassing the complete plant
China. Some species have a very wide natural range development: cotyledons, juvenile leaves, scale leaves

Genome Mapping and Molecular Breeding in Plants, Volume 7

Forest Trees
C. Kole (Ed.)
© Springer-Verlag Berlin Heidelberg 2007
30 C. Plomion et al.

(cataphylls), and secondary leaves. Cotyledons vary in each consisting of a cell mass protected by an in-
number from 4 up to 24 in P. maximartinezii. There are tegument. The micropyle through which the pollen
up to three resin ducts in the cotyledons and either one tube penetrates is turned toward the axis. After pol-
or two vascular bundles (Farjon 1984). Primary leaves lination the seed cone closes its scales by expansion.
are single, generally helically arranged, and acicular, The seed cone then grows rapidly. Pollen germinates
and they are produced in most species only during the and produces the male gametophyte, with two sperm
first growth season but for a longer period in a few nuclei. Fertilization takes place later, about one year
species. Cataphylls, the nonchlorophyllous primary after pollination in temperate pine species. Seed cone
leaves produced on shoots, occur in an extension of maturation requires one growing season after fertil-
the helical arrangement of the primary leaves and sub- ization for many species, and even a third for some
tend all shoot structures, but they are typically small species. In tropical climates the cycle can be short-
and subulate or lanceolate. Secondary leaves, the nee- ened because of the lack of winter dormancy. Individ-
dles, appear by the end of the first growing season, or ual seed cones thus persist for 2, up to even 3, years
later in some species. They are the most common pine on the same tree for most pine species. Seeds contain
leaves, permanently green, and metabolically active, an embryo embedded in the remaining megagameto-
ranging in length from 2 to 50 cm (generally smaller phyte and the seed coat. The seed wings derive from
ones occur in subalpine or aridity-adapted species). tissue on the adaxial face of the seed scale. Even in
They are borne on dwarf shoots axillary to cataphylls, species with vestigial or absent wings, remnants of
in fascicles of one (P. monophylla, with circular section basal wing tissue are present on the seed scale, on the
needles) to eight needles, usual numbers per fascicle seed, or on both. Seeds are mostly wind dispersed.
being two, three, or five. A fascicle is initially bound In some species, birds are important seed-dispersal
together by a basal sheath that may then fall off or vectors (P. albicaulis). Seed cones are serotinous in
persist, but actual leaf fall involves the entire fascicle some species and open only following exposure to fire
structure. The number of leaves in the fascicles de- (P. banksiana).
termines the transverse leaf shape (Farjon and Styles
1997). Stomata are arranged in several longitudinal Systematics and Phylogeny
lines along the entire leaf length. There are gener- The genus is divided into subgenera, sections, and
ally two or more resin ducts in needles. The number subsections. Various classifications have been pro-
of vascular bundles (one or two as in cotyledons) is posed in this genus since Linnaeus. Recent ones ob-
the major trait for the identification of the main di- tain support from DNA phylogenetics to identify re-
visions of genus Pinus. The trunk is usually single, lated species. Many phylogenetic studies have been
erect, and columnar. The branches are grouped into carried out in pines. Some of the first studies in-
pseudowhorls (often called clusters), at least when volved restriction patterns of the chloroplast genome
young. Bark patterns in pines result mainly from fis- (Strauss and Doerksen 1990; Govindaraju et al. 1992;
suring due to expansion growth and to the formation Krupkin et al. 1996). More recent classifications, in-
of scales that eventually fall off. cluding a large number of species, were established
Pine species are monoecious. Pollen cones are rel- from nuclear sequences (ITS, Liston et al. 1999) and
atively small and soft; these ephemeral structures chloroplast sequences (rbcL: Gernandt et al. 2005), the
consist of an axis with many helically arranged mi- chloroplast genome being paternally inherited in the
crosporophylls. Two microsporangia are attached to genus Pinus (Neale and Sederoff 1989). Some stud-
the underside of each microsporophyll. These cones ies have also focused on subsets of the genus Pinus:
open in spring, at least for temperate-climate species, subgenus Pinus (Geada López et al. 2002), section
and release large quantities of pollen into the air. Parrya (Gernandt et al. 2003), and Eurasian species
Pollen is of the bisaccate type giving it great buoyancy (Wang et al. 1999). Comprehensive classifications of
in the air. Pine pollen can be blown over long distances. the genus Pinus were earlier established by Gaussen
Its morphology is very similar for all pine species. Seed (1960) and Van der Burgh (1973) using morpholog-
conelets are found in most species at the ends of new ical and anatomical traits. Later, Price et al. (1998)
twigs, taking the position of a lateral bud. They are and Gernandt et al. (2005) also included molecu-
usually located on the higher branches. They consist lar data and identified monophyletic subgenera, sec-
of scales, the megasporophylls, arranged around an tions, and subsections. Some features are consistent,
axis; on each megasporophyll lie two separate ovules, but variations are noticed between the classical and
 Chapter 2 Pines 31

molecular approaches. The main division into two consists of species with five-needled fascicles, thin
subgenera according to the number (one or two) of cone scales, terminal position of spines on seed cone,
leaf vascular bundles has been recognized by these and several other features absent in the subsections
authors with various subgenera names (Haploxylon Krempfianae and Gerardianae of Southeast Asia al-
and Diploxylon, sometimes called, respectively, soft ready differentiated by Van der Burgh (1973) and Price
pines and hard pines, and, more recently, Strobus et al. (1998). The three subsections of the Quinque-
and Pinus named from type species as recommended foliae section share a deciduous fascicle sheath. The
by botanical nomenclature code; [Link] section Parrya consists of the subsections Cembroides,
tree?group=Pinus&contgroup=Pinaceae). The taxo- Nelsoniae, and Balfourianae; they share an American
nomic position of a singular species with flat sec- distribution, the external position of resin ducts (as in
ondary leaves P. krempfii is not fully agreed. It has subsections Krempfianae and Gerardianae), and thick
been considered as a third monospecific subgenus cone scales (again as in subsections Krempfianae and
by Gaussen (1960), while molecular data place it as Gerardianae). Subsection Nelsoniae shows persistent
a member of the subgenus Strobus (Wang et al. 2000; fascicle sheath not found in other species of subgenus
Gernandt et al. 2005). Most species belong to the sub- Strobus. Most monophyletic groups cannot be identi-
genus Pinus. Subgenus Pinus species are character- fied from unique morphoanatomical traits.
ized by thick seed-cone scales and persistent fasci- Two of the 11 subsections consist of American
cle sheaths. The numerous sections proposed by Van species and Eurasian species. Sections Pinus and
der Burgh (1973) for this subgenus have subsequently Quinquefoliae have an Asiatic origin according to
been grouped into two sections Pinus and Trifoliae chloroplast data. The subsection Strobus lineage
(Gernandt et al. 2005), the latter being called New would have then evolved in America before coming
World diploxylon pines by Price et al. (1998). The sec- back to Eurasia. Few dispersal events to eastern
tion Trifoliae consists of American species distributed North America have probably occurred to explain the
into subsections Australes (septal, internal, or medial presence there of the limited number of species of the
needle resin ducts), Ponderosae (internal or medial subsection Pinus. The development and utilization
needle resin ducts), and Contortae (medial needle of low-copy-number nuclear genes (Syring et al.
resin ducts): most of these species are characterized 2005) should provide new insights to solve remaining
by three-needled fascicles (Table 1). classification problems. Most ancient pine fossils
Each of these subsections groups two or more have been dated to the early Cretaceous (Millar 1998).
previously described subsections. P. leiophylla and They have been found in China, North America,
P. lumholtzii are now clustered within subsection Aus- and Europe (which was very close to eastern North
trales and are not further differentiated. The section America at that time). They did not further refine the
Pinus is divided into subsections Pinus and Pinaster. putative geographic origin of genus Pinus.
Species within this section, with few exceptions, grow
in Eurasia and northern Africa. Subsection Pinaster, Hybridization
characterized by the lack of a spine on the umbo of Interspecific hybridization occurs in pines but is lim-
the cone scale, includes P. pinaster as the type. More ited mostly to related species within a subsection.
recently, all of the other species included in the sec- Some species such as P. engelmannii, P. jeffreyi, and
tion Pinaster as defined by Van den Burgh (1973) were P. ponderosa are compatible in a number of different
found to cluster within the Australes subsection of the combinations (Liston et al. 1999). Natural hybridiza-
genus (Gernandt et al. 2005). Species of the subsec- tions are often indicated by the introgression of the pa-
tion Pinus including the type species of the genus, ternally inherited chloroplast genome. Barriers occur
P. sylvestris, were previously grouped into a section at different stages, from the failure of pollen germina-
called Sylvestres by Van der Burgh (1973). The sub- tion to failure at embryogenesis (Ledig 1998). A few
genus Strobus has been divided into two sections: species are postulated to have been derived from in-
Quinquefoliae and Parrya. They differ from the sec- terspecific hybridization. They also exhibit the highly
tions Strobus and Parrya of Van der Burgh (1973) conserved chromosome number in pines (2n = 24).
and Price et al. (1998) by the transfer of the subsec- This is the situation for P. densata that has been shown
tions Krempfianae and Gerardianae from the section to combine nuclear polymorphisms of P. tabuliformis
Parrya into the section Quinquefoliae, which also in- and P. yunnanensis with the chloroplast genome of the
cludes the subsection Strobus. The subsection Strobus latter (Wang et al. 2001) and probably several other
32 C. Plomion et al.

Table 1. Systematics of the genus Pinus, according to Gernandt et al. (2005)

Subgenus Section Subsection Number Best-known species Distribution

of species

Pinus Pinus Pinus 17 P. sylvestris, P. kesiya, Eurasia,

P. merkusii North America
Pinaster 7 P. pinaster Mediterranean, Asia
Trifoliae Contortae 4 P. banksiana, P. contorta America
Australes 26 P. elliottii, P. radiata, America
P. taeda
Ponderosae 17 P. jeffreyi, P. ponderosa America
Strobus Parrya Balfourianae 3 P. balfouriana America
Cembroides 11 P. cembroides, America
P. culminicola
Nelsoniae 1 P. nelsonii Central America
Quinquefoliae Gerardianae 3 P. bungeana Asia
P. gerardiana
Krempfianae 1 P. krempfii Asia
Strobus 21 P. cembra, P. lambertiana, America, Eurasia
P. strobus

species. P. densata exchanged genes with ancestral cobs et al. 2000; Hizume et al. 2002; Liu et al. 2003; Cai
populations prior to its isolation with local differ- et al. 2006). Doudrick et al. (1995) developed a FISH-
enciation (Ma et al. 2006). Combinations of parental based karyotype for P. elliottii var. elliottii using 18S-
traits and selection for adaptation to new conditions 25S and 5S rDNA probes and CMA (chromomycin A3 )
favored colonization of new territories by the hybrid and DAPI (4’,6-diamidino-2-phenylindole) banding
species. that distinguished all 12 homologous pairs of chro-
mosomes. They went further to suggest that the pre-
sented karyotype might be useful as a standard or
2.1.2 reference karyotype for Pinus. Lubaretz et al. (1996)
Cytogenetics, DNA Content, used computer-aided chromosome analysis on the ba-
and Genome Composition sis of chromosome length, chromosome arm length
ratio, and the positions of rDNA (18S-28S and 5S)
Sax and Sax (1933), Mergen (1958), and Khoshoo and telomere (Arabidopsis-type telomere repeat se-
(1961) were the earliest to describe the karyotypes of quence or A-type TRS) detected with FISH to dis-
various conifer species. They found that species of the criminate three chromosomes of P. sylvestris. Hizume
genus Pinus were diploid with 24 chromosomes (2n et al. (2002) used four probes [45S rDNA, 5S rDNA,
= 2x = 24). The chromosomal complements generally PCSR (CMA-band specific repeat), and A-type TRS] in
consist of 10 or 11 pairs of large homobrachial (meta- developing FISH-based karyotypes for four different
centric) chromosomes and one or two pairs of smaller pine species (P. densiflora, P. thunbergii, P. sylvestris,
heterobrachial (submetacentric) chromosomes (Say- and P. nigra). Liu et al. (2003) and Cai et al. (2006) used
lor 1961, 1964, 1972, 1983). Several attempts have been FISH to establish rDNA positions in several species of
made to construct chromosome-specific karyotypes the Pinus and Strobus subgenera, respectively. Cur-
for various pine species using traditional cytogenet- rent work in P. taeda (subgenera Pinus, section Pi-
ics techniques, viz., C-banding, Giemsa, and fluores- nus, subsection Australes) (Fig. 1) emphasizes an im-
cent banding (Borzan and Papes 1978; MacPherson proved chromosome preparation technique (based on
and Filion 1981; Drewry 1982; Saylor 1983; Hizume Jewell and Islam-Faridi 1994 and Islam-Faridi and
et al. 1989, 1990). More recently, fluorescent in situ Mujeeb-Kazi 1995) and statistical analyses of chro-
hybridization (FISH) has been utilized in several pine mosome arm lengths and FISH signal positions and
species (Doudrick et al. 1995; Lubaretz et al. 1996; Ja- intensities to develop a reference karyotype and cy-
 Chapter 2 Pines 33

Fig. 1. A fluorescent in situ hybridization (FISH) photomicrograph of Pinus taeda metaphase chromosomes showing 23 of the
24 chromosomes (bar = 10 µm). The red signals detect the 18S-28S rDNA sites, and the green signals detect the Arabidopsis-type
telomere repeat sequence (A-type TRS) sites

togenetic map for use in physical genome mapping subgenus and the section. The genome of subgenus Pi-
in the subsection Australes (Islam-Faridi et al. 2003, nus is lower (54.0 pg/2C) than that of subgenus Strobus
Islam-Faridi et al. 2007). Comparison of these results (65.6 pg/2C). This variation has been related to seed
with those obtained in other subsections suggests that mass. Relationships of such variation with ecologi-
a subsection-specific karyotype may be required for cal conditions, such as drought tolerance, have also
more robust physical mapping across the entire genus been investigated (Wakamiya et al. 1996). Various en-
as a whole (Islam-Faridi et al. 2007). vironmental conditions appear to be related to DNA
Another feature of the pine genome is its large content, including latitude of range and invasiveness
physical genome size expressed in DNA content. (Grotkopp et al. 2004). An extremely large genome
Pines exhibit some of the largest DNA contents per size is common to other gymnosperms (reviewed by
diploid cell in the plant kingdom ranging from ca. Ohri and Khoshoo 1986; Murray 1998; Leitch et al.
44 pg (P. banksiana) to 75 pg (P. gerardiana) (1 pg = 2001; Grotkopp at al. 2004).
960 Mb; Arumuganathan and Earle 1991) based on At the whole-genome level, reassociation kinetics
laser flow cytometry (Grotkopp et al. 2004). For in- data (i.e., Cot analysis) indicates that 25 to 30% of
stance, the size of the maritime pine (P. pinaster) the pine genome corresponds to low- to single-copy
genome has been estimated to be between 51 and sequences, while 70 to 75% corresponds to highly re-
60 pg/2C (Chagné et al. 2002; Grotkopp at al. 2004), peated sequences (Miksche and Hotta 1973; Rake et al.
which is about seven times the size of the human 1980; Kriebel 1985; Peterson et al. 2006). The repeti-
genome (7 pg/2C; Morton 1991) and 170-fold larger tive sequences of pine have not been studied in much
than the genome of model plant Arabidopsis thaliana detail. However, it appears that the majority of repeti-
(0.3 pg/2C). The DNA amount varies according to the tive DNA consists of repeats of low sequence complex-
34 C. Plomion et al.

ity (Schmidt et al. 2000), retrotransposons (Kamm simple silviculture (monocultures and clear-felling);
et al. 1996; Kossack and Kinlaw 1999; Friesen et al. their wood is easily processed and utilized for a wide
2001), and 18S-5.8S-25S rDNA genes. As shown by variety of end uses (lumber, pulp and paper, par-
molecular, cytological, and linkage-mapping studies ticleboard, etc.); and even if species lack naturally
(Friesen et al. 2001; Scotti et al. 2005), most of these durable heartwood, treatment with preservatives is
repeat sequences are present at multiple loci and are easy.
presumably interspersed among other sequences, al- Various minor forest products come from pines
though they tend to form loose clusters that surround (Burdon 2002). Some pine plantations, especially
gene-rich islands. The remainder of the genome is P. pinaster in the Landes of southwest France, were
composed of low-copy sequences of which an un- established largely for resin production. Resin pro-
known portion are genes. If the pine genome (1C duction from pines was also a major economic activ-
∼ 25 to 30 pg; see above) is similar to Arabidopsis ity in the southeast of the USA, Mediterranean basin,
in having about 30,000 expressed genes and an aver- northern India, and east and Southeast Asia. Resin
age gene size of 2,000 bp including introns and UTRs products are still recovered from chemical pulping of
(The Arabidopsis Genome Initiative 2000), only 0.5% pines. Foliage, in the form of litterfall, has been used
of its genome is likely to be transcribed. In compari- for a range of purposes. The bark is used for a vari-
son, 54% of the Arabidopsis sequence and 5 to 6% of ety of purposes, and that of many species has a high
mammalian genomes are transcribed (Rat Genome tannin content. For a few species with large seeds, no-
Sequencing Project Consortium 2004). tably P. pinea, P. edulis, and P. sibirica, seeds have been
Gene duplication and the formation of complex a prized item of the human diet. Edible fungi, repre-
gene families have been widely cited as a potential senting the fruiting bodies of mycorrhizal symbionts,
cause of the abundance of low-copy DNA in the pine are often collected from pine stands.
genome. Kinlaw and Neale (1997) suggested that levels Pine species are also used widely for shelter and
of multiplication were greater in conifers than in other the interlinked purposes of revegetation, soil stabiliza-
plant species, and transcriptional profiling studies tion, and soil conservation, as befits their tolerance of
have noted surprising levels of transcriptome com- exposure and degraded soils in certain species. Most
plexity (Lorenz and Dean 2002). However, considering of the widely planted species of pines are used in some
the relatively low proportion of a conifer genome rep- degree for one or more of these purposes, often with
resented by expressed sequences, it is unlikely that timber production as a major bonus. Ornamental and
the evolution of multigene families alone can ex- festive use is common, with P. sylvestris and P. virgini-
plain the enormous size of pine nuclear genomes. ana being very widely grown for Christmas trees.
It is possible that a relatively large proportion of
low-copy sequences in pine are pseudogenes and/or
highly diverged repeat sequences, and indeed there 2.1.4
is evidence supporting both possibilities (Elsik and Classical Breeding Objectives
Williams 2000; Rabinowicz et al. 2005).
In most pine breeding programs, the setting of goals
was originally done subjectively, based on perceptions
2.1.3 of the main traits limiting profitability, of the vari-
Economic Importance ability and heritability of the traits concerned, and of
the genetic correlations (both favorable and adverse)
Pine species play an especially important role in among traits (Burdon 2004). The setting of breeding
modern plantation forestry worldwide and now form goals is crucial to any tree-improvement program, and
a large part of both the annual wood harvest and the yet most pine tree breeding programs do not have for-
immature plantation forests that will provide wood mally defined breeding objectives. Various reasons for
in the future (Burdon 2002). Pines enjoy such great this include complexity of the forest processing indus-
popularity because the large number of species allows try, difficulties in determining the relationships be-
choice for widely varying site and environmental con- tween selection traits and end uses, and long rotation
ditions; the volume of production of some species can ages that create uncertainty about their use (Apiolaza
be high to very high, even under unfavorable site con- and Greaves 2001). Earlier attempts of most improve-
ditions; they are well suited for reforestation and for ment programs have been to improve growth, form,
 Chapter 2 Pines 35

climatic adaptation, and resistance to certain diseases genetic gain (Wei et al. 1997; Rosvall et al. 1998; Lam-
(Cahalan 1981; Danjon 1995; Paul et al. 1997; Shel- beth 2000; Alazard 2001; Olsson et al. 2001; Plomion
bourne et al. 1997). et al. 2001; Burdon and Kumar 2004). In the species
Most recent efforts have been diverted at improv- that are easy to propagate, the concept of using clonal
ing traits related to pulp and paper and solid-wood replication of individuals within families as a means
products (Shelbourne et al. 1997; Chambers and Bor- of genetic testing is being implemented in order
ralho 1999; Greaves 1999; Lowe et al. 1999; Pot et al. to increase the efficiency of genetic improvement.
2002; Kumar 2004). Wood density and fiber mor- Significant genetic gains from such a strategy have
phology (e.g., tracheid length and coarseness) have been reported for P. radiata (Matheson and Lindgren
been reported to be crucial for pulp yield and quality. 1985; Shelbourne 1992) and P. taeda (Isik et al. 2004).
Wood properties such as wood density and stiffness Pinus patula, P. taeda, and P. elliottii, are planted
are crucial selection traits for the improvement of in South Africa in summer rainfall zones, and their
wood stiffness, while compression wood, spiral grain, average productivity is 15 m3 ha−1 yr−1 (Du Toit et al.
and microfibril angle are candidate traits for reducing 1998). Brazil, with its humid summers, largely uses
in-service instability (Shelbourne et al. 1997; Ivkovic P. caribaea, P. taeda, and P. elliottii. Pines in Brazil pro-
et al. 2006). Heartwood, resinous defects, and inter- duce 8 to 30 m3 ha−1 yr−1 on rotations of 20 to 25 years.
nal checking are also being considered as selection P. radiata is grown as large plantations in the temper-
traits to develop germplasm for appearance-grade ate southern hemisphere countries of Australia, Chile,
products (Shelbourne et al. 1997). There are some sit- New Zealand, and South Africa. The mean annual in-
uations, especially involving exotic species and dis- crement over 25 years is often 25 to 30 m3 ha−1 yr−1
ease resistance, etc., where hybrids between selective (Lamprecht 1990). However in New Zealand, growth
species are desirable to make genetic improvement rates of up to 50 m3 ha−1 yr−1 have been recorded
in the breeding-objective traits (Hyun 1976; Byun on the best sites and as low as 11 m3 ha−1 yr−1 on
et al. 1989; Blada 1994; Nikles 2000; Shelbourne 2000). very dry sites (Burdon and Miller 1992). The aver-
Pine species vary widely in their amenability to vari- age productivity of P. pinaster in southern France is
ous forms of vegetative propagation (Hartmann et al. about 10 m3 ha−1 yr−1 but could reach 20 to 25 m3
1990). Ease of propagation was generally a minor cri- ha−1 yr−1 on the best sites. Deployment of genetically
terion for species selection in breeding programs of improved loblolly pine in the USA has been reported
pines. Currently with improved technology, selection to yield up to about 21 m3 ha−1 yr−1 (McKeand et al.
is based more on silvicultural performance and wood 2003). Dhakal et al. (1996) reported a realized gain
quality. in volume of about 22% in a slash pine improvement
program in the USA. Wood-quality traits that are cur-
rently being included, in addition to growth, form,
2.1.5 and health traits, in the breeding objectives of various
Classical Breeding Achievements pines species appear to be under moderate to strong
genetic control (Burdon and Low 1992; Hannrup et al.
Various selection and deployment strategies are being 2000; Atwood et al. 2002; Gwaze et al. 2002; Pot et al.
used for different pine species. Species selection fol- 2002; Kumar 2004), and predicted genetic gains from
lowed by provenance and family-within-provenance selection appeared to be in the order of about 10%
selection, and establishment of seed orchards are for traits such as wood density and stiffness (Kumar
quite common first steps across various species 2004).
including P. radiata (Falkenhagen 1991; Matziris Clonal forestry (CLF) represents the large-scale
1995), P. oocarpa (Moura et al. 1998), P. caribaea propagation and deployment of selected clones that
(Zheng et al. 1994), P. strobus (Beaulieu et al. 1996), have been clonally tested. Deployment of tested clones
P. sylvestris (Quencez and Bastien 2001), P. pinaster by CLF is being increasingly employed with P. radiata
(Alía et al. 1995, 1997; Danjon 1995), P. taeda (Jiang in New Zealand (Sorensson and Shelbourne 2005),
et al. 1999; Lopez-Upton et al. 2000), P. tecunumanii P. taeda in the USA (Stelzer and Goldfarb 1997), and
(Hodge and Dvorak 1999), and P. contorta (Cahalan pine hybrids in Australia (Walker et al. 1996). In prin-
1981). In advanced-generation breeding programs, ciple, CLF offers additional genetic gains from cap-
forward selections and/or combined selection turing nonadditive effects, which are not captured via
(among- and within-family) are the major sources of sexual propagation, plus the benefits of greater uni-
36 C. Plomion et al.

formity and predictability in performance resulting and biogeographic history (the distributions of many
from a lack of genetic segregation. Genetic gains from species have been affected by Pleistocene glacial ad-
CLF have been predicted to be considerably higher vances), even if, in some cases, human activities also
than those from family forestry in P. radiata (Aimers- played a relevant role (e.g., P. pinaster in Portugal;
Halliday et al. 1997), P. taeda (Stelzer and Goldfarb Ribeiro et al. 2001).
1997), and P. strobus (Park 2002). Self-fertilization in pines generally occurs at a low
level (Muona and Harju 1989), and a high outcross-
ing at the mature seed stage is maintained. There is
2.1.6 evidence of selection at the embryonic stage so that
Molecular Diversity the number of inbreds is already low at the seedling
stage (Kärkkäinen and Savolainen 1993). Selection af-
The majority of studies aiming to monitor the level ter the seedling stage is still severe. This has been
and distribution of genetic diversity in Pinus natural shown by observing the survival of selfed seedlings in
populations were based on the use of neutral markers. P. sylvestris (Muona et al. 1987) and in P. leucodermis
Indeed, molecular markers, such as microsatellites, (Morgante et al. 1993).
provided useful information on historical demogra- The mode of inheritance has a major effect on the
phy and population evolution. partitioning of genetic diversity in pines, with studies
The use of isozyme electrophoresis significantly based on maternally inherited markers (transmitted
increased the amount of data on the genetic struc- by seeds only) having significantly higher GST values
ture of populations (Petit et al. 2005). These data re- than those based on paternally or biparentally inher-
vealed that pine species had high genetic diversity ited markers for pine (Burban and Petit 2003; Petit
within populations and only low levels of differen- et al. 2005). In fact, the chloroplast and mitochondrial
tiation among populations. For 28 north temperate genomes are generally paternally and maternally in-
pines, genetic differentiation (GST ) averaged 0.076 herited in pines, respectively (Petit and Vendramin
(Ledig 1998). Exceptions are represented by P. pinea 2006). In pines, GST is nearly always larger at mi-
(G.G. Vendramin et al. personal communication), Pi- tochondrial DNA markers than at chloroplast DNA
nus resinosa, a species that has a vast range across markers. On the other hand, there is no significant dif-
northeastern North America (Echt et al. 1998; Wal- ference between GST at biparentally inherited mark-
ter and Epperson 2001), and P. torreyana (Ledig and ers and at paternally inherited markers in pines (Petit
Conkle 1983; Provan et al. 1999) where near absence of et al. 2005). This is expected considering that both the
variation was observed, and by Mexican pine species cpDNA and half the nuclear genomic complement are
where higher differentiation among populations was dispersed by pollen and by seeds, i.e., they use the
observed, probably because their natural distribu- same vehicles to achieve gene flow.
tions are more highly fragmented by physiography Chloroplast and mitochondrial DNA markers al-
than those of species at more northerly latitudes. low describing phylogeographic structure in pines.
Other pine species with disjunct populations and re- While chloroplast DNA generally exhibits the high-
stricted gene flow also showed higher differentiation est diversity, phylogeographic inferences from these
among populations: for example, 16 to 27% for P. ra- markers can be blurred by extensive pollen flow. Low
diata and 22% for P. muricata (Wu et al. 1999). On population structure due to extensive pollen flow
the other hand, experimental evidence indicates that has been inferred in P. pinaster (Burban and Petit
seeds of some pines (e.g., P. palustris; Grace et al. 2004) 2003), P. sylvestris (Robledo-Arnuncio et al. 2005),
have the potential to disperse greater distances than P. canariensis (Gómez et al. 2003), and P. albicaulis
previously reported, which partly contributes to the (Richardson et al. 2002). In contrast, the mitochon-
low levels of genetic differentiation observed in these drial markers, despite their generally lower level of
species. diversity in pines (but this holds for all conifers; So-
In general, the typical distribution of the genetic ranzo et al. 2000; Gugerli et al. 2001), generally pro-
diversity within and among populations of Pinus vide a clear picture of nonoverlapping areas colonized
species is correlated with their mating system and life from different refugia (e.g., in P. pinaster; Burban and
history (pines, for example, are wind-pollinated and Petit 2003).
tend to be predominately outcrossing) (Hamrick and Neutral markers also have allowed investigating
Godt 1996; J. Duminil et al. personal communication) spatial genetic structure (SGS) in natural pine popu-
 Chapter 2 Pines 37

lations. A generally weak within-population structure basic population genetic parameters such as the pat-
has been described. P. pinaster showed a fine-scale tern of nucleotide diversity and LD for each particular
structure at the seedling stage with a patch size of ca. species and candidate gene set.
10 m that seems to be produced by restricted seed flow First estimates indicate that nucleotide diversity
(González-Martínez et al. 2002). Pines with a heavy varies considerably between plant species. Interest-
seed (differences in seed dispersion capability play ingly, the pines (e.g., P. sylvestris, Dvornyk et al. 2002;
an important role), such as P. pinaster, are expected García-Gil et al. 2003; P. taeda, Brown et al. 2004;
to have a short dispersal distance, thus producing González-Martínez et al. 2006a; P. pinaster, Pot et al.
a fine-scale structure. However, fine-scale structure 2005a) are not among the most variable species, con-
often does not persist as stands mature. For exam- tradicting expectations from the results obtained us-
ple, within-population genetic structure in Mediter- ing neutral markers and their life history character-
ranean pines may be affected by postdispersal events istics. First evidences seem to show that broadleaved
(e.g., mortality due to the severity of the Mediter- species (e.g., Populus, Ingvarsson 2005; Quercus, Ta-
ranean climate and animal-mediated secondary dis- ble 2.7 in Pot et al. 2005a) display higher levels of
persal during the summer period) that may mod- nucleotide diversity than pines.
ify the original spatial structure (González-Martínez Markers in specific functional regions of the
et al. 2002). Logging can also play a role in decreasing genome need to be statistically analyzed in order to
spatial structuring, as observed in P. strobus (Mar- test for the possibility that these regions might have
quardt and Epperson 2004), suggesting that man- experienced different selective pressures. In unstruc-
agement practices can alter natural spatial patterns, tured populations, standard neutrality tests might be
too. applied. When variation is structured in populations,
It should be stressed that the presence of fine-scale a relatively easy approach is the comparison of
structure is uncommon in Pinus species. Epperson genetic differentiation estimates, such as Wright’s
and Allard (1989), studying the spatial pattern of al- F-statistics, among markers tagging a putative gene
lozyme alleles within P. contorta ssp. latifolia stands, under selection and neutral markers or expected
found a lack of structure in the distributions of most distributions computed using coalescence theory (see
genotypes. Neutral markers by definition do not re- reviews in van Tienderen et al. 2002; Luikart et al.
flect selective processes and therefore are not used as 2003). If population divergence (FST ) is higher for
an indicator of the population adaptive potential to the gene-targeted marker with respect to divergence
a changing environment (Morin et al. 2004). Recent estimates obtained from random markers, this might
and well-established markers to detect functional ge- indicate divergent selection and local adaptation
netic variation are single nucleotide polymorphisms for the tagged gene (van Tienderen et al. 2002). Pot
(SNPs), which are particularly useful for finding genes et al. (2005a) found a high differentiation among
under selection and studying the dynamics of these populations at Pp1 (glycine-rich protein homolog)
genes in natural populations. gene in P. pinaster, higher than at the neutral level.
The availability of high-density markers, such as This result is consistent with diversifying selection
SNPs, opens the possibility of studying, by associa- acting at this locus in this species. On the other
tion genetics, the molecular basis of complex quanti- hand, the absence of differentiation observed for the
tative traits in natural populations, taking advantage gene CeA3 (cellulose synthase), compared with the
of the fact that genetic markers in close proximity to significant level observed at neutral markers may
causal polymorphisms may be in linkage disequilib- indicate balancing selection acting on this gene.
rium (LD) to them. The magnitude and distribution Recent studies on pines reveal a rapid decay in
of LD determine the choice of association mapping LD with physical distance. LD declines very rapidly
methodology. Extension and distribution of LD de- within 200 to 2,000 bp in Pinus taeda (Brown et al.
pend on many factors including population history 2004; González-Martínez et al. 2006a), Pinus sylvestris
(e.g., the presence of population bottlenecks or ad- (Dvornyk et al. 2002 García-Gil et al. 2003). A rapid
mixture) and the frequency of recombination. decay of LD in pines is consistent with what is ex-
In order to avoid false associations, the optimiza- pected from outcrossing species with large effective
tion of LD mapping requires a detailed knowledge of population size.
38 C. Plomion et al.

2.2 1992). Conkle (1981) located more loci, but still not
Construction of Genetic Maps enough to cover the pine genome. Proteins revealed by
two-dimensional polyacrylamide gel electrophoresis
In genetics, mapping is defined as the process of (2D-PAGE; O’Farrell 1975) presented the advantage
deducing schematic representations of DNA. Three of being multiplexed compared to isozymes. Impor-
types of DNA maps can be constructed depending on tantly, proteins can be easily characterized by mass
the landmarks on which they are based: spectrometry (e.g., Gion et al. 2005) and may be rec-
– Physical maps, whose highest resolution would be ognizable by sequence similarity to others proteins
the complete nucleotide sequence of the genome published in sequence databases, therefore provid-
– Genetic maps, which describe the relative positions ing functional markers expressed in the tissues ana-
of specific DNA markers along the chromosomes, lyzed. Two-dimensional protein markers were devel-
determined on the basis of how often these loci are oped only in P. pinaster (reviewed in Cánovas et al.
inherited together 2004 and Plomion et al. 2004). Although proteins pro-
– Cytogenetic maps, a visual appearance of a chro- vided physiologically relevant markers to map the
mosome when stained and examined under a mi- expressed genome, this time-consuming technique
croscope failed to provide enough markers for genetic appli-
cation, which requires full genome coverage, such as
To provide a first glimpse of the pine genome, high- linkage mapping and quantitative trait loci (QTL) de-
resolution genetic maps have been established for sev- tection.
eral pine species using different types of molecular
markers and following different strategies that are re-
viewed in the following sections. RFLPs
Restriction fragment length polymorphism (RFLP)
markers were developed in the early 1990s for P. taeda
2.2.1 (Neale and Williams 1991). They offered a sufficient
Development of Molecular Markers in Pines number of markers for high-density genome map-
ping in pine. However, this labor-intensive and time-
The construction of a linkage map relies on the avail- consuming technique was only applied to P. taeda
ability of enough molecular markers to detect link- (Devey et al. 1994) and P. radiata (Devey et al. 1996).
age between them. Each type of marker technology
has advantages and limitations. Many factors (e.g.,
polymorphism information content, level of polymor- RAPDs and AFLPs
phism exhibited for the mapping progeny, mode of In the mid-1990s PCR-based multiplex DNA finger-
inheritance, genome size) can influence the devel- printing techniques provided very powerful tools to
opment of a particular technique and the choice of generate dense linkage maps in a short period of
a marker system for a given purpose (e.g., genetic time. Random amplified polymorphic DNA (RAPD;
mapping, quantitative trait loci analysis, survey of ge- Williams et al. 1990) and then amplified fragment
netic diversity, forensic applications). The purpose of length polymorphism (AFLP; Vos et al. 1995) became
this section is to briefly review the different types of the most popular marker technologies in conifers.
molecular marker techniques that have been devel- Despite their biallelic nature and dominant mode
oped in pines and used for genetic mapping appli- of inheritance (which was actually not an issue
cations. We will not present the details of each tech- for mapping with haploid megagametophyte, see
nique. Both the review by Cervera et al. (2000a) and Sect. 2.2.2.), these markers tremendously boosted up
the references cited in Table 2 will provide the reader genetic analysis in most forest tree species including
with the necessary information for understanding the pines (reviewed in Cervera et al. 2000b).
scientific basis of each technique.
Nuclear Microsatellites
Isozymes and Proteins In contrast to other plant species, few polymorphic
The first markers developed for pine were isozymes. single-copy nuclear microsatellite markers or sim-
Linkage studies were carried out on more than 10 ple sequence repeats (SSRs) have been reported in
species for about 15 loci (reviewed by Tulsieram et al. pines (reviewed in Chagné et al. 2004). The genome
Table 2. Characteristics and applications of molecular markers

Feature Protein markers DNA markers

Hybridization-based PCR-based
Isozyme 2D-PAGE RFLP RAPD AFLP SSR SNP

Tissue quantity (g) 0.5–2 1 na na na na na

DNA quantity (mg) na na 10 0.02 0.25–0.5 0.02 0.005–0.02
DNA quality na na high Medium Medium Medium Medium
Allelism and mode Multiallelic biallelic, Biallelic Biallelic Biallelic Multiallelic Biallelic
of inheritance codominant codominant, codominant dominant dominant codominant codominant
or dominant
Number loci analysed 1–5 several tens 1-6 2–30 20–100 1–10 1–several hundred
per assay (multiplexing) (multiplexing)
Polymorphism Low Low Medium Medium Medium High Mediuma
Development cost Low Low High Low Medium High High
Cost per analysis Low Medium High Low Medium Low Low
Amenable to automation Low Low Low Medium Medium High High
Technical demand Low High High Low Medium Low Medium
Reproducibility Highb Mediumb High Low–highc High High High
Review (Müller–Starck (Plomion (Brettschneider (Rafalski 1998) (Cervera (Echt and (Syvanen 2001)
1998) et al. 2004) 1998) et al. 2000b) Burns 1999)

Applications
Certification - - - - + +++ ++
Diversity + + ++ + + +++ +++
Phylogeny ++ + ++ ++ ++ ++ +++
Mapping - - ++ ++d ++d +++ +++
Comparative mapping + + ++ - - +++ +++
QTL analysis + + ++ + + +++ +++
Association studies + + + - - ++e +++

a Sequence dependent (exon vs UTR and intron)

b Under the same environmental conditions, type of tissue, and age
c Between laboratories vs single laborytory
d To construct dense maps
Chapter 2 Pines

e Used as control loci

39
40 C. Plomion et al.

structure of these species, characterized by a large based markers (ESTP: EST polymorphisms). There
physical size, with a large amount of repeated se- are basically two groups of technologies used to de-
quence (Sect. 2.1.2), has been the main obstacle to tect nucleotide polymorphisms (SNPs and INDELs),
the development of useful markers using classical either based on the knowledge of nucleotide vari-
SSR-enriched library approaches (e.g., Auckland et al. ants or not. Up to now, techniques based on the de-
2002; Guevara et al. 2005a; C.S. Echt and C.D. Nelson, tection of differences in the DNA stability (denat-
unpublished results). In addition, the ancient diver- urating gradient gel electrophoresis, DGGE; Myers
gence time between coniferous species (Price et al. et al. 1987), conformation (single-strand conforma-
1998) and the complexity of their genomes means tion polymorphisms, SSCPs; Orita el al. 1989) under
that transferability of single-copy SSRs among genera specific polyacrylamide gel conditions, or heterodu-
and even within Pinus is generally poor, resulting in plex cleavage (TILLING, targeting induced local le-
a large proportion of amplification failure, nonspe- sions in genomes; Colbert et al. 2001) have been
cific amplification, multibanding patterns, or lack of successfully applied in pines (Plomion et al. 1999;
polymorphism (Echt and Nelson 1997; Mariette et al. Temesgen et al. 2001; Chagné et al. 2003; Ritland et al.
2001). In an attempt to circumvent these genome- 2006). With the decrease of sequencing costs and the
related problems, Elsik and Williams (2001) removed availability of pine cDNA sequences, more targeted
most of the repetitive portion of the genome us- and precise approaches are now possible (Pot et al.
ing a DNA reassociation kinetics-based method, and 2005b). The bioinformatics assembly of ESTs into
Zhou et al. (2002) targeted the low-copy portion of the large contigs (i.e., unigenes) has also made it pos-
genome using an undermethylated region enrichment sible to identify putative SNPs. Le Dantec et al. (2004)
method. Both approaches yielded remarkable enrich- identified a set of 1,400 candidate SNPs in P. pinaster
ment for useful SSR markers in P. taeda. SSRs made contigs containing between 4 and 20 sequence reads.
from low-copy, undermethylated, and total genomic This represents a great resource of molecular mark-
DNA yielded mappable markers (Nelson et al. 2003; ers for this species that can be used to map candi-
Zhou et al. 2003). P. taeda SSRs developed by Elsik and date genes, study LD, and develop comparative or-
Williams (2001) and Zhou et al. (2002) transferred thologous sequence markers for comparative genome
quite well between American hard pines (Shepherd mapping. In addition to SNPs and INDELs, a large
et al. 2002a) but were shown to be less transferable set of microsatellite markers have been developed
in the phylogenetically divergent Mediterranean hard from P. taeda and P. pinaster expressed sequence tags
pines (Chagné et al. 2004; González-Martínez et al. (ESTs) (Echt and Burns 1999; Chagné et al. 2004; Echt
2004). Interestingly, perfect trinucleotide SSRs trans- et al. 2006). These markers present the advantage of
ferred from American to Mediterranean pines better bei ng highly polymorphic and located in coding re-
than other motifs (Kutil and Williams 2001). A num- gions.
ber of nuclear SSR markers have been developed for
P. radiata, almost all of which are based on the more
frequently polymorphic dinucleotide repeat motifs 2.2.2
(Smith and Devey 1994; Fisher et al. 1996, 1998; De- Haploid- and Diploid-Based Mapping Strategies
vey et al. 2003), and used in a number of applications.
Polymorphic chloroplast microsatellite loci have also The construction of a genetic map requires two
been identified and applied (Cato and Richardson components: first, a segregating population (map-
1996; Kent and Richardson 1997). More recent SSR ping pedigree) derived from a cross between parental
discovery efforts have been undertaken in both New trees that are heterozygous for many loci and, sec-
Zealand and Australia and have been most commonly ond, a set of molecular markers segregating in the
applied to fingerprinting (Kent and Richardson 1997; progeny according to Mendelian ratios. Linkage map
Bell et al. 2004) and QTL mapping applications (Devey construction is based on the statistical analysis of
et al. 2004a). polymorphic markers in the mapping population,
considering that the distance between two loci is
EST Polymorphisms related to the probability of observing a recombi-
With the availability of sequence data obtained by ran- nation event between them. There is a number of
dom sequencing of pine cDNAs (Sect. 2.5.2), there is mapping software to facilitate automated analysis
now a clear trend toward the development of gene- ([Link]
 Chapter 2 Pines 41

Haploid- or Half-Sib-Based Mapping Strategy (i.e., intercross mating type), or 1:1:1:1, if they have
In conifers, the haploid megagametophyte constitutes different genotypes (i.e., fully informative mating
an ideal plant material for genetic mapping. This nu- type). These segregation data are then subdivided
tritive tissue surrounding the embryo is derived from into two independent data sets that separately contain
the same megaspore that gives rise to the maternal ga- the meiotic segregation data from each parent, and
mete. Therefore, it represents a single meiotic event independent maps are constructed for each parent.
in the parent tree that is genetically equivalent to A sex-average map is then constructed using an out-
a maternal gamete. The dominant nature and bial- bred mapping program, which uses fully informative
lelic mode of inheritance of RAPD and AFLP is not and intercross markers to serve as common anchor
an issue for genetic mapping with haploid megaga- points between each parental data set. Compared to
metophytes. However, quantitative traits can only be “megagametophyte progeny,” full-sibs can be grafted
measured on half-sib seedlings, limiting the detection and/or propagated by cuttings, thereby constituting
of QTLs at the first stages of tree development. Thus, a perpetual population, analogous to recombinant
this approach is not applicable to the analysis of QTLs inbred lines in crop plants. The use of such clonally
for economically important traits in well-established propagated progeny obviously increases the preci-
plantations, unless the megagametophytes were col- sion of quantitative measurements and therefore
lected and saved, which has generally not been the enhances the QTL detection power (Bradshaw and
case. Foster 1992).

Diploid- or Full-Sib-Based Mapping Strategy

Different strategies have been followed to construct 2.2.3
genetic linkage maps of Pinus: the “pseudotestcross Genetic Mapping Initiatives in Pines
strategy,” the “F2 inbred model,” and the “three-
generation outbred model.” The pseudotestcross In this section and Table 3 we summarize what has
strategy is mainly based on selection of single-copy been done in terms of linkage map construction in the
polymorphic markers heterozygous in one parent genus Pinus with emphasis made on the most studied
and homozygous null in the other parent and species. In addition, some maps have been published
therefore segregating 1:1 in their F1 progeny as together with QTL studies and will be found in the
in a testcross. Grattapaglia and Sederoff (1994) references cited in Sect. 2.3.
introduced the term two-way pseudotestcross to
define this mapping strategy, where two independent Maritime Pine
genetic linkage maps are constructed by analyzing Linkage maps of the Maritime pine (Pinus pinaster
the cosegregation of markers in each progenitor. The Ait.) genome were first constructed by analyzing the
efficiency of this strategy, as well as for the haploid cosegregation of proteins extracted from megagame-
mapping strategy, depends on finding individual tophytes collected during the germination of the em-
trees that are heterozygous for many loci, which bryo. Bahrman and Damerval (1989) were the first to
is quite easy using arbitrarily primed PCR assays report a linkage analysis for 119 protein loci using 56
(RAPD and AFLP) in highly heterozygous outcrossed megagametophytes of a single tree. Extending this ap-
tree species such as pines. The F2 inbred model is proach, Gerber et al. (1993) reported a 65-locus link-
based on a three-generation pedigree for which the age map covering one fourth of the pine genome, using
grandparents are treated as inbred lines. In the F2 18 maritime pine trees with an average of 12 megaga-
generation, three genotypes occur at any locus – AA, metophytes per tree. A more conventional pedigree
AB, and BB – segregating 1:2:1. The three-generation (inbred F2 ) was used to map 61 proteins using hap-
outbred model (Sewell et al. 1999) is an extension loid (Plomion et al. 1995a) and diploid (Plomion et al.
of the pseudotestcross strategy. Within a single 1997; Costa et al. 2000) tissues of the same seedlings.
outbred pedigree, any given codominant marker will In the latter case, protein loci were found on each chro-
segregate in one of three different ways. When one mosome (Thiellement et al. 2001). As stated above,
parent is heterozygous and the other is homozygous, the advance of PCR-based markers has allowed the
segregation will be 1:1 (i.e., testcross mating type). construction of saturated linkage maps, in a short
When both parents are heterozygous, segregation will period of time, with no prior knowledge of DNA se-
be either 1:2:1, if both parents have the same genotype quence (Plomion et al. 1995a, b; Costa et al. 2000).
Table 3. Summary of genetic linkage maps of Pinus species
42

Species Number of LGs Total number RAPD, SSR EST-based Others Total genetic Genome References
pedigree (+pairs, triplets) of linked AFLP distance (cM) saturation
marker loci

Pinus pinaster
Haploid (1 tree) Not estimated 119 – – – 119 (2D proteins) Not estimated Not estimated Bahrman and
C. Plomion et al.

Damerval 1989
Haploid (8 trees) Not estimated 65 – – – 65 (2D proteins) 530 25% Gerber et al. 1993
Haploid (H12 self) 13(+5) 263 251 – – – 1223 90% Plomion et al. 1995b
Haploid 1236 90%
(H12 open pollinated)
Haploid (H12 self) 12 463 436 – – 27 (2D proteins) 1860 100% Plomion et al. 1995a
Haploid (H12self) 11 94 94 – – – 1169 Not estimated Plomion and
O’Malley 1996
Diploid (H12 self) 11 94 94 – – – 1354 Not estimated
Haploid (H12 self) 13 398 235+127 – – 36 (2D protein) 1873 93.4% Costa et al. 2000
F1 - AFOCEL 12 759 738 14 7 – 1994 Not estimated Ritter et al. 2002
9.103.3 × 10.159.3 12 620 620 – – – 1441 Not estimated Chagné et al. 2002
(consensus)
9.103.3 × 10.159.3 12 326 276 – 50 – 1638 Not estimated Chagné et al. 2003
(consensus)
Pinus taeda
Various pedigrees Not estimated – – – – 20 isozymes Not estimated Not estimated Conkle et al. 1981
Base pedigree 20 75 – – 90 6 Not estimated Not estimated Devey et al. 1994
Haploid (Tree 10-5) 16 458 458 – – – 1727 97% Wilcox 1995
Base and QTL pedigrees 18 357 67 – 257 12 1300 76.4% Sewell et al. 1998
Tree 7-56 12 508 508 – – – 1741 100% Remington et al. 1999
Base and QTL pedigrees 20 265 253 12 – – 1281 75.4% Devey et al. 1999
Base pedigree 15 51 – 51 – – 795 46.7% Zhou et al. 2003
Pinus sylvestris
49-2 14 261 261 – – – 2638 Yazdani et al. 2003
F1 of P315 × E1101 12(+4) 179 179 – – – 1000 50% Hurme et al. 2000
F1 of AC3065 × Y3038 ♀: 12 188 188 – – – 1645 98% Yin et al. 2003
♂: 12 245 245 – – – 1681
F1 of E635 × 12(+3) 260 194 4 61 – 1314 66–85% Komulainen et al. 2003
E1101
Table 3. (continued)

Species Number of LGs Total number RAPD, SSR EST-based Others Total genetic Genome References
pedigree (+pairs, triplets) of linked AFLP distance (cM) saturation
marker loci

Pinus radiata
3-generation 22 208 41 2 165 – 1382 Not estimated Devey et al. 1996
Haploid (full-sib seed) 14 267 267 – – – 1665 93% Emebiri et al. 1998
S1 19 172 168 4 – – 1117 56% Kuang et al. 1999b
Pseudotestcross 20 235 224 11 – – 1414 85% Wilcox et al. 2001a
(Parent 850.055)
Pseudotestcross 21 194 185 9 – – 1144 77% Wilcox et al. 2001a
(Parent 850.096)
Two full-sib families 12 311 – 213 98 – 1352 Not estimated Wilcox et al. 2004
Pinus elliottii
Tree 8-7 13(+9pairs) 73 73 – – – 782 64–75% Nelson et al. 1993
18-62 × 8-7 17(+12pairs) 129 129 – – – 1146 Not estimated Kubisiak et al. 2000
D4PC40 × D4PC13 15 154 63 – 41 (RFLPs) 1115 Not estimated Brown et al. 2001
45 5 isozyme
Pinus palustris
Tree 3-356 16(+6pairs) 133 133 – – – 1635 85% Nelson et al. 1994
P. elliotti and
P. palustris hybrids
F1 of - 3-356 × H-28 ♀: 18(+3 pairs) 122 122 – – – 1368 81% Kubisiak et al. 1995
♂: 13(+6 pairs) 91 91 – – – 953 62%
BC1 of 488 × 18-27 ♀: 17 133 133 – – – 1338 91% Weng et al. 2002
♂:19 83 83 – – – 995 81%
P. elliotti and
P. caribea var.
hondurensis hybrids
F1 of 2PEE1-102 × ♀: 24 125 117 8 – – 1548 82% Shepherd et al. 2003
1PCH1-63 ♂: 25 155 145 10 – – 1823 88%
Chapter 2 Pines
43
44 C. Plomion et al.

The two-way pseudotrestcross mapping strategy was SSR markers tested, only 9 were mapped in both
used to construct genetic linkage maps of maritime species; however, these codominant markers along
pine using AFLP markers (Chagné et al. 2002; Ritter with several codominant RFLP markers were useful
et al. 2002) as well as AFLP, SAMPL, SSR, and gene- in defining homeologous LGs between the species.
based markers (N. de María and M.T. Cervera, un- Temesgen et al. (2001) added 56 ESTP markers to
published results). Comparing the total map distance the consensus map developed earlier by Sewell. The
of genetic maps constructed based on haploid and DGGE method proved quite useful for assaying ESTP
diploid progeny from the same Maritime pine tree, markers and suggested a general method for plac-
a higher rate (28%) of recombination in the pollen ing genes on the maps since the markers were de-
parent was found (Plomion and O’Malley 1996). Such veloped from expressed sequences. Additional ESTP
a significant difference between male and female re- markers were developed and used to identify an-
combination was also reported in other pine species chored reference loci based on their sequence sim-
(Moran et al. 1983; Groover et al. 1995; Sewell et al. ilarity between species and their nature to map to
1999). conserved locations in more than one species (Brown
et al. 2001). Zhou et al. (2003) mapped 51 SSR mark-
Loblolly Pine ers, covering 795 cM on 15 LGs, in a three-generation
Loblolly pine (Pinus taeda L.) has been used exten- pedigree with 118 progeny. The markers, developed
sively for genetic mapping including the development from loblolly pine libraries of three types, were not
of an early map based on 20 isozyme loci that in- found to be clustered within the genome, further high-
cluded five linkage groups (LGs) (Conkle 1981). More lighting the value of SSR markers in genome map-
recently, maps have been constructed in several pedi- ping.
greed populations using several types of DNA-based
markers. Devey et al. (1994) published the first map Radiata Pine
utilizing 90 RFLP and six isozyme loci. The map was Over the past 15 years, a range of DNA-marker-
based on a three-generation pedigree with 95 progeny based linkage maps have been constructed for this
and revealed 20 LGs. Genomic mapping was success- species. Wilcox (1997) briefly reviewed mapping stud-
fully used by Wilcox et al. (1996) to define a single gene ies undertaken up till that date. Although results were
locus for resistance to an isolate of the fungus (Cronar- summarized from seven studies involving construc-
tium quercuum f. sp. fusiforme) that causes fusiform tion of eight linkage maps, only one map had actu-
rust disease. O’Malley et al. (1996) described a RAPD- ally been published by that date, consisting of 208
based map of clone 7-56, a top P. taeda parent. This (mostly) RFLP, SSR, and RAPD markers (Devey et al.
map had been constructed several years prior to pub- 1996, 1999). This map consisted of 22 LGs and cov-
lication. The first consensus map was produced by ered 1,382 cM. All of the other maps reviewed by
Sewell et al. (1999) combining data from two three- Wilcox (1997) were constructed using RAPD markers,
generation pedigrees, including the pedigree used either using haploid megagametophytes, or diploid
by Devey. RFLP, RAPD, and isozyme markers were tissues using a pseudotestcross approach. The num-
placed on the integrated or consensus map contain- ber of markers used in these studies ranged from
ing 357 loci and covering about 1,300 cM of genetic 124 to 290, with only one of the maps having LGs
distance on 18 LGs. The first complete genome map equal to the haploid number of chromosomes and
was developed by Remington et al. (1999). They uti- the remainder ranging between 14 and 22 LGs. To-
lized haploid megagametophyte samples from an in- tal map length estimates were undertaken in three
dividual mother tree to develop a map based on 508 studies ranging between 1,978 and 3,000 cM. Subse-
AFLP markers. This map revealed 12 LGs equaling the quent to these earlier studies a number of maps were
basic number of chromosomes and about 1,700 cM of published, some of which were included in Wilcox’s
genetic distance. Their analysis suggested that this 1997 review. Using 222 RAPD markers to genotype
distance saturated the genome, in slight contrast to 93 megametophytes, Emebiri et al. (1998) constructed
an earlier estimate of 2,000 cM based on data from a linkage map that covered 14 LGs and spanned a total
three species of pines (Echt and Nelson 1997). SSR distance of 1,665 cM. Kuang et al. (1999a) described
markers developed in radiata pine (P. radiata) were a map constructed using megagametophytes of 198
used in a comparative mapping project between radi- S1 seeds that had been genotyped with 168 RAPD
ata and loblolly pine (Devey et al. 1999). Of the 20 and four microsatellite markers. The resulting map
 Chapter 2 Pines 45

consisted of 19 LGs, covering 1,116.7 cM, which was Longleaf Pine, Slash Pine, Caribbean Pine,
estimated to cover 56% of the genome. Because this and their Hybrids
was constructed using an S1 family, elevated levels
of segregation distortion were observed. Wilcox et al. Longleaf pine (P. palustris Mill.), slash pine (P. elliottii
(2001a) published framework maps of both parents Engelm. var. elliottii), and Caribbean pine (P. caribaea
of 93 full-sib progenies based on a total of 429 AFLP, Morelet.) are hard pines of subsection Australes found
RAPD, and SSR markers. These parent-specific maps along the coastal plains of the southeastern United
were constructed using a pseudotestcross strategy and States, eastern Central America, and the Caribbean
covered 1,414 and 1,144 cM in 20 and 21 LGs, re- islands. Although loblolly pine (Pinus taeda L.) has
spectively. These maps have subsequently been added been planted on millions of acres that were once typ-
to, using over 300 SSR and EST markers, and have ically occupied by these species, in many situations
been reduced to 12 LGs (Wilcox et al. 2004). De- these pines are proving to be the preferred timber
vey et al. (1999) published a comparative map of species due to their adaptation to the coastal-plain
radiata and loblolly pine based on RFLP, SSR, and soils and the associated natural disturbances such
RAPD loci and showed that the highly syntenic na- as frequent fires and hurricanes (Wahlenberg 1946;
ture of Pinus applies to these two economically im- Shoulders 1984). A number of genetic maps, con-
portant species. Overall estimates of map length ap- sisting primarily of RAPD markers, have been con-
pear to be similar to that of loblolly pine (Wilcox et al. structed for slash pine, longleaf pine, and their hy-
2001a). brids (Nelson et al. 1993, 1994; Kubisiak et al. 1995),
with the main goal being to use these marker maps as
a tool for dissecting the inheritance of specific traits
Scots Pine of interest and for use in marker-assisted-selection
Early mapping work in P. sylvestris has been based (MAS) strategies within tree-improvement programs
on isozyme loci (Rudin and Ekberg 1978; Szmidt and (Kubisiak et al. 1997, 2000; Weng et al. 2002). Some
Muona 1989). These maps contained no more than 20 markers significantly linked to traits of interest have
loci. The number of RFLP markers developed for Scots been converted to more easily scorable markers, such
pine had been very low (Karhu et al. 1996). Thus, the as sequence characterized amplified region (SCAR),
next mapping efforts were based on RAPD makers to aid selection efforts (Weng et al. 1998). Brown
segregating in haploid megagametophytes. Yazdani et al. (2001) assembled a genetic linkage map for
et al. (1995) mapped 261 markers in 14 LGs in a tree slash pine using a variety of markers [RAPDs, ex-
that is part of the breeding program. Hurme et al. pressed sequence tag polymorphisms (ESTPs), re-
(2000) made a low-coverage RAPD map for an F1 tree striction fragment length polymorphisms (RFLPs),
that was a result of north × south cross, such that and isozymes]. An additional genetic map for slash
alleles for important quantitative traits were assumed pine and one for Caribbean pine, using amplified
to segregate in the same cross. The map with the best fragment length polymorphisms (AFLPs) and mi-
genome coverage, so far, was constructed by Yin et al. crosatellite or SSR markers, were created using an F1
(2003). The AFLP mapping in a full-sib family resulted hybrid population (Shepherd et al. 2003). The num-
in two maps, for each of the breeding program par- ber of markers mapped and the genetic distances
ents. The map lengths for the two parents, based on covered by some of the published maps for these
about 200 framework markers, were about 1,645 and species and their hybrids are summarized in Ta-
1,681 cM for the male and female trees, respectively, ble 3. Unlike loblolly pine, significantly less effort
with very high estimated genome coverage. Most re- has been focused on comparative mapping across
cently, Komulainen et al. (2003) mapped about 60 gene these species. However, studies have shown synteny
based markers in the F1 progeny of a north × south across slash and longleaf pines using RAPDs (Ku-
cross. This map also contained markers that had been bisiak et al. 1995, 1996), slash and loblolly pines using
previously developed for P. pinaster (Plomion et al. ESTPs as anchored reference loci (Brown et al. 2001),
1999), and others that had been used for P. taeda. and slash and Caribbean pine using AFLPs (Shepherd
Most importantly, the homologous markers allowed et al. 2003). In addition, SSR markers look promising
defining the correspondence between 12 LGs in the for further comparative analyses across these species
two species. More markers were later added to this (Shepherd et al. 2002a; C.D. Nelson and C.S. Echt, pers.
map (Pyhäjärvi et al. unpublished). comm.).
46 C. Plomion et al.

Other Pines that their genome might be relatively well conserved

Genetic linkage maps have also been constructed in among species. Approaches for evaluating genome
other pine species including P. brutia (Kaya and Neale similarity have used cytogenetics and linkage map
1995), P. contorta (Li and Yeh 2001), P. edulis (Travis comparison. Although cytogenetics can provide
et al. 1998), P. massoniana (Yin et al. 1997), P. strobus a direct idea about the conservation between
(Echt and Nelson 1997), P. thunbergii (Kondo et al. different genomes (Hizume et al. 2002), most of
2000; Hayashi et al. 2001). These were based mainly the interspecies comparisons have been carried
on RAPDs and AFLPs. out using genetic maps. The same tools have also
been developed for other applications such as gene
mapping and QTL detection. Comparative genome
2.2.4 mapping aims to measure the conservation of gene
Genetic vs. Physical Size and Practical Implication content (synteny) and order (colinearity) among
chromosomes and uses orthologous loci as anchor
These mapping studies have led to the conclusion points between maps. Comparative genome mapping
that the total genetic distance of the pine genome has been successfully used in grasses to explain the
is around 2,000 cM (Gerber and Rodolphe 1994), i.e., genome evolution of cereals (Moore et al. 1995), for
about 167 cM per chromosome. Given a physical size choosing a model species for whole-genome sequenc-
of 25 pg/C, one unit of genetic distance (1 cM) would ing (rice, Ware et al. 2002; [Link]
therefore correspond to 13 × 106 nucleotides (13 Mb), English/Projets/Projet_CC/organisme_CC.html),
while it represents 0.23 Mb in the model plant species and for transfering genetic information between
Arabidopsis! Such high genetic/physical size ratio ob- related species, such as the position of candidate
viously hampers the characterization of QTLs by fine- genes (Schmidt 2002). The lack of genome sequence
mapping and positional cloning approaches. Hence, for a Pinaceae species has made comparative map-
as will be discussed in Sects. 2.3 and 2.4, the only ping even more important as the primary tool for
way for understanding the molecular basis underly- integrating genetic information across species.
ing quantitative trait variation is the candidate-gene To define the syntenic relationships among
approach in which genes are identified a priori as phylogenetically related pine species, orthologous
likely candidates for the trait of interest and their markers (i.e., homologous DNA sequences whose
polymorphisms tested against quantitative trait vari- divergence follows a speciation event and whose
ation. Interestingly, despite the 56-fold difference be- sequence and genome location is conserved between
tween Pinus and Arabidopsis chromosomes, the num- different species) are used. A first example was
ber of crossings over per chromosome was found to provided by Devey et al. (1999), who aligned the
be highly conserved between both genera: 2 to 4 chi- genetic maps of P. taeda and P. radiata using RFLPs
asmata per bivalent (1 chiasma = 50 cM; Ott 1991). and SSRs. This first effort was further consolidated
A comparison of genome lengths among evolution- in the frame of the Conifer Comparative Genomics
ary divergent pines found P. pinaster, P. palustrus, and Project ([Link] Low-
P. strobus to have essentially identical rates of recombi- copy cDNA PCR-based markers were developed
nation (Echt and Nelson 1997). Thus, genetic mapping in loblolly pine (Harry et al. 1998; Brown et al.
studies carried out in pines have clearly demonstrated 2001; Temesgen et al. 2001; Krutovsky et al. 2004).
that the mechanism of crossing over is conserved on These markers showed a relatively good PCR cross-
a chromosomal basis, and independent of physical amplification rate between pine species because they
map size and the fraction of coding DNA. target conserved coding regions, showing a relatively
high polymorphism rate and a low number of
paralogous amplification when PCR primers were
2.2.5 chosen carefully. They were used to study the synteny
Comparative Mapping: Toward the Construction between species belonging to the Family Pinaceae,
of a Unified Pine Genetic Map which included pines, along with other important
conifers such as spruces and firs. These markers
All pine species have the same number of chro- made it possible to assign LG homologies for 10 out
mosomes (i.e., n = 12) as well as a similar genome of 12 chromosomes between P. elliottii and P. taeda
size. Moreover, they are all diploid, which suggests (Brown et al. 2001; [Link]
 Chapter 2 Pines 47

ccgp/[Link]), 10 out of 12 between of pine species for which maps have not yet been
P. pinaster and P. taeda (Chagné et al. 2003), and 9 developed.
out of 12 between P. sylvestris and P. taeda (Komu- As more conifer ESTs become available in public
lainen et al. 2003). About 30 or 40 ESTP markers databases (329,531 in Pinus teada, 132,531 in Picea
were proved to be useful in demonstrating large glauca, 27,283 in Pinus pinaster, 28,170 in Picea engel-
areas of synteny between each species pair. While mannii × Picea sitchensis, 80,789 in Picea sitchensis,
this comparison was only of low density, these 6,808 in Pseudotsuga menziesii, 7,639 in Cryptome-
pioneering studies suggested that pine genomes ria japonica: EMBL 19 March 2006), a computational
did not show any apparent chromosomal rear- approach could be used for in silico development of
rangement. They also provided an indication that putative orthologous EST-based markers (Fulton et al.
gene content and gene order is conserved, as is 2002), as was recently illustrated between loblolly pine
illustrated for LG 6 in Fig. 2. Current efforts are and Douglas fir by Krutovsky et al. (2004). Such re-
being expended to add more markers common sources should help to define the precise syntenic re-
to P. radiata and P. taeda (P. Wilcox, personal lationship across conifers and establish a framework
communication). From an application point of view, for comparative genomics in Pinaceae.
these comparisons provide a set of markers that can The alignment of genetic maps of P. pinaster and
be used for constructing framework genetic maps P. taeda made it possible to discover putative con-

Fig. 2. Synteny in the Pinus

genus (linkage group 6):
alignment of genetic maps
of five pine species
48 C. Plomion et al.

served QTLs between the two species (Chagné et al. periments of Sax (1923) and Thoday (1961), and it is
2003). Those are QTLs for wood density and wood now well known that a small number of segregating
chemical composition traits located on LGs 3 and 8, loci are involved in the genetic control of quantitative
respectively. The same observation has been made trait variation. These genes act together to provide
in P. radiata (Telfer et al. 2006). Preliminary studies a quantitative difference and are referred to as quan-
indicate wood-density QTLs are colocating more fre- titative trait loci or QTLs (Geldermann 1975). The
quently than would be expected by chance between basic theory of using genetic markers to detect QTLs
these two species (Telfer et al. 2006), and analyses was introduced by Sax (1923). Initially, the applica-
are being extended to other wood property traits. tion of this theory was limited by the lack of avail-
Moreover, candidate genes coding for functions that able segregating markers; however, rapid advances in
are linked to wood formation have been mapped DNA marker-based technologies since the 1980s have
in the same regions, which indicates that they may now made it possible to genoptype hundreds of ge-
be involved in the molecular control of those traits. netic markers to construct dense genetic linkage maps
These first examples of the application of compara- (Sect. 2.2.3) and further to carry out a comprehensive
tive genome mapping in pines show that comparative search of QTLs along the genome. Sewell and Neale
genome mapping can be used to verify QTLs across (2000) and Guevara et al. (2005b) recently reviewed
species and that the same genes may be involved in the science of QTL mapping in pine trees.
the genetic control of the same traits. In the first part of this section, devoted to the ge-
netic dissection of agronomically important traits, we
will present an update of the studies that were car-
ried out in pines. The specificity of pines regarding
2.3
the different types of populations and strategies used
Genetic Architecture to detect QTLs will be presented. Then, the main re-
of Complex Traits sults of these studies will be discussed. In the second
part, a perspective on the identification of diagnostic
Pine tree improvement is hampered by different in- markers for pine-tree breeding will be discussed.
herent characteristics: (1) the time needed to reach
sexual maturity, (2) the time lag required to evaluate
field performance (e.g., growth; Kremer 1992), and 2.3.1
(3) in some cases, the cost of phenotyping (e.g., wood- Strategy and Methods Used for QTL Detection
quality-related traits). This makes breeding of these in Single Family Pedigrees
species a slow process compared to that of most com-
mon crop plants. In addition, selection of these traits Pines are characterized by late maturity (longevity),
remains imprecise because environmental effects are an outbred mating system, and a high genetic variabil-
rather high for most traits of interest. Heritabilities ity (Hamrick and Godt 1990). Their outbred mating
for height, diameter, volume, branching traits, and system and high genetic load have hampered the de-
bole taper, i.e., straightness, are in the range of 0.1 to velopment of inbred lines, the material of choice from
0.3, and only slightly higher (0.3 to 0.6) for wood and a QTL mapping perspective. In this context, specific
end-use properties (reviewed by Cornelius 1994). In populations and statistical methods were developed
this context, any tool directed toward selection pro- specially for forest trees, and pines in particular. Until
cesses that would improve the evaluation of genetic recently, most of the QTL mapping efforts were fo-
value and also reduce the generation time would be of cused on single family pedigrees. However, given the
considerable value. high rate of polymorphism encountered in pines and
Most traits important to forestry, such as biomass the relative lack of stability of QTLs in different genetic
production, wood quality, and biotic and abiotic stress backgrounds, methods aimed at validating markers
resistance are complex quantitative traits. In the the- linked to the traits of interest in more complex pedi-
ory of quantitative genetics it is assumed that the grees or even in unrelated genotypes are emerging.
heredity of a quantitative trait can be ascribed to the In this section, we will review the type of population,
additive effects of a large number of genes with small statistical methods, and main results that have been
and similar actions, modulated by environment. This obtained using simple mapping pedigrees. The limita-
assumption has been questioned since the early ex- tions of this approach will be underlined, and the use
 Chapter 2 Pines 49

of complex pedigree designs or unrelated populations inbred F2 model. Although RAPD markers were used,
to unravel the genetic architecture of complex traits genetic information from progeny and corresponding
will be presented. Finally, the relevance of integra- megagametophytes were utilized to overcome the lim-
tive approaches combining classical QTL studies and itations associated with dominant markers using an
transcriptome and proteome analysis, together with F2 model. It was also proposed to use trans-dominant-
studies of molecular evolution, will be highlighted. linked markers to overcome the problem of dominant
markers on the sporophytic phase of F2 trees (Plomion
et al. 1996c).
Single Family Designs
Two-Generation Half-Sib Design With this popula-
Two-Generation Full-Sib Design The advantage of
tion structure, the effects of two maternal QTL alleles
naturally high levels of heterozygosity in outbred for-
are averaged over a large pollen pool (Hurme et al.
est trees can be utilized in a two-generation popula-
2000). This type of structure allows one to test the
tion structure, where it may be expected that trees
stability of the effect of the maternal alleles in differ-
chosen as potential parents will likely be heterozy-
ent genetic backgrounds. In addition to these classical
gous for some number of QTLs, which will then seg-
mapping designs, particular populations (F1 S) were
regate in the F1 . Typically, “plus trees” are identified
especially developed to analyze inbreeding depression
and are used as parents of a QTL mapping popu-
in P. taeda and P. radiata (Kuang et al. 1999a,b; Rem-
lation. Various studies have taken advantage of this
ington and O’Malley 2000a,b; Williams et al. 2001).
two-generation design to analyze each parent under
a pseudotestcross model (Kumar et al. 2000; Lerceteau Two-Generation Full-Sib and Half-Sib Design An
et al. 2000; Ball 2001; Shepherd et al. 2002b; Weng extension of the pseudotestcross QTL mapping strat-
et al. 2002; Markussen et al. 2003; Yazdani et al. 2003). egy, in which QTLs are defined in a narrow genetic
This model is well suited for dominant markers. How- background, Plomion and Durel (1996) show that
ever, the main limitation is that the phenotypic effects a “general” value of a “specific” QTL detected in a full-
inherited from each parent are analyzed individually, sib family could be easily evaluated, provided that
even though the genetic contribution of each parent both parents of the full-sib were involved in maternal
simultaneously contributes to the phenotypic vari- half-sib (open-polinated or polycross) families. Such
ation in the progeny population. Consequently, the two-generation pedigrees are widely available in most
genetic information in the four progeny classes of pine breeding programs that involve the simultaneous
an outbred pedigree is collapsed into only two geno- estimation of specific and general combining abilities
typic classes, thereby reducing the robustness of the of selected trees. However, this strategy has never been
analysis. Of course, if codominant markers are used, tested experimentally.
a consensus map can be built precisely to detect allele
effects from both progenitors simultaneously.
Methods Used for QTL Mapping
Three-Generation Full-Sib Design In the three- in Single Family Designs
generation outbred population structure, two crosses Regardless of the population structure and size
are made among four unrelated grandparents, where (Beavis 1994), several factors must be considered for
each mating pair is selected among individuals dis- successful QTL detection. Statistical methodology
playing divergent phenotypic values for the trait (e.g., significantly influences the accuracy of QTL position
Groover et al. 1994). From each grandparental mat- and effect estimation. Simple statistical methods
ing, a single phenotypically intermediate individual such as analysis of variance (ANOVA) have opened
is chosen as a parent. Presumably, these intermediate the way to the development of more powerful
parents are heterozygous for both marker and QTL QTL detection methods, integrating information
alleles and are potentially heterozygous for different available at multiple markers: interval mapping (IM),
allelic pairs that display a divergent phenotypic ef- composite interval mapping (CIM), and multiple
fect. This three-generation full-sib structure is typ- interval mapping (MIM).
ically designed for QTL analysis under an outbred The first method, called single marker analysis,
model and has been used extensively (Table 4). Alter- proposed by Edwards et al. (1987), is the simplest one.
natively, Plomion et al. (1996a, b) utilized the selfing ANOVA is performed with one marker at a time, on
ability of P. pinaster to design experiments that fit an the genotypic classes defined by a single marker. This
 50

Table 4. QTL and marker-trait association detected in pines: pedigree structure and methods

Species Objective Population Replicationsb Clonal Genetic QTL analysis Software Reference
(size)a replicates markers (analytical model)c

Pinus elliottii Detection F1OB (186) 1 No RAPD SG (54 : 27 genotypes SAS Kubisiak et al. 2000
from each tail), SM, MM
C. Plomion et al.

Pinus elliottii Detection F1OB (89) 2 sites (60,29) No AFLP, SSR SM, CIM QTLcartographer Shepherd et al. 2005
var elliottii ×
Pinus caribaea
var hondurensis
(Pinus palustris Mill × Detection, F1OB (258) 3 sites (82,83,93) No – SM, IM SAS, Weng et al. 2002
Pinus elliottii Engl) × stability (age) MAPMAKER/QTL
Pinus elliottii Engl
Pinus sylvestris Detection F1OB 1 No AFLP IM Qgene Lerceteau et al. 2000
Detection, F2HS 2 No RAPD SG (1994 : 48 vs. 48, SAS, Multimapper/ Hurme et al. 2000
Stability 1996 48 vs. 48), SM, OUTBRED
(time replicate) Bayesian QTL analysis
Detection F1OB 1 No RAPD SM SAS Yazdani et al. 2003
Pinus pinaster Detection, F2S (120) 3 (age) No RAPD IM MAPMAKER/QTL Plomion et al. 1996a
Stability (Age)
Detection F2S 1 No RAPD, AFLP, CIM QTLcartographer Costa 1999
proteins
Detection F2OB (186) 1 No AFLP CIM MULTIQTL Brendel et al. 2002
Detection F20B (186) 1 No AFLP, EST CIM MULTIQTL Pot et al. 2005
Detection F1OB (80) 1 No AFLP, SSR, EST IM SAS Markussen et al. 2002
Detection, F20B (186) 1 No AFLP, EST CIM MULTIQTL Pot 2004
stability
(age, season)
Detection F2S (120) 1 No RAPD IM MAPMAKER/QTL Plomion et al. 1996b

a F1OB: two-generation outbred design, F2OB: three-generation outbred design, F2S: three-generation inbred design, F2HS: Two-generation half-sib design, CD: complex design (in this
case more information is provided)
b When the traits were measured on one site and at one given time, 1 is indicated, otherwise more details are provided
c SM: single-marker analysis, MM: multiple-marker analysis (without map information), IM: interval mapping, CIM: composite interval mapping, SG: selective genotyping, BSA: bulk

segregant analysis
Table 4. (continued)

Species Objective Population Replicationsb Clonal Genetic QTL analysis Software Reference
(size)a replicates markers (analytical model)c

Pinus radiata Detection, F1OB (174) 1 No RAPD SM, BSA (9 vs 9) - Emebiri et al. 1997
stability (age)
Detection, F1OB (80-93) 3 times (age) No RAPD, AFLP, IM, MTM - Kumar et al. 2000
stability (age) SSR
Detection, F1OB (93) 2 times (age) No RAPD, AFLP, Bayesian approach Splus Ball 2001
stability (age) SSR
Detection CD 1 No SSR - - Kumar et al. 2004
Detection, validation, 3 unrelated F1OB 2 sites No RFLP and SSR SG (Juvenile wood GENSTAT Devey et al. 2004a
stability (400 each) density : 50 vs 50;
(genetic background) Diameter at breast
height : 100 vs 100), SM
Detection, Detection: 1 site for the No RFLP + SSR SM within family GENSTAT Devey et al. 2004b
validation 6 related F2OB (202) detection population + SM among families
Verification: 2 sites for the Yes
1 F2OB (400) verification population
Pinus taeda Detection F20B (Detection 6 sites No RFLP SG, SM Home Groover et al. 1994
population : 172) (19 to 35 trees per sites) made
Detection F20B (Detection 6 sites No RFLP IM Home Knott et al. 1997
population : 172) (19 to 35 trees per sites) made
Detection, stability 2 F2OB (Base : 95, 2 sites for BASE (48+47), No RFLP, RAPD, IM Home Kaya et al. 1999
(age, genetic Detection : 172) 6 sites for Detection Isozymes made
background) (19 to 35 trees per sites)
Detection, F2OB (91) 5 times (age) No SSR IM Home Gwaze et al. 2003
stability (age), made
development of
statistical methods
Detection, stability F20B (Detection 6 sites No RFLP IM Home Sewell et al. 2000
(age, season) population : 172) (19 to 35 trees per sites) made
Detection, stability F20B (Detection 6 sites No RFLP IM Home Sewell et al. 2002
(age, season) population : 172) (19 to 35 trees per sites) made
Detection, validation, 1 F2OB D : 6 sites No RFLP + EST IM QTL Brown et al. 2003
Chapter 2 Pines

stability detection (D) : 172, (19 to 35 trees per sites) express

(genetic background) validation (V) : 457
51
52 C. Plomion et al.

method suffers from several limitations. First, it does resulting number of possible interactions. Thus, the
not provide any information regarding the location of problem is no longer genetic but resides in the ability
the QTLs in the genome, and furthermore the defini- to test all the possible genetic models and select the
tion of the QTL effect is largely inaccurate given the one that best fits the observed data (model selection).
inability to separate small-effect QTLs with tight link- Until now, the MIM algorithm implemented in QTL
age from QTLs of large effect but more distant linkage. Cartographer developed by Kao et al. (1999) and Zeng
In order to improve the efficiency of QTL mapping et al. (2000) has not been directly used in pines, but
through the use of genetic map information, Lander Bayesian approaches, also based on model selection,
and Botstein (1989) developed the IM method. This have been applied in these species (Hurme et al. 2000;
approach allows QTLs to be detected in the intervals Ball 2001).
defined by the markers. The IM has been widely used The simultaneous analysis of multiple correlated
in pines (Groover 1994; Plomion et al. 1996a, b; Knott traits has also been incorporated into algorithms in
et al. 1997; Kaya et al. 1999; Costa and Plomion 1999; order to improve QTL detection efficiency (Jiang and
Costa et al. 2000; Sewell et al. 2000, 2002; Brown et al. Zeng 1995; Korol et al. 1995, 1998). Although these al-
2003). This method is, however, rather limited as QTL gorithms have not been specifically used up till now,
detection is made in a linear way, i.e., the same test there is much interest in such algorithms, especially
is applied at each point of the interval without taking for such traits as annual growth or wood density. Ad-
into account the results of successive tests. ditionally, application of specific genotyping strate-
More recently, statistical approaches have been de- gies has further maximized the efficiency of QTL map-
veloped to increase the statistical power of QTL detec- ping. One such strategy, called selective genotyping,
tion and have received increasing attention in pines has been extensively used in pines (Groover et al. 1994;
(Costa 1999; Brendel et al. 2002; Pot et al. 2005b). One Hurme et al. 2000; Kubisiak et al. 2000; Devey et al.
of these approaches is CIM, developed by Zeng (1993a, 2004a). Another strategy, termed bulked segregant
b) and Jansen (1993), which combines IM with mul- analysis (BSA), commonly used to study qualitative
tiple regression. Like IM, this method evaluates the traits, has also been applied to analyze quantitative
presence of a QTL at multiple analysis points across traits (Emebiri et al. 1997).
each interlocus interval. However, at each point it also Besides statistical procedures, it has been clearly
includes in the analysis the effect of one or more shown that the power of QTL detection largely de-
markers elsewhere in the genome. However, although pends upon the quality of the phenotypic assess-
IM and CIM brought significant improvements, both ment. Poor phenotypic assessments, i.e., imprecise
methods lack dimensionality: i.e., only a single QTL measurements, generally result in QTLs with true ef-
is being searched at a time. fects being left undected, but in some circumstances
The method proposed by Kao et al. (1999) and (but to a lesser extent) might even lead to the detection
Zeng et al. (2000) called MIM differed from the pre- of false QTLs. Without the possibility of developing F3
vious methods through the implementation of multi- populations or recombinant inbred lines to precisely
QTL models. The selection of the model (i.e., QTL estimate the value of the traits, clonally propagated
number, position, effect, and interactions) that best material has become the material of choice for for-
fits the data follows an iterative process. After iden- est tree geneticists (e.g., Scotti-Saintagne et al. 2004).
tification of QTL number and position by CIM, the However, this type of material has only been rarely
MIM strategy consists of looking for additional QTLs used in pines (Devey et al. 2004b). This is most likely
through forward-backward selection cycles while in- due to the large capital investments and technical ex-
tegrating interaction information between the differ- pertise needed to clonally propagate pines and the
ent QTLs. Within each iterative cycle, QTL position wide variability noted among specific genotypes in
and effect are reevaluated. MIM also allows the evalu- their ability to produce rooted cuttings.
ation of QTL epistasis. However, it is important to note
that, although this strategy provides significant im- 2.3.2
provements over CIM, it does not allow the detection QTL Discovery in Single Family Pedigree Designs
of nonsignificant QTLs at the individual level. There-
fore, even if the concept aimed at detecting multiple Twenty-six QTL studies aimed at detecting associa-
QTL effects is simple, its implementation is relatively tions between molecular markers and trait variation
complex given the number of potential QTLs and the have been performed in seven pine species. However,
 Chapter 2 Pines 53

most of the efforts have been concentrated in three show continuous variation, suggesting polygenic con-
species, P. pinaster, P. taeda, and P. radiata (Tables 4 trol of the traits, only a limited number of QTLs per
and 5). In addition, studies on inbreeding depres- trait (Table 4: between 0 and 8) have been detected.
sion have been carried out in P. radiata and P. taeda This number is in general smaller than in annual crops
(Kuang et al. 1999a, b; Remington and O’Malley 2000a, and has led to a smaller proportion of the phenotypic
b; Williams et al. 2001). variation being explained (Table 5). As stated in the
chapter on Eucalypts, “the limited power to detect
QTL in forest trees compared to crop species may be
Qualitative Traits
due to the high environmental and developemental
Associations between qualitative traits and molecu-
variation in tree plantations, as well as to the small
lar markers have led to the identification of markers
size of the analyzed populations” (mainly between 91
linked to resistance to different rust diseases: pine nee-
and 200 genotypes, Table 4). However, QTL analyses
dle gall midge in P. thunbergii (Hayashi et al. 2004),
carried out by Brown et al. (2003) and Devey et al.
white pine blister rust in P. lambertiana Dougl (De-
(2004a) using large mapping populations (>400 indi-
vey et al. 1995; Harkins et al. 1998), and fusiform rust
viduals) do not support the latest hypothesis. Brown
disease in P. taeda (Wilcox et al. 1996). A major gene
et al. (2003) reported QTL effects that were two- to
controlling the biosynthesis of δ-3 carene has also
threefold smaller than those reported by Sewell et al.
been mapped in P. pinaster (Plomion et al. 1996b).
(2000, 2002) for the same traits. Such divergence likely
Most of these studies have been based on intraspecific
represents more accurate estimates of the QTL ef-
mapping populations. Additionally, two interspecific
fects owing to the larger segregating population an-
crosses were used to increase the level of polymor-
alyzed. These results suggest that most QTL studies
phism and the range of variation for branch archi-
performed in pines, with the exception of the analy-
tecture in the mapping population (Shepherd et al.
ses carried out by Devey et al. (2004a, b) and Brown
2002b).
et al. (2003), have yielded an overestimation of the
QTL effects, which was also suggested by Beavis et al.
Quantitative Traits (1994). Consequently, as underlined by Wilcox et al.
Growth has been the most studied trait in QTL map- (1997), the genetic determinism of most target traits
ping studies in pines. To date, a comparable num- for pine breeding is likely to be explained by small-
ber of studies have been achieved for wood and end- effect genes, rather than any moderate- to large-effect
use properties. These two classes of traits have been genes. However, a larger number of QTL experiments
studied either globally (e.g., height growth, specific with larger pedigree sizes will be required to validate
gravity) or after decomposition into simpler compo- this hypothesis.
nents (e.g., growth unit, ring density). Traits involving Compared to the small number of QTLs detected
adaptation to the environment have also been stud- at any one maturation stage, the analysis of QTL stabil-
ied, including tree response to heavy metal, drought, ity along a cambial age or seasonal gradient (Table 4)
and cold stress, as well as bud phenology (Costa 1999; revealed a significant increase in the number of QTLs
Hurme et al. 2000; Kubisiak et al. 2000; Lerceteau et al. detected. For instance, the simultaneous analysis of
2000; Brendel et al. 2002; Yazdani et al. 2003). different maturation stages combined with seasonal
variation allowed for the detection of 7 to 23 QTLs for
QTL Results wood-specific gravity, the percentage of latewood, and
The diversity of population types, population sizes, microfibrilar angle (Sewell et al. 2002). Similar results
marker types, QTL detection methods, detec- were reported by Pot (2004), who observed a total of
tion thresholds, and variation of phenotypic trait 30 QTLs for wood density, 42 for wood heterogeneity,
measurements have made the comparison between and 33 for radial growth when measured over several
QTL experiments a difficult task. However, general cambial ages.
observations can be made regarding QTL number,
position, phenotypic effect, and stability. QTL Stability In their review, Sewell and Neale (2000)
pointed out that “before a commitment to marker-
QTL Number and Effect Most QTLs have been de- aided selection or breeding (MAS/MAB) can be made
tected at a single maturation stage and in a single en- in tree breeding, QTL that have been detected must be
vironment. Therefore, although all the analyzed traits verified in different experiments, as well as in differ-
 54

Table 5. QTLs and marker-trait associations detected in pines: number of QTLs and phenotypic variance explained

Species Type of trait Trait #QTlsa %Variance Note Reference

(abbreviation used in the article) explained by
each QTLb

Pinus elliottii Adaptative traits Aluminium tolerance 3∗ 15.61all Kubisiak et al. 2000
C. Plomion et al.

Pinus elliottii Wood-quality traits Average branch angle (AVBRA) 0 – Sheperd et al. 2002b
var elliottii × and growth traits Average number of branch per whorl (AVBRN) 3∗∗∗ 12–18
Pinus caribaea Average branch diameter (AVBRD) 2∗∗∗ 16–17
var hondurensis Average whorl spacing (AVWS) 2∗∗∗ 15–19
Regularity of whorl spacing (CVWS) 1∗∗∗ 17
Stem class (SC) 0 –
Number of distinctively large, steep angled 0 –
branches observed per tree (RAM)
Number of "leaders" (DL) 0 –
Bark thickness (AVBT) 2∗∗∗ 11–12
Relative bark thickness (RBT) 0 –
Trunk height (HT) 3∗∗∗ 13–21
Overbark diameter at breast height (OBDBH) 1∗∗∗ 16
Underbark diameter at breast height (UBDBH) 1∗∗∗ 17
Basic density (BD) 1∗∗∗ 14
(Pinus palustris Mill × Growth traits Total height (month 7,16,29,41) 5lod2 3.6–11 Weng et al. 2002
Pinus elliottii Engl) × Stem diameter 6lod2 4–10
Pinus elliottii Engl Height increments 5lod2 4.3–8.5
Diameter increments 5lod2 4.2–11
Pinus sylvestris Growth, wood-quality, Tree height (TH) 3lod2 11.5–12.2 Lerceteau et al. 2000
and adaptive traits Trunk diameter at breat height (DBH) 2lod2 9.3–15
Trunk diameter 0.5m from ground (D0.5) 2lod2 12.6–13.3

a∗ significant at chromosomewide level (p < 0.05),

∗∗∗ significant at genomewide level (p < 0.05),
lodx lod threshold value provided by authors (IM),
pv=x significance level used for single-marker analysis,
 : unique QTLs, defined as subset of QTLs (suggestive or significant) that map within ca. 15 cM of one to another and have the same general profile for their parental and interaction

effects (magnitude and direction of effects)

b when the % of phenotypic variance explained by each QTL was not provided, the phenotypic variance explained by all QTLs is indicated all
Table 5. (continued)

Species Type of trait Trait #QTlsa %Variance Note Reference

(abbreviation used in the article) explained by
each QTLb

Branch diameter of average branch at fourth 0 –

branch level from terminal bud (BDA)
Branch angle (BA) 1lod2 15.9
Basic wood density measured with Pilodyn (PIL) 0 –
Frost hardiness, critical temperature giving mean 2lod2 12.1–21.1
injury (CTm)
Frost hardiness, estimated temperature causing 2lod2 11.3–22.7
slight injury to 50% of needles (CT50)
Tree volume calculated with DBH (VOL) 2lod2 10.7–16.7
Tree volume calculated with D0.5 (VOL0.5) 2lod2 11–14.6
Adaptive traits Bud set 3 2.1–12.7 Hurme et al. 2000
Frost hardiness 7 3–11.1
Adaptive traits Cold acclimation 16p=0.01 9–19.5all Yazdani et al. 2003
Growth traits Height growth 11p=0.01 10.8–30.8all
Pinus pinaster Growth traits Total height week 15 2∗∗ 7–12 Plomion et al. 1996a
Total height week 38 3∗∗ 6.2–11.5
Total height week 92 1∗∗ 10
Fertile zone length (LF)_ cycle 5 1∗∗ 19.6
Fertile zone length (LF) adjusted to NSU_ cycle 5 1∗∗ 11.8
Number of stem units in the fertile zone (NSU)_ cycle 5 1∗∗ 20.4
Mean stem unit length (MSUL)_ cycle 5 1∗∗ 6
Germination date (GERMD) 2∗∗ 15–17
Hypocotyl length (Lhypo) 2∗∗ 9.5–12.7
Megagametophyte weight (MW) 2∗∗ 6–8.6
Adaptive Delta 13C 3∗∗∗ + 5∗∗ 4.7–12.4 pop : id to Brendel et al. 2002
and growth traits Ring width 1∗∗∗ + 5∗∗ 5.9–18.1 Pot et al 2006
and Pot 2004
Wood-quality Total height 0 – pop : id to Pot et al. 2006
and growth traits Mean density of all the rings 2∗∗ 4.6–4.8 Brendel et al 2002
Density estimated by Pilodyn penetration 2∗∗ + 1∗∗∗ 3.7–8.8 and Pot 2004
Chapter 2 Pines

Density heterogenity 3∗∗ 5.1–8

Lignin content 7∗∗ 5–9
55
 56

Table 5. (continued)

Species Type of trait Trait #QTlsa %Variance Note Reference

(abbreviation used in the article) explained by
each QTLb

Water extractives content 2∗∗ +1∗∗∗ 5.8–8.8

C. Plomion et al.

Acetone extractive content 0 –

Alpha cellulose content 4∗∗ 3.7–7.8
Hemicellulose content 3∗∗ + 1∗∗∗ 4.5–8.1
Lignin composition 3∗∗ 5.1–9.4
Kraft pulping yield adjusted to Kappa number 1∗∗∗ 8.8
Kappa index 3∗∗ 4.7–6.4
Arithmetic fiber length 2∗∗ + 1∗∗∗∗ 4.9–12.3
Weighted fiber length 2∗∗ + 1∗∗∗ 6.1–11.9
Fibre width 2∗∗ + 1∗∗∗ 5.4–8.2
Coarseness 2∗∗ + 1∗∗∗ 4.4–11.9
Curl 1∗∗ + 1∗∗∗ 6–10.2
Zero span tensile value 5∗∗ + 2∗∗∗∗ 4.9–7.3
Wood-quality Alpha cellulose content 5∗∗ 5.98–12.24 Markussen et al. 2002
and growth traits Lignin content 6∗∗ 5.98–13.17
Pulp yield 7∗∗ 9.65–18.43
Extractives content 2∗∗ 4.26–13.36
CIE brightness 3∗∗ 7.76–13.74
Mean wood density 8∗∗ 6.17–12.95
Minimum wood density 3∗∗ 5.09–8.58
Maximum wood density 6∗∗ 7.50–14.51
Diameter 7∗∗ 5.04–14.51
Height growth 3∗∗ 4.93–9.78
Secondary metabolism δ3Carene content 1∗∗∗ 26.5 Plomion et al. 1996b
Pinus radiata Growth traits Stem growth index 2p=0.01 9.23–10.56 Emebiri et al. 1997
Wood-quality trait Wood density 1–5 years old 1∗∗∗ – Pop identical Kumar et al. 2000
Wood density 6–10 years old 0 – to Ball 2001,
Wood density 14 years old 0 – LG3 only, multitrait
model analysis
Wood-quality trait Wood density 1–5 year-old 1∗∗∗ – Pop identical Ball 2001
Table 5. (continued)

Species Type of trait Trait #QTlsa %Variance Note Reference

(abbreviation used in the article) explained by
each QTLb

Wood density 14 year-old 0 – to Kumar et al.

2000, LG1
and 3 only
Growth and Diameter at breast height 2∗∗∗ _ 0 – Kumar et al. 2004
wood-quality traits Stem straightness 2∗∗∗ _0 –
Branching cluster frequency 2∗∗∗ _0 –
Wood density 1∗∗∗ _0 –
Growth and Diameter at breast height Verification: Verification: 2.2all Devey et al. 2004a
wood-quality traits 2p=0.05
Bridging pop : 0
Juvenile wood density Verification : Verification: 14.1all
8p= 0.05

Bridging pop :
4p=0.05
Biotic stress tolerance Resistance to Dothistroma 4p=0.05 in the detection: not estimated, Devey et al. 2004b
needle blight detection population, validation: 1.76–4.8
4p<0.05 in the
validation population
Pinus taeda Wood-quality trait Wood specific gravity 5p=0.05 23all Groover et al 1994
Wood-quality trait Wood specific gravity 4∗∗ , 1∗∗∗ – re-analysis Knott et al. 1997
of Groover
et al. 1994
Growth traits Height increment Base: 6∗∗∗ , Base: 23.1–30.5all , Detection Kaya et al. 1999
detection: 7∗∗∗ detection: 7.3–11.7all population: same as
Diameter increment Detection : 8 Detection : 12.5–59.5all Groover et al 1994,
Knott 1997,
Sewell 2000, 2002,
Brown et al 2003.
Growth traits Total height 2 years 2∗∗ 7.9–10.8 Gwaze et al. 2003
Total height 3 years 1∗∗∗ 10.3
Chapter 2 Pines

Total height 4 years 1∗∗∗ 12.2

Total height 5 years 1∗∗∗ 12.2
57
Table 5. (continued)
58

Species Type of trait Trait #QTlsa %Variance Note Reference

(abbreviation used in the article) explained by
each QTLb

Total height 10 years 1∗∗∗ 10.5

Growth rate 1∗∗∗ 11.3
C. Plomion et al.

Wood-quality trait Wood specific gravity (individual rings 23 5.4–15.7 Same population Sewell et al. 2000
and composite rings, measured for early as Groover et al.
and latewood) 1994 and
Knott et al. 1997
Percentage of latewood (individual rings 16 5.5–12.3
and composite rings)
Average microfibrilar angle (individual 7 5.4–11.9
rings and composite rings, measured
for early and latewood)
Wood-quality trait Cell wall content 8 5.3–12.7 Same population Sewell et al. 2002
as Groover et al.
1994 and
Knott et al. 1997
Wood-quality traits Wood-specific gravity V : 18 , V : 1.7–5.7, Detection population: Brown et al. 2003
U : 5 U : 1.8–4.4 same as
Percentage of latewood (individual rings V : 12 , V : 1.7–5.7, Groover et al 1994,
and composite rings) U : 5 U : 1.8–4.4 Knott et al. 1997,
Average microfibrilar angle (individual V : 4 , V : 1.7–5.7, Sewell et al. 2000, 2002
rings and composite rings, measured for U : 2 U : 1.8–4.4
early and latewood)
Cell wall content V : 10 V : 1.7–5.7
 Chapter 2 Pines 59

ent genetic and environmental backgrounds.” Indeed, 1994; Knott et al. 1997; Kaya et al. 1999; Sewell et al.
estimation of QTL stability (position and phenotypic 2000, 2002; Shepherd et al. 2002b; Weng et al. 2002;
variation explained) is one of the most critical factors Brown et al. 2003; Devey et al. 2004a, b), in only one
if QTL analysis is to be performed for application in study did the authors analyze QTL stability at different
MAS. This question has been addressed at different sites (Groover et al. 1994).
levels, including the stability of QTLs across differ- Genetic effect: a complete understanding of the
ent developmental stages (ontogenic or cambial age genetic variability of the traits of interest will rely on
effect), time points during the growing season (sea- the analysis of multiple populations as all the ma-
sonal effect), environments, and among diverse ge- jor genes involved in the genetic control of a given
netic backgrounds. trait are unlikely to be polymorphic in a single family.
Ontogenic effect: Given the long-lived character- On a more practical side, QTL stability across differ-
istic of forest trees, it has been questioned whether or ent genetic backgrounds is a prerequesite to marker-
not the same genomic regions would control quan- assisted breeding (MAB) in multiparental tree breed-
titative traits (e.g., annual growth, density) as trees ing programs. As underlined by Brown et al. (2003),
mature. Several experiments have been conducted MAB will reveal its full potential under two scenar-
to answer this question (Plomion et al. 1996a; Eme- ios: (1) if a genetic marker in full LD with molecular
biri et al. 1997; Kaya et al. 1999; Sewell et al. 2000, polymorphism causing trait variation at the popula-
2002; Weng et al. 2002; Brown et al. 2003; Gwaze et al. tion level is discovered or (2) if the gene (polymor-
2003; Pot 2004). In general, a rather low QTL sta- phism) underlying a QTL is identified. Given their
bility has been noted across different maturation allogamous reproductive system and their recent do-
stages for growth (Brown et al. 2003; Pot 2004). In mestication, pines are characterized by high levels
only one case did the authors report on the detec- of genetic diversity and low levels of LD (reviewed
tion of the same QTL regions at different matura- in González-Martinez et al. 2006b). The combination
tion stages (Gwaze et al. 2003). Conversely, Kaya et al. of these two factors (high diversity and low LD), to-
(1999) did not find a single “common” QTL con- gether with their perennial characteristics (matura-
trolling growth rate through tree development. The tion, environmental heterogeneity), is likely to con-
same trend – low QTL stability across maturation tribute to QTL instability across genotypes. Yet few
stages – was observed for wood density, a trait that studies have addressed this important issue in pines.
presents a higher juvenile-mature correlation than Kaya et al. (1999) did not find any QTL shared be-
growth (e.g., Williams and Megraw 1993, Hannrup tween pedigrees, while Brown et al. (2003) and Devey
and Ekberg 1998). This unexpected result suggests et al. (2004a, b) found only a small fraction of the de-
that some of the wood-density QTLs probably re- tected QTLs to be common across different genetic
flect the genotypic response to annual climatic vari- backgrounds.
ation. Recently, Rozenberg et al. (2002) reported on
the alteration of wood-density profiles in response to Multiple-Trait QTL Analysis As stated ealier,
drought. More interestingly, they showed that the al- multiple-trait QTL analysis will likely become very
teration of annual wood-density profiles (the pres- important for breeding purposes since pine breeding
ence of a false late wood ring in the early wood is a multitrait process. In several QTL experiments,
zone) in response to drought was genetically con- more than one trait (e.g., growth, wood quality)
trolled. was studied. In most cases, colocalizations between
Seasonal effect: Sewell et al. (2000, 2002) and Pot QTLs for different traits were observed, which might
(2004) analyzed the seasonal stability of growth and be expected for highly correlated traits. Possibly
wood-quality QTLs for traits measured in spring more important was the occurrence of multiple
(early wood) or summer (late wood). Overall, half of colocalizations in the genome, suggesting the effect
the detected QTLs were specific to one type of wood of pleiotropic genes rather than the existence of
(early vs. late wood). These results agree with recent physically linked genes controlling different traits. It
transcriptome studies that reveal that different sets of should be noted that QTL clusters were also observed
genes are regulated throughout the growing season for traits that were not phenotypically correlated
(Le Provost et al. 2003; Egertsdotter et al. 2004). (Brown et al. 2003; Pot et al. 2005b), suggesting
Environmental effect: although several studies strong environmental/developmental effects masking
have been performed in multisite trials (Groover et al. genetic correlation.
60 C. Plomion et al.

2.3.3 and 5 females). Parental trees were genotyped and

Future Direction on QTL Mapping the association between parental genotypes and the
performance of 200 full-sib-generated families were
As underlined in the previous section, the most reli- analyzed according to the strategy used in larch and
able QTLs – from a breeding perspective – are those eucalyptus by Arcade et al. (1996) and Verhaegen et al.
that have been consistently detected at different de- (1998), respectively. This analysis allowed the iden-
velopmental stages, in different environments, and in tification of several significant associations between
diverse genetic backgrounds. However, it is impor- markers and traits. However, it is important to note
tant to remind the reader that most of the studies that none of the marker-trait associations found using
performed so far in pines have been based on single full-sib family performance were identified when the
pedigree analysis, and that only a handful of exper- parental general combining ability was regressed on
iments have attempted to validate QTLs across dif- the allelic frequencies of the marker. This result sug-
ferent genetic backgrounds (Brown et al. 2003; Devey gests that the first associations detected were probably
et al. 2004a, b). Considering that pine improvement biased due to population structure (only five females),
involves the deployment of many families/clones, the as in larch and eucalyptus.
genetic stability of marker-trait association is a pre- Pan-genomic LD mapping in pines will require an
requisite before any extended use of molecular mark- extremely high marker density, given the low extent
ers is considered in operational breeding programs. of LD as estimated so far within the few genes ana-
There have been two major attempts to identify diag- lyzed. Alternatively, marker-trait association can be
nostic markers for MAB either using neutral markers performed on selected candidate genes.
spanning the genome or selecting candidate genes
based on their coincidence with QTLs. The Candidate-Gene Approach
Candidate genes (CGs) can be proposed based on
Complex Designs for Detecting the coincidence between QTLs and known functional
Marker-Trait Associations: LD Mapping genes putatively involved in the genetic control of
As underlined in Guevara et al. (2005b), while QTL the trait. Such positional CGs have been described
identification is based on physical LD generated in one in P. taeda (Brown et al. 2003) and P. pinaster (Chagné
or a few generations of crossing, association mapping et al. 2003; Pot et al. 2005b) for wood-quality-related
or LD mapping takes advantage of events that created traits. In P. taeda, colocalizations between genes in-
associations in the past to find a statistical associ- volved in monolignol biosynthesis (4CL, C4H, C3H,
ation between molecular markers and a phenotype and CcOAOMT) and QTLs for wood density were
on a much finer scale. As presented in Sect. 2.1.6., observed (Brown et al. 2003). In P. pinaster, a sin-
five studies have provided valuable information re- gle candidate gene-QTL colocalization was found
garding the extent of LD in pines. In all cases, it was between KORRIGAN, a gene involved in the hemi-
found that LD extended only over short distances. In cellusose/cellulose biosynthesis, and QTLs for hemi-
P. taeda, Brown et al. (2004) and González-Martínez cellulose and fiber characteristics.
et al. (2006a) revealed a rapid decay of LD within 800 Other types of analysis can be used to select
to 2,000 bp in candidate genes for wood quality and CGs related to plant adaptation. As underlined
drought-stress response. In P. sylvestris a rapid decay by González-Martínez et al. (2006b), “standard
of LD was also detected by Dvornyk et al. (2003) and neutrality tests applied to DNA sequence variation
García-Gil et al. (2003). The genotyping of a breeding data can be used to select candidate genes or amino
population of P. radiata with microsatellites (Kumar acid sites that are putatively under selection for
et al. 2004 also yielded the same result, i.e., no signif- association mapping.” Unusual patterns of nucleotide
icant LD was observed between pairs of genetically diversity and/or population differentiation have
linked markers, suggesting that LD decreases rapidly been detected in P. taeda (González-Martínez et al.
with physical distance. 2006a), P. pinaster, and P. radiata (Pot et al. 2005a).
LD mapping is only in its infancy in pines. To our Pot et al. (2005a) observed singular patterns of
knowledge, only one study has reported the use of nucleotide diversity in three genes: a glycin-rich
molecular markers (SSRs) in unrelated trees (Kumar protein homolog that was found to be up-regulated
et al. 2004). Marker-trait associations were analyzed in late wood-forming tissue (Le Provost et al. 2003);
in a P. radiata trial consisting of 45 parents (40 males CesA3, a cellulose synthase gene; and KORRI-
 Chapter 2 Pines 61

GAN, a membrane-bound endo-1,4-beta-glucanase From a commercial perspective, the key objective

involved in cellulose/cellulose biosynthesis. As men- is assurance of genetic gain by ensuring that seedlots
tioned before, colocalization between KORRIGAN and/or clones are true to intention. The applicability
and wood-quality QTLs has been reported (Pot et al. and efficacy of various DNA marker systems is de-
2005b). In P. taeda, although the action of neutral pendent, therefore, on the way in which genetic gain
processes cannot be completely ruled out to explain is delivered. In conifers, such delivery can come in
the patterns of nucleotide diversity observed for a number of different forms (and costs), even within
CcoA-OMT1, several characteristics of its nucleotide the same breeding program.
diversity seem to indicate the action of natural
Open-Pollinated Seed
selection on this gene. The colocalization of this gene
Open-pollinated seed from a mixture of selected seed
with a QTL of water use efficiency (Brendel et al. 2002;
orchard material generally consists of a few dozen ma-
Pot et al. 2005b) and the differential expression of
ternal parents, with little if any control of the pollen
this protein under different watering regimes (Costa
source. This is a common means of seed production,
et al. 1998) clearly emphasize the putative role of
particularly in commercial Pinus species in the south-
this gene in wood-trait variation. Association studies
eastern United States. In such cases, there may be
should now be used to validate this hypothesis.
a need to ensure no contributions from unwanted
maternal parents, which can be achieved by geno-
2.4 typing megagametophytes with sufficient markers,
Marker-Assisted Breeding assuming maternal parentage information is main-
tained for all seed. Furthermore, markers could be
This section discusses the application of informa- used to quantify the relative contributions of each
tion from DNA polymorphisms in conifer-type maternal parent to individual seedlots if seed counts
tree-improvement programs. Note here that such are not available. A variation on this method of seed
information includes not only purposefully designed production is used in mass pollination techniques,
markers but also information obtained directly from such as supplemental mass pollination or liquid pol-
(re)sequencing, for which specific markers may not lination. In these approaches, receptive conelets are
have been designed. Applications of information pollinated using pollens from selected genotypes by
from DNA polymorphisms fall into four generic various means, but the conelets themselves are not
areas: audit and quality control, elucidation of actually covered to prevent fertilization by unwanted
genetic phenomena, population management, and pollens.
selection and breeding. We discuss each of these Control-Pollinated Seed
aspects below, covering both existing and potential Control-pollinated seed, where pollens from selected
applications. Furthermore, we provide examples parents are used to fertilize selected seed parents,
where appropriate, as well as comments regarding in the process excluding pollen from other sources
the current status of each of these applications. using bags to cover receptive cones and/or under-
taking pollinations in contained greenhouse facili-
ties. This is usually undertaken to produce full-sib
2.4.1 families or individual half-sib families with known
Quality Control and Audit selected pollen parents. Such methods are used in
both breeding and commercial production for species
Development of an array of DNA marker systems over such as P. radiata. Vegetative propagation, some-
the past two decades has provided tree breeding pro- times involving in vitro technologies, can be used to
grams with a range of tools to achieve basic aspects of amplify genotypes, particularly for commercial pro-
quality control that hitherto could only be addressed duction, largely because of shortages of seed and/or
with difficulty, if at all. Seedlot and/or clonal fidelity cost of seed production. DNA markers – particu-
have been the key concerns for both commercial and larly codominant multiallelic marker systems such
research applications. Indeed, this area was the ear- as microsatellites – have been developed for such
liest commercial application of DNA markers in tree purposes (e.g., Devey et al. 2003). In addition, pa-
breeding programs and is still the most widespread, ternal parentage can sometimes be evaluated via pa-
at least in coniferous species. ternally inherited chloroplast markers. Such mark-
62 C. Plomion et al.

ers have been developed and applied in P. radiata 2.4.2

for checking paternal inheritance (Kent and Richard- Elucidation of Genetic Phenomena
son 1997), although judiciously chosen nuclear DNA
markers may also suffice. In general, markers have All genetic gain ultimately depends on DNA polymor-
shown that misidentification is common in breeding phisms. Knowledge of the nature and effects of the
programs; for example, Bell et al. (2004) estimated polymorphisms has the potential to generate far more
that 2.6% of parents were misidentified in a sam- gain than the use of purely phenotypic data – on the
ple of an Australian breeding population of P. radi- parents and/or their progeny. Information from DNA
ata and that 8.4% of offspring of ten families were polymorphisms enables the elucidation of phenom-
not consistent with expected parentage. These results ena such as understanding of the genetic architecture
also indicate that a proportion of misidentification of trait variation, revelation of population structure
of open-pollinated seedlots is possible. Furthermore, and history, and detection of selection fingerprints,
our experience at Scion with putative full-sib fami- all of which can have direct or indirect applications in
lies used for gene mapping experiments has revealed tree breeding.
very few such families – produced either commer- The genetic architecture of trait variation can be
cially or by research groups – are completely con- defined as the frequencies, location, magnitude, and
sistent with expected parentage (unpublished data). mode(s) of action of quantitative trait loci/nucleotide
DNA markers, particularly codominant marker sys- (QTL/N) effects underpinning quantitative traits.
tems, have therefore been useful in checking and as- While QTL mapping has been very informative in this
suring parentage. regard, the results are relevant only to the pedigree(s)
used, not to whole populations. Association genetics
Clonal Production may, therefore, be more relevant for understanding
Clonal production is used either for experimental pur- the genetic landscape of trait variation in forest
poses (such as clonal tests) or for mass-propagation trees.
of tested clones for clonal forestry. In these cases,
genotyping is undertaken to ensure that ramets do Estimation of Genetic Variance Structures
represent the desired genotype(s). Similarly, ram- and Heritabilities
ets deployed in seed orchards raise the same issue, DNA markers offer new ways of obtaining some
although sometimes misidentification of parents is key knowledge that is fundamental to tree improve-
detected via parentage testing of the seed obtained ment regarding genetic parameters, in particular ge-
via methods described above. For applications where netic variances, heritabilities, and correlations among
genotype fidelity is needed, “profiling” marker sys- traits. Such knowledge informs the breeder not only
tems such as RAPDs and AFLPs is useful in that they about the feasibility of breeding for certain traits or
are generally cheaper to both develop and use for combinations thereof, but also how it might be effi-
this specific application, particularly as the high level ciently undertaken.
of polymorphism revealed overcomes issues associ- For obtaining genetic gain, a trait must be both
ated with dominance. However, in more recent years, variable and heritable. Genetic correlations between
fingerprinting kits have tended to utilize codomi- economic traits can be a major constraint if they are
nant markers (see references above), as these have adverse, but they can provide the breeder with great
been developed for other purposes (above) and are opportunities if they are favorable. Even genetic cor-
generally adequate for clonal fingerprinting, particu- relations between noneconomic indicator traits and
larly if enough marker loci are used (e.g., Kirst et al. economic ones can be used to great advantage if
2005b). the correlations are strong and the indicator traits
DNA markers have, therefore, been developed highly heritable. For genetic variances and heritabil-
and utilized extensively in tree breeding pro- ities, if they have to be estimated from seed collected
grams, for both verifying commercially deployed in natural stands, then the coefficients of relation-
materials and for ensuring that experimental ship within seed-parent families need to be known.
materials meet the requirements. Even so, some Indeed, the relative contributions of inbreeding as
programs still do not universally implement or rely such and finite effective numbers of (unrelated) pollen
on such genotyping, largely due to the expenses parents per seed parent represent important infor-
involved. mation. While in the past isozymes were a valuable
 Chapter 2 Pines 63

tool for providing such information, DNA markers check these assumptions and provide empirical data
can be much more powerful for the purpose, and for more accurate estimates. Similarly, sequence data
with a greater range of applications. This issue of from genic regions can reveal evidence of selection:
nonrandom components in the mating system can for example, balancing selection was detected by Cato
be important with native stands of conifers (Burdon et al. (2006b) in a gene associated with wood density
et al. 1992), despite their wind pollination, even if and growth rate in P. radiata. Krutovsky and Neale
it is less acute than in insect-pollinated species like (2005) found evidence for selection in three of 18
eucalypts (Hodge et al. 1996). The wind pollination genes in Douglas-fir. Such evidence – which can be
of conifers tends to reduce the nonrandomness of generated on a relatively small subset of genotypes –
mating, while the conifers’ mechanism of archego- could be an effective prescreen for genes more likely
nial polyembryony probably reduces the inbreeding to be associated with trait variations, although some
component still further. For estimating genetic cor- caveats apply regarding the power to detect the effects
relations between traits, knowledge of coefficients of of selection (Wright and Gaut 2005).
relationship within such families is likely to be much
less crucial. As we will mention later, a priori family
information as such may not be essential for this pur- Assignment of Gene Function
pose, but it can be supplemented with information Knowledge of gene function, if acquired, offers the
from markers. greatest long-term opportunities to capture genetic
gain, using either endogenous variation or genes in-
Population Structuring and History troduced by conventional breeding or genetic engi-
The structuring and history of populations influ- neering. LD mapping and association genetics can
ence both the availability of genetic variation for provide clues in the search for quantitative trait nu-
the breeder to exploit and the potential for asso- cleotides (QTNs) regarding which genes have func-
ciation genetics to contribute. While large, essen- tional roles in trait variation. While relatively short
tially panmictic populations cannot be expected to stretches of disequilibrium (usually hundreds to low
have appreciable across-family linkage disequilib- thousands of base pairs) represent constraints for
rium (LD), cryptic structuring may exist that gen- breeding applications, a key advantage is the po-
erates significant disequilibrium that could be use- tential for assignment of function – even identifi-
ful for breeders. For example, localized population cation of individual QTNs. However, because of the
bottlenecks, followed by coalescences, could easily size of conifer genomes and the lack of genomic
cause this. Such LD could provide valuable clues sequence, identifying candidate genes will be cru-
to “metapopulation” history. Despite wind pollina- cial. Possible approaches are described in the fol-
tion, various factors can generate population struc- lowing section and in more detail by Wilcox et al.
ture in conifers (Mitton 1992). Interesting possibil- (2006).
ities of structure exist in populations derived from In carefully selected cases, genetic transformation
recent admixtures. In P. radiata, the exotic, domesti- can be used to verify the role of a QTN in generating
cated “landraces” still have large elements of the wild phenotypic variation. The costs of achieving a trans-
state. Interestingly, they evidently represent a genet- formation, and often the regulatory issues, will de-
ically recent fusion of two of the native populations, mand a highly selective application of this approach.
Año Nuevo and Monterey (Burdon et al. 1998), which On the other hand, knowledge of gene function may
may provide a basis for some admixture disequilib- be useful for identifying genes to target for genetic
rium. transformation, to create new variations of use for
Polymorphisms revealed by DNA sequence data breeders, and to provide commercial cultivars. There
derived from both genic and nongenic regions can re- are no reports yet of cloning QTLs from tree species,
veal much about the genetic history of those regions. partly due to the large number of candidates within
Departures from Hardy-Weinberg equilibrium could QTL confidence intervals, but also because of the time
reveal the presence of previously undetected genetic required for trait expression of transformants arising
phenomena such as the presence/absence of inbreed- from complementation studies. Nonetheless, associa-
ing. Indeed, genetic variance (and gain) estimates are tion genetics will be a key step in increasing resolution
based on assumptions regarding the relatedness of and, in some cases, identifying putative QTN for fur-
parents used in genetic tests. Such data can be used to ther analyses.
64 C. Plomion et al.

2.4.3 diversity will need to be cross-referenced with the

Population Management functional diversity since the two classes of diversity
are not necessarily closely coupled, at least among
Pedigree Reconstruction and Detection species (Morgante and Salamini 2003; Paran and Za-
of Genetic Contamination mir 2003). Components of DNA diversity obviously
Breeding populations represent the “engine room” for include percentage of polymorphic genes (in either
capturing the additive gene effects that allow cumula- the coding or regulatory regions), percentage of base
tive genetic gain over successive generations, through pairs that are polymorphic, particularly for QTN that
recurrent cyles of selection, intermating, evaluation, must exist in coding or regulatory regions), and allele
selection, and so on. Maintaining full pedigree has frequencies for the polymorphisms. Such informa-
been favored on the grounds that it helps preclude tion, in conjunction with knowledge of magnitudes
inbreeding and maintains effective population size. of QTN effects, provides a benchmark for monitoring
However, the expense of maintaining full pedigree changes in diversity, for any forest trees.
can limit the size of the breeding population that can Loss of low-frequency alleles is an obvious
be handled, raising the question of whether larger manifestation of a decrease in genetic diversity.
populations can be handled if pair crossing is not Paradoxically, abrupt increases in the frequencies of
mandatory. Moreover, there may be situations where such alleles, as can occur through genetic drift, can be
the breeder has reason to resort to material, e.g., com- a manifestation of the same phenomenon. While the
mercial stands, in which pedigree has been sacrificed inherently outbreeding genetic systems of forest trees
but which has the advantage of huge numbers (Burdon may be able to cope with significant inbreeding in the
1997). wild, through selection for balanced heterozygotes,
Modern marker technology has major potential such mechanisms may be impeded under conditions
for pedigree reconstruction (Lambeth et al. 2001; Ku- of artificial breeding. In principle, almost any sort of
mar et al. 2006), at least in open-pollinated families genetic marker should be able to manifest the losses
of known seed parentage. Complete reconstruction of of alleles or sharp fluctuations in their frequencies.
predigree will be more challenging, especially with Nevertheless, it seems preferable to know what genes
the wind pollination that characterizes conifers. In are of particular current or contingent importance
maintaining gene resources that underpin breeding and monitor their frequencies. This, however, will
populations, controlled crossing is typically far too depend on knowing the functions of the genes.
expensive, yet there may be a call for quantifying
and even detecting individual cases of contamination Provision for Biotic Crises
(Burdon and Kumar 2003). Different types of markers While it appears QTNs typically exert minor individu-
may be required for this purpose, and the task may be al effects in conifers, disease-resistance genes can rep-
challenging, but it appears inherently feasible. Pedi- resent a notable exception (Burdon 2001). Such genes
gree reconstruction is discussed later, in connection can both have large effects (despite outward appear-
with selection. ances of classical quantitative inheritance) (e.g., Kin-
loch et al. 1970; Wilcox et al. 1996) and be present at
Tracking and Maintenance of Genetic Diversity low frequencies (R.D. Burdon and P.L. Wilcox personal
A key element of population management is mainte- communication). These genes can also have the fea-
nance of genetic diversity, to give the breeder flexibil- ture of gene-for-gene specificities between host geno-
ity in both the short and long terms and to safeguard types and fungal strains (pathotypes), which has im-
continuing long-term genetic gain. While pedigree in- portant implications for ensuring durability of disease
formation is an indicator of genetic diversity, it does resistance against pathogen mutations (Burdon 2001).
not provide definitive information in itself. Achieving Establishing the nature of such genes can require care-
that poses significant challenges. fully planned mating between parents and inocula-
The “gold standard” for functional genetic di- tion studies based on single-spore isolates, preferably
versity will usually be performance in well-designed backed up with genomic studies. Here, as in other ar-
and properly located common-garden genetic experi- eas, comparative genomics can have a major role, at
ments. However, such experiments are expensive and least in studying resistance genes of lower specificity.
often slow to deliver results. The use of DNA poly- The implications of such patterns of genetic varia-
morphisms is clearly much quicker, but such marker tion within both hosts and pathogens, which are called
 Chapter 2 Pines 65

pathosystems, can be major for both population man- vantages include underrepresentation of some spe-
agement and selection. Disease resistance can be very cific families due to either chance or factors affecting
important in plantation forestry, especially with ex- fertilization, prevalence of inbreeding in some fami-
otics that are being grown in the absence of natural lies distorting breeding-value estimates of those fam-
or other pathogens. Conserving low-frequency genes ilies, difficulties in detecting full-sib-specific combin-
of large effect, which are sometimes important in ing ability in full-sib families, and genotyping costs.
conifers, can pose a threefold challenge. A large popu- Most of these disadvantages could be overcome, al-
lation may be needed to find trees with such resis- though further research is needed to fully evaluate
tance. The requisite population size can be increased various possible application scenarios. To date, this
by the desirability of obtaining resistance genes in un- approach is currently being evaluated in the context
related pedigrees. It may be further increased by the of operational breeding in New Zealand with Eucalyt-
desirability of combining (“pyramiding”) genes in the pus and P. radiata.
genetic material that represent diverse mechanisms of
resistance to the pathogen, toward ensuring durability Estimation of Heritability Based
of resistance against mutation or genetic shifts in the on Marker-Ascertained Relatedness
pathogen population. In implementing this approach, Because molecular markers have the potential to es-
identification of resistance genes will be of enormous timate relatedness between genotypes, methods to
help. estimate the heritability of phenotypic characteris-
The information on forest-tree pathosystems, in tics have been proposed and evaluated (Andrew et al.
general, is still very sketchy, although there are re- 2005; Kumar and Richardson 2005) without the need
sistance genes identified for some coevolved forest to have any prior knowledge of genetic relationships
pathosystems (e.g., Wilcox et al. 1996). Nevertheless, (see above). Such methods have the potential bene-
what is known indicates that preparations for biotic fit of allowing one to obtain information from exist-
crises in the form of new fungal diseases should in- ing forests (e.g., natural forests or commercial plan-
clude having very large population resources avail- tations), thereby obviating the need for genetic test-
able. In at least some cases, it is very doubtful whether ing involving progenies, thus speeding up the gener-
breeding programs effectively contain such a provi- ation of information, particularly for species where
sion (Burdon and Gea 2006). little a priori information is available. Some disad-
vantages involve the reliability of marker-based es-
2.4.4 timates, particularly as assumptions need to be ei-
Selection and Breeding ther made or checked regarding genetic structure
and prior levels of relatedness, as well as the reliabil-
Pedigree Reconstruction as an Alternative ity of phenotypic information, particularly if sourced
to Maintaining Pedigree Records from forests rather than purpose-designed common-
A proposed application (Lambeth et al. 2001) is to garden tests with appropriate controls. These ap-
use markers in genetic tests to reconstruct pedigrees proaches have not been extensively investigated but
retrospectively, as opposed to maintaining pedigree may have a role. In part, this may be due to the fact that
information throughout the life of a genetic test. This existing approaches are well established and success-
involves applying pollen mixes of known composi- ful, as well as to the existence of infrastructure such
tion to a range of seed parents, planting the resul- as extensive testing involving progenies. Kumar and
tant offspring in designed experiments with limited Richardson (2005) compared phenotype-based with
or no maintenance of family information (Kumar et al. marker-based heritability estimates for wood density
2006). Upon subsequent measurement, pedigree iden- in P. radiata and found very little difference. Simi-
tity is ascertained via DNA markers. Benefits of such larly, Andrew et al. (2005) reported nonzero estimates
an approach include logistical simplicity, cost reduc- of heritability for a number of foliar defence-related
tions for breeding and testing, potentially better es- chemicals in Eucalyptus melliodora, some of which
timates of genetic parameters and increased gains, were consistent with independent estimates based
and assurance of parentage (see above). Furthermore, on phenotype (see Andrew et al. 2005 and references
fewer financial costs are incurred if experiments are therein). Such approaches may be of use also for as-
abandoned or lost prior to remeasurement – which sociation genetics, particularly in revealing cryptic
does happen in tree-improvement programs. Disad- population structuring in natural forests, as well as
66 C. Plomion et al.

for selecting maximally unrelated individuals for es- Wu (2002) evaluated the tradeoff between proportion
timating LD. of variance explained by markers and shortening of
selection interval. For traits of moderate heritability
(0.2 to 0.4) marker-assisted early selection (MAES)
Within-Family Selection Based was more efficient than phenotypic selection, if they
on Marker-Trait Associations Derived explained more than 5 to 10% of additive genetic vari-
from Pedigreed QTL Mapping Populations ance and allowed the selection interval to be reduced
Such selection is based upon selecting individuals by half. For later-expressed traits, even less variance
within known full- or half-sib families where marker- explained by markers was necessary for markers to be
trait associations have been previously determined. more efficient than phenotypic selection. Wu (2002)
Genotypes with the desired multilocus genotypic also evaluated a number of other scenarios, including
composition are selected from within families. In combining phenotypic selection with markers, and
theory, this can be applied within families in breed- found that MAES is only marginally more efficient
ing populations and/or production populations. than phenotypic selection except in cases of relatively
Generic benefits include earlier selection, increased low juvenile-mature phenotypic correlations. Given
selection intensity, and potentially cheaper selection genetic architectures of QTLs explaining a few per-
(Stromberg et al. 1994). In conifer improvement, cent each (Wilcox et al. 1997; Brown et al. 2003; De-
breeding populations are usually based on a few vey et al. 2004a, b), and limitations for the selection
hundred parent genotypes; thus it is largely imprac- of up to ten unlinked markers per family linked to
tical and very expensive to detect associations for such QTLs, the effectiveness of early selection is lim-
each genotype independently (Strauss et al. 1992; ited for traits with low-moderate age-age phenotypic
Johnson et al. 2000). Rather, application of markers correlations, particularly once costs of QTL detection
is more likely to be restricted to populations where are taken into account. Time savings for traits of de-
specific families are advanced, as for example in layed expression or low juvenile-mature correlation
elite populations or in small nucleus- or mainline will, however, depend on the prior establishment of
breeding populations. QTL/phenotype associations.
A number of situations have been evaluated either All the above studies investigating the applica-
for species-specific scenarios or from a wider theo- tion of markers for breeding population advancement
retical perspective. One of the earliest such studies have generically shown that while there are scenar-
was that by Strauss et al. (1992), who concluded that ios where markers could be used to generate genetic
within-family marker-assisted selection (MAS) was of gains, the financial gains could be quite limited. This
limited or no value unless a high proportion of addi- is in part due to the cost of genotyping as well as the
tive genetic variation could be explained by markers length of time, meaning that additional costs will be
for traits of low heritability but high value and where incurred. A further limitation stems from the fact that
selection intensities within families were high com- most breeding objectives involve multiple quantita-
pared to that among families. They also concluded tively inherited traits whose genetic architectures are
that difficult- or expensive-to-measure traits such as likely to involve predominantly small-effect genes (see
wood quality or resistance to certain diseases showed references above). This means limited genetic gains
the most potential for MAS. Williams and Neale (1992) from individual markers. Furthermore, relatively few
evaluated the relative efficiency (RE) of MAS and also markers can be used to select within any particular
concluded RE was greatest where (a) traits were not family: for a simple two-genotype-per-marker sce-
expressed sufficiently early to enable a reduction in nario, it would be necessary to generate and geno-
generation length as well as (b) with lower-heritability type thousands of offspring per family to have suffi-
traits where markers explain a high proportion of ad- cient power to generate the optimal ten-marker-locus
ditive genetic variability for the traits. However, re- genotype. If such selection were restricted to a limited
sults from many QTL mapping experiments have in- range of traits, there would also be the need to aug-
dicated that the genetic architecture of quantitatively ment marker information with phenotypic records to
inherited traits is dominated by genes of small effect effectively address the breeding objective; therefore
(Wilcox et al. 1997; Sewell and Neale 2000; Brown et al. the opportunity for earlier selection could be lost. On
2003; Devey et al. 2004a, b), thus limiting the oppor- the other hand, few studies have investigated the ef-
tunity to explain sufficient variation with markers. ficacy of marker-based or marker-assisted selection
 Chapter 2 Pines 67

over multiple generations. Application over gener- characteristics. Some private companies in the USA
ations is likely to reduce costs, although not in a man- and New Zealand have undertaken or are still un-
ner linearly proportional to gains. This is owing to dertaking MAS on a limited scale, but there is as yet
the fact that additional revenues in future generations no widespread uptake either for breeding-population
are discounted relative to cost, as costs are effectively advancement or as a tool for more immediate genetic
incurred early (e.g., Johnson et al. 2000). gains. Nonetheless, there is still some potential, as
An exception to the limitations outlined above is some of the above studies have indicated, as well as
the selection for genes of large effect such as major- other possible areas of application yet to be explored.
gene disease resistance (see references above). In It may be, however, that other technologies, e.g., GAS
breeding populations, frequencies of resistance genes (see below) or genetic modification, may supersede
could be increased much faster via marker-based se- MAS.
lection than phenotypic selection; markers could ob-
viate most of the need for field and/or greenhouse Combined Among- and
screening by allowing only those individuals to be de- Within-family Selection Based upon Results
ployed in field tests (for screening for other traits) from Association Genetics Experiments
that have the favorable marker phenotypes. Where Advances in genomics technologies over the past 5
gene pyramiding is crucial for durable resistance, this to 10 years have made possible almost unrestricted
marker-based approach could be especially valuable. access to any region of tree genomes, particularly se-
As with all marker-assisted (or based) selection sce- quences within and associated with expressed genes.
narios, such genes need to be detected, which in- Variation in genic regions and the associated regula-
curs costs early in the breeding cycle. However, such tory regions form the basis of phenotypic variation.
costs are likely to be lower in that such large-effect LD mapping and association genetics are key tools
genes are less financially burdensome to detect, and in correlating such sequence variation with observed
in cases where resistance is valuable, financial gains trait variation. Unlike marker-trait associations from
can be considerable, particularly if undertaking de- pedigreed QTL detection populations (i.e., families),
velopment of disease-resistant (or tolerant) breeds. associations derived from association genetics have
Opportunities for MAS have also been evalu- applicability for both within- and among-family se-
ated for production populations in species such as lection. The term gene-assisted selection (GAS) has
Douglas-fir (Pseudotsuga menziesii) (Johnson et al. been used to describe this method of selection (Wilcox
2000) and Pinus radiata (Wilcox et al. 2001b), and et al. 2003, 2006; Wilcox and Burdon 2006); it dif-
for clonal deployment in P. radiata (Kumar and Gar- ferentiates the manner in which markers are found
rick 2001). Results in general indicate that gain from (i.e., via association genetics) and utilized from MAS
MAS is possible in P. radiata for a range of options (above).
but marginal for Douglas fir, implying rotation length Because association genetics is relatively new to
and product value are both important. For example, forest trees (Neale and Savolainen 2004; Wilcox and
Wilcox et al. (2001b) showed that even modest gains in Burdon 2006; Wilcox et al. 2006) – and many other
physical traits of 3.0 to 3.4% resulted in product value plant species (Flint-Garcia et al. 2003) – very few quan-
gains in excess of 9% and internal rates of return rang- titative analyses have been undertaken to date that
ing from 9.1 to 21%. However, a key condition here involve detailed specific strategies for incorporation
was the need to achieve adequate multiplication rates into tree breeding programs. Wilcox et al. (2006) de-
via vegetative propagation. Propagation technologies scribed a range of applications within tree breeding
are key to MAS being cost-effective. Johnson et al. programs, which are summarized below.
(2000) showed that modest genetic and financial gains Conifer breeding programs can be generically
were possible for MAS in Douglas fir for production characterized as consisting of highly heterozygous
population applications, although results were highly genotypes in hierarchically arranged (and managed)
dependent upon assumed genetic architectures and populations, with multitrait breeding objectives. To be
relatively large areas of plantations were needed to effective, marker-trait associations derived from as-
justify the extra costs associated with MAS. sociation genetics must integrate within this scheme,
Given all of the above, it is not surprising that as with MAS. A key generic benefit of GAS in this
relatively few breeding programs are actively pur- regard is application to both within- and among-
suing MAS, particularly for quantitatively inherited family selection, in contrast to MAS, which is re-
68 C. Plomion et al.

stricted to those families in which the associations breeding programs. We expect this will change as re-
have been detected. Association genetics, therefore, sults from association genetics and LD experiments
has application to all levels of population hierarchies, become available.
from essentially unimproved germplasm, through to Indeed, the outlook from results to date is cau-
advanced-generation lines developed for clonal de- tiously favorable, as marker-trait associations have
ployment. been reported for a number of tree species, includ-
A further benefit would be selection at the seedling ing P. taeda (Brown et al. 2004), P. radiata (Cato et al.
stage, much sooner than full trait expression – as with 2006b), and Eucalyptus nitens (Thumma et al. 2005),
MAS. For most commercially important pine species, although all of these need to be independently veri-
phenotypic selection is applied at 6 to 12 years of age, fied. Results from a range of other plant species are
with the onset of reproductive competency occurring also emerging (see Gupta et al. 2005 for a review), also
slightly earlier. Opportunity exists for early selection with encouraging messages.
to either increase selection intensities via a multistage Given the size of gymnosperm genomes and the
approach as with MAS (Wu 2002) or undertake an lack of widespread genomic DNA sequence for most
additional round of selection prior to finalizing geno- hardwood species, it appears that sequences within
types for deployment (Wilcox et al. 2001b). Where on- and associated with expressed genes are likely to be
set of reproductive maturity precedes trait expression, investigated. But how should such sequences be cho-
there is also opportunity to reduce breeding-cycle sen? Possible approaches are as follows.
length, although this would require known marker-
– Identifying gene sequences using EST libraries
trait associations for at least the majority of traits in
from both angiosperms and conifers, and complete
the breeding objective(s). The opportunity for selec-
genomic sequences from model angiosperm
tion among as well as within families provides even
species to select numerous genes for high-
greater opportunity to shorten breeding cycles com-
throughput (re)sequencing. This approach is being
pared to MAS.
used in P. taeda (see [Link]
An additional benefit is cost reduction relative to
adept2).
phenotypic selection. Once marker-trait associations
– Expression information at the mRNA level, via in
are detected, genotyping costs are generally some-
silico expression profiling or application of a num-
what lower as compared to phenotypic evaluation.
ber of gene expression profiling technologies.
Such evaluations, however, are considerably more ex-
Protein-expression information is also available
pensive than obtaining sufficient markers for genome
for some species.
coverage for linkage and QTL mapping as well as for
– Selecting genes on the basis of biochemical role(s)
MAS purposes, as virtually all polymorphisms within
and/or known sequence motifs.
and associated with expressed genes need to be de-
– Information from colocalization of expressed genes
tected – which, on a whole-transcriptome basis, en-
and QTL from pedigreed QTL mapping popula-
tails major effort. Such information nonetheless has
tions, within either the species of interest or related
other potential applications (e.g., prescreening genes
species.
for genetic modification experiments), so the costs
– Combinations of the above, for example gene-
could be spread over several funding sources and ap-
or protein-expression information combined with
plications.
QTL mapping (Kirst et al. 2004).
Because of the breadth of potential selection ap-
– Genes shown to be involved in trait variation via
plications within any one breeding program, as well
genetic modification experiments.
as the fact that conifer breeding programs all differ in
some regard from one another, the application of as- Technical advances in “omics” technologies may also
sociation genetics will need case-by-case evaluation – offer a number of other approaches.
as has been the case with MAS – ultimately needing The undoubted importance of comparative ge-
numerical simulation to evaluate different scenarios. nomics, especially for establishing gene function,
However, although a number of breeding programs means that genomic information is needed on other
are involved in some of the basic research required species, in addition to the classical model species like
for detecting marker-trait associations, as yet there Arabidopsis. This need is actually being met, to some
are no published reports of evaluations of strategies degree at least. In conifers, parallel research is pro-
that incorporate association genetics within conifer ceeding in P. radiata in Australasia, P. taeda in the
 Chapter 2 Pines 69

southeastern USA, and P. pinaster in Europe. The ev- 2.5

idently close synteny among these species promises Genomics Resources
great opportunities to apply genomic information on for the Genus Pinus
any one species to the others. Using these as model
species within conifers should allow one to capital-
ize on the most favorable model features of all the 2.5.1
species at once. Similarly, whole-genome sequences Efforts Toward Complete Genome Sequencing
of Populus and Eucalyptus will allow for even more
rapid advances in these species. BAC Libraries
Should association genetics be utilized by tree Large-insert genomic DNA libraries in which each
breeders, then a number of other requirements need DNA clone is stored and archived individually (i.e.,
to be addressed. These include: ordered libraries) are a fundamental tool in mod-
ern genomics research. The most popular large-insert
– Populations and analytical methods for adequate
vector is the bacterial artificial chromosome (BAC),
detection of marker-trait association (see Ball
which, despite its name, is not a chromosome but
2005),
rather a modified bacterial plasmid (F-factor) that
– Suitable laboratory and bioinformatics capabilities
can stably carry large inserts (ca. 50 to 400 kb) (Zhang
functionally integrated into tree-improvement pro-
and Wu 2001). BAC libraries have been made for a host
grams, and
of taxa and employed in a variety of applications (re-
– Development of strategic alliances among appro-
viewed by Zhang and Wu 2001).
priate entities to effectively manage resource asso-
Despite the importance of pine and other
ciation genetics programs in a manner that allows
conifers and the value of BAC libraries to genomic
operational implementation.
research, BAC resources are extremely limited for
The above issues are discussed in more detail by gymnosperms. To our knowledge, there are three
Wilcox et al. (2006). gymnosperm BAC libraries in existence – all of them
for pine. The first pine BAC library, constructed by
2.4.5 Islam-Faridi et al. (1998), affords 0.05× coverage of
Summary of Applications the P. taeda genome. A second library, developed
of DNA Polymorphisms in Conifer Breeding by Claros et al. (2004) for P. pinaster, provides
considerably higher genome coverage (0.32×),
Information from DNA polymorphisms has a wide although the probability of finding a gene of interest
range of applications for tree improvement, including is still relatively low (27%). However, the P. pinaster
quality control, pedigree reconstruction, elucidation library is reportedly being expanded to 3× (Claros
of genetic phenomenon, monitoring and maintenance et al. 2004), which would provide 95% probability
of genetic diversity, and selection and breeding based of finding a given genome sequence (Fig. 3). In
upon polymorphisms associated with trait variation. September 2004, the US National Science Foundation
To date, only some of these applications have been im- funded a 3-year project that included construction
plemented in operational breeding programs – largely of a 10× BAC library from the P. taeda genotype
those associated with quality control. Some potential “7-56” (Peterson et al. 2005) as a primary objective
applications, such as association genetics and pedi- (Fig. 4). When completed, this library will be, to
gree reconstruction as part of operational testing, are our knowledge, the single largest BAC library ever
largely in the research and development phase, while made with roughly 1.7 to 2 million individual clones.
others such as within-family MAS have been suffi- As of this writing, the 7-56 BAC library affords
ciently evaluated to identify at least some strategies for 2× coverage of the pine genome. When the 7-56
implementation. The lack of uptake across the spec- library reaches 3× coverage (ca. June 2006), it will be
trum of potential applications is likely due to cost, gridded onto BAC macroarrays and screened with
which is particularly important in tree-improvement STS and molecular genetic markers including those
programs, which usually take years to recoup such associated with genes of economic importance. The
costs. Nonetheless, technological advances will in- resulting information will be utilized to isolate and
crease the scope of applications for tree improve- sequence intact pine genes. Additionally, marker
ment. localization will represent the start of STS-based
70 C. Plomion et al.

Fig. 3. The relationship between genome coverage (×) provided by a BAC library and the probability (p) of finding a particular
sequence of interest in that library. A 0.05× BAC library, like the P. taeda BAC library generated by Islam-Faridi et al. (1998),
affords only a 4.8% chance (p = 0.048) of finding a sequence of interest (triangle), while a 0.32× library, like the current P. pinaster
library (Claros et al. 2004), affords a considerably better chance (p = 0.27; circle). The P. taeda 7-56 library (Peterson et al. 2006)
currently has a size of 2× (p = 0.86; four-point star) and will be used to isolate genes of interest once it reaches 3× coverage (p
= 0.95; five-point star). At 10× coverage (p = 0.9999; eight-point star), the P. taeda 7-56 BAC library will be of adequate size for
essentially any use including genome-wide physical mapping and genome sequencing

physical mapping in pine. Comparison between the – Genome structure: Very little is known about the
P. pinaster and P. taeda libraries should provide structure of any conifer genome, and relationships
considerable insight into genome evolution within between genes, repeats, and pseudogenes have only
Pinus. been tangentially explored (e.g., Elsik and Williams
Need for Pine Genomic Sequence 2000; Rabinowicz et al. 2005 ; Peterson et al. 2006).
Although EST sequencing in pine is relatively ad- BAC and shotgun sequencing of pine DNA will pro-
vanced (Sect. 2.5.2), there are few genomic sequence vide a more detailed understanding of the sequence
data available for pine or any other gymnosperm. structure of the pine genome. Such information will
As of 6 March 2006, the longest continuous gym- be essential in the development of an efficient strat-
nosperm/conifer genomic sequence in GenBank was egy for sequencing the gene space of pine.
only 6,884 bp. While ESTs provide the coding regions – Differential gene expression: The lack of genomic
of many expressed genes, an understanding of gene sequence for pine has severely limited the study
function and regulation requires knowledge of those of the noncoding regulatory regions of its genes.
noncoding sequences that coordinate gene expression Information on these regions is essential if we
in response to biotic and abiotic cues (e.g. promoters, are to understand differential gene expression in
enhancers, silencers). Such information necessarily pine and manipulate conifer genes in a useful and
comes from sequencing genomic DNA. controllable manner (No et al. 2000). To deter-
As the pine genome is ten times larger than that mine the general tissue/development/environment
of maize (the largest plant genome to be the target of specificity of the regulatory sequences for a given
a full-scale genome sequencing effort, NSF 2005), it gene, the expression profiles of that gene can be
may be a while before whole-genome sequencing of examined; since there is already a considerable
pine becomes a reality. Nonetheless, pine genomics amount of expression profile data available (e.g.,
can be greatly advanced by even relatively modest se- Egertsdotter et al. 2004; Lorenz et al. 2006), asso-
quencing of genomic DNA, especially sequencing of ciating regulatory sequences with specific tissues
large continuous pieces of DNA such as BAC inserts. and/or developmental events could be initiated im-
The following are just a few areas of pine genome re- mediately after sequencing. Comparison of regula-
search that will benefit from more extensive sequenc- tory sequences from genes with similar expression
ing of genomic DNA. patterns can be used to gain insight into the ac-
 Chapter 2 Pines 71

Fig. 4. Constructing the Pinus taeda 7-56

BAC library. A Genetix QPixII robot picks 7-56
BAC colonies off of agar in a “Q-Tray.” Picked
clones are used to inoculate media in 384-well
microtiter plates (left side of image)

tivation/repression mechanisms underlying major pine genes (e.g., Krutovsky et al. 2004; Le Dantec
developmental events. et al. 2004). However, sequence changes in the non-
– Comparative genomics of conifers: Pine coding regulatory regions of pine genes may pro-
BAC/shotgun sequences and comparative ge- vide as much, if not more, information on pheno-
nomic approaches can be utilized to advance the typic diversity than coding sequence.
study of other gymnosperm genomes, which will – Association mapping: Relatively long DNA
afford considerable insight into the evolution of stretches produced by BAC sequencing will facil-
this important group of organisms. Comparison itate genomewide studies of LD, which, in turn,
of orthologous regions of conifer genomes (e.g., will fuel association mapping projects (e.g., Brown
through sequencing of BACs and/or BAC contigs et al. 2004; Krutovsky and Neale 2005).
recognized by common markers) will provide – Sequencing adaptive and economic genes of pine:
high-resolution means of investigating conifer There is considerable interest in sequencing intact
sequence evolution (see Stirling et al. 2003 and Yan genes of economic importance and/or particular
et al. 2004 for angiosperm examples). value to conifer/gymnosperm research. Potential
– Gymnosperm/angiosperm comparisons: Infor- target genes include those associated with outcross-
mation garnered from pine genomic sequence ing (Williams et al. 2001), embryogenesis (Ciavatta
should provide nearly limitless opportunities to et al. 2001), disease resistance/susceptibility (e.g.,
explore comparative genome evolution between Devey et al. 1995; Morse et al. 2004; Kayihan et al.
gymnosperms and angiosperms. BAC sequences 2005), abiotic stress resistance (e.g., Chang et al.
from pines can be compared with existing plant 1996; Dubos and Plomion 2003; Dubos et al. 2003 ;
genome sequences as a means of evaluating macro- Krutovsky and Neale 2005; Lorenz et al. 2006), and
and microsynteny as well as sequence conserva- wood properties (Brown et al. 2003; Devey et al.
tion/divergence between pines and angiosperms. 2004a; Krutovsky and Neale 2005; Pot et al. 2005b).
As pine has been widely used as an outgroup in Complete gene sequences can be utilized to explore
angiosperm comparative research, pine genomic pine/conifer evolution and facilitate pine improve-
sequences should enable investigations by many ment through traditional breeding, marker-aide se-
angiosperm research groups throughout the lection, and genetic engineering. Of note is the fact
world. that a BAC containing a target gene may well contain
– Polymorphism discovery and characterization: EST other genes of interest. The study of these “bonus”
sequencing has afforded tremendous insight into genes will likely be as important as studying target
polymorphisms within the coding sequences of genes themselves.
72 C. Plomion et al.

– Exploring the evolution of repeat sequences: Re- Sequencing the Pine Genome
cently a 454 Life Sciences Genome Sequencer 20
(Margulies et al. 2005) was used to generate 100 Mb C. Plomion: Will the pine genome be sequenced?
of sequence from the P. taeda genotype 7-56 (Pe- D.G. Peterson: Yes.
terson et al. 2006). This sequence should provide C. Plomion: When?
considerable insight into pine genome structure, D.G. Peterson: Not this week.
although the short length of 454 reads (ca. 100 bp
each) limits the utility of the sequence. While pseu- The complete sequencing and assembly of the Ara-
domolecule contigs have been assembled, 100 Mb bidopsis and rice genomes have afforded considerable
affords only 0.046× coverage of the pine genome, insight into the evolution of higher plants including
and consequently it is likely that any 454 contig of pines. For example, Kirst et al. (2003) compared the
significant length is a conglomeration of sequences genome sequence of Arabidopsis with pine ESTs and
in a repeat sequence family rather than an individu- showed that gymnosperms and angiosperms pos-
al element found in native DNA. However, repeat sess highly similar gene complements. Other plant
sequences obtained from sequenced BACs and/or genomes, most notably poplar, sorghum (Sorghum
contigs assembled from Sanger/capillary shotgun bicolor L.), and maize (Zea mays L.), are current
reads can presumably be used as scaffolds on which subjects of full-scale genome sequence efforts, and
454 reads can be aligned. The depth and shape of the the genome sequences of these plants will further
resulting alignments should permit detailed analy- facilitate understanding of plant genome evolution
sis of the evolution of major repeat-sequence fami- and function (Paterson et al. 2005). The poplar se-
lies in pine. quence, in particular, may provide information that
– Genetic engineering: Regulatory sequences will help advance understanding of wood formation
obtained from genomic sequences can be at- in trees (Tuskan et al. 2004). However, angiosperms
tached to reporter genes and used to study and gymnosperms diverged from a common ances-
promoter/regulator specificities (e.g., No et al. tor more than 300 million years ago (Bowe et al. 2000),
2000). Such testing will provide insight into pine and consequently the utility of angiosperm sequences
gene regulation, which will eventually enable in the study of pine and other gymnosperms will
insertion of constructs containing pine promoters be limited. Ultimately, the best means of advancing
and genes/alleles of value into conifers. In addition, pine and conifer genomics is complete sequencing of
the genomic sequences can serve as sources of a conifer genome. While it is now theoretically fea-
molecular markers that can be utilized in tree sible to sequence the genome of any organism, the
improvement through MAS. large, repetitive nature of conifer genomes will likely
prevent them from being targets of full-scale genome
Physical Mapping sequencing for at least a few years. However, the like-
Given the enormous size of the pine genome and the lihood that the pine genome will be sequenced is high
time and financial resources required to conduct phys- based upon past and present investments made by
ical mapping research, it is likely that complete phys- the NSF (including construction of the P. taeda 7-56
ical mapping of the pine genome will be too costly to BAC library) and other granting agencies and growing
pursue, at least with present technology. Because DNA worldwide interest in pine as a biofuel/carbon seques-
sequencing technologies are advancing rapidly (e.g., tration crop (Jackson and Schlesinger 2004; Perlack
Margulies et al. 2005), it is probable that complete et al. 2005).
shotgun sequencing of the pine genome will become While whole-genome sequencing may have
affordable before complete physical mapping does. to wait, sequencing of gene-rich BACs is likely
However, physical mapping of gene-rich genomic re- to begin in relatively short order. Additionally,
gions has been utilized to advance understanding of reduced-representation sequencing (RRS) methods
important chromosomal regions in many species (e.g., (see Peterson 2005 for review) are being used to
Folkertsma et al. 1999; Sanchez et al. 1999; Dilbirligi investigate sequence subsets of the pine genome and
et al. 2004; Barker et al. 2005), and indeed, physical are likely to play a role in eventual whole-genome
mapping of gene-rich regions is a goal of the re- sequencing as well. Two RRS techniques that have
search group constructing the P. taeda 7-56 BAC li- been successfully utilized in maize genome ex-
brary. ploration are methylation filtration (MF) and Cot
 Chapter 2 Pines 73

filtration (CF)1 (Whitelaw et al. 2003; Springer et al. Lamoureux et al. 2005) or used directly as a pyrose-
2004). quencing substrate. With regard to the latter op-
tion, Peterson et al. (2006) recently sequenced Cot-
– MF is based on the preferential hypermethylation
filtered P. taeda 7-56 DNA including isolated highly
of retroelements and hypomethylation of genes ob-
repetitive, moderately repetitive, and low-copy se-
served in some plants (Rabinowicz et al. 1999). In
quences using the new 454 Life Sciences Genome
MF, genomic DNA fragments are ligated into a plas-
Sequencer 20 (see Margulies et al. 2005 for details
mid containing an antibiotic-resistance gene and
on this instrument). These data are currently being
the resulting recombinant molecules are used to
evaluated, although initial results look promising.
transform a bacterial strain containing enzymes
CF has proven highly effective in wheat, where it
that preferentially cleave hypermethylated DNA.
provides 19.5-fold enrichment for low-copy DNA
Linearization of a hypermethylated recombinant
(Lamoureux et al. 2005).
molecule results in its loss and makes its host
cell susceptible to an antibiotic in the growth
2.5.2
medium. Consequently, only hypomethylated se-
Genomic Tools to Identify Genes
quences are successfully cloned. While a useful
of Economic and Ecological Interest
gene-enrichment tool in maize and sorghum (Ra-
binowicz et al. 1999; Bedell et al. 2005), MF has
proven considerably less effective when applied to Characterization of the Pine Gene Space
the “mega genomes” of pine and wheat (Rabinow- A large-scale expressed sequence tag (EST) sequenc-
icz et al. 2005). ing project can generate a partial sequence for a large
– CF is rooted in the principles of Cot analysis, the proportion of genes from a given organism (Adams
study of DNA reassociation in solution (Peterson et al. 1991). As only genic sequences are analyzed re-
et al. 2002). When sheared DNA is heated to 100 ◦ C, gardless of genome size, this approach is considered
the two complementary strands of each double he- the most cost-efficient strategy of gene discovery for
lix come apart in a process known as denatura- organisms whose genome sequences are not yet avail-
tion. If the denatured DNA is slowly cooled, com- able (Rudd 2003), especially for species, such as pines,
plementary DNA strands find each other and form with exceedingly large genomes primarily composed
double helices (duplexes). The rate at which a par- of nongenic repetitive elements. For this reason, such
ticular DNA sequence finds a complementary se- projects have been initiated for many organisms in-
quence with which to pair is proportional to the cluding the numerous agronomically important plant
number of times that sequence (and hence its com- species (Rounsley et al. 1998). To rapidly scan for all
plementary sequence) is found within the genome. the protein coding genes and to provide a sequence
In other words, repetitive sequences renature more tag for each gene of the pine genome, large collections
quickly than low-copy sequences. In CF, sheared of ESTs has been developed in P. taeda and, to a lesser
genomic DNA is denatured and allowed to reas- extent, in P. pinaster.
sociate for a period of time in which only repet-
Loblolly pine ESTs Three major EST sequencing
itive DNA sequences are likely to form duplexes.
projects, mainly funded by the NSF, have been
The double-stranded repetitive DNA is then sepa-
conducted so far in P. taeda. A first project, “Ge-
rated from single-stranded low-copy DNA using hy-
nomics of wood formation in Loblolly pine,” was
droxyapatite chromatography. The low-copy DNA
concerned with gene discovery in wood formation
and/or the repetitive DNA can then be cloned and
([Link] It resulted in ca.
sequenced (Peterson et al. 2002; Yuan et al. 2003;
60,000 ESTs placed in the public domain. A second
1 Cot filtration was originally called “Cot-Based Cloning and
project, entitled “Transcriptome responses to envi-
ronmental conditions in Loblolly pine roots,” aimed
Sequencing” (CBCS; Peterson et al. 2002) and later “high
Cot” (HC; Yuan et al. 2003) sequencing . However, the term
at completing this information by targeting especially
“Cot filtration” and the acronym CF are becoming more roots undergoing a variety of biotic and abiotic
widely used because (a) the acronym “CBCS” can also stand stresses ([Link]
for “clone-by-clone sequencing,” (b) “high Cot” is frequently It resulted in 140,000 additional ESTs. A third
confused with “high copy,” and (c) Cot components may be project, “Genomics of Loblolly pine embryo-
sequenced without prior cloning (Peterson et al. 2006). genesis,” seeks to identify and characterize
74 C. Plomion et al.

Fig. 5. Pine sequence

resources available at EMBL
([Link]/srs), on 28
February 2006

an additional 85,000 ESTs derived from so- unigene set of 45,500 genes ([Link]
matic and zygotic embryos ([Link] scripts/tgi/T_index.cgi?species=pinus), representing
tdb/e2k1/pine/[Link]). probably a major fraction of the coding genome of
pine. This resource has laid the foundation for fu-
Maritime pine ESTs Early molecular biology ture identification, characterization, and cloning of
studies paved the way for the construction of cDNA economically important genes in pines and should
libraries from different woody tissues (Cantón provide breeders with a wealth of information to sig-
et al. 1993; García-Gutiérrez et al. 1995), roots nificantly enhance and speed up the breeding pro-
(Dubos and Plomion 2003) and buds (C. Collada, cess.
M.A. Guevara, and M.T. Cervera, unpublished Besides this resource, a significant number of ESTs
results) of maritime pine. Funded by the European have not been released yet in the public domain and
Union (GEMINI project), INRA (lignome project), are mainly held by private companies. For instance,
and CNS (ForEST project), 28,000 high-quality Strabala (2004) reported that in August 2001, 344,279
sequences have been generated in about equal ESTs from P. radiata were sequenced by AgriGenesis
number from wood-forming tissues, roots, and buds. Biosciences Ltd. based in New Zealand.
These ESTs have been made available at the EMBL
database and at the following URL: [Link] Microarrays: New Nools to Study the Functioning
outils/SAM/COMPLETE/. The P. pinaster cDNA of the Pine Genome
clones are available for the scientific community, ESTs represent snapshots of the genes expressed in
at the Platform for Integrated Clone Management a given tissue or at a given developmental stage and
PICME Web site ([Link] A Web- provide a good representation of the expressed re-
based environment for assembly and annotation of gions of the genome if a sufficient number of different
collections of ESTs (Le Provost et al. in prep.) was libraries from a range of tissues and developmen-
used to analyze the P. pinaster ESTs. tal stages and after various environmental challenges
are analyzed. To this end, several cDNA libraries
A comprehensive unigene assembled for the user have been produced from pine roots, needles, dif-
community The number of pine ESTs in public ferentiating xylem, and buds (see previous section).
databases has increased dramatically during the past These genes have enabled the development of mi-
decade (Fig. 5), especially for P. taeda, ranking at croarrays to identify genes whose expression varies
the 19th position in dbEST with 329,469 ESTs, and in response to various environmental and develop-
to a lesser extent for P. pinaster (about 28,000 ESTs). mental cues (mainly drought-stress response and
The other pine species, P. elliottii, P. banksiana, P. pat- wood formation). These studies, which will likely
ula, and P. sylvestris, represent less than 2,000 se- increase in the future, are listed in Table 6 with
quences. In 2005, the Pinus genus was introduced others that have used other transcriptomic (SAGE,
in the TIGR gene index (Quackenbush et al. 2001), SSH, cDNA-AFLP) and proteomic (2D PAGE MS/MS)
where a comprehensive collection of all publicly avail- approaches to study the functioning of the pine
able pine ESTs has been clustered into a tentative genome.
Table 6. Transcriptome and proteome analysis in pines

Species Methodology Study Tissue Authors

P. contorta Microarrays Adventitious root development Roots Brinker et al. 2004

P. halepensis, P. taeda cDNA library, Northern blot, Drought response in two Roots, needles Sathyan 2005
RT-PCR pine species
P. pinaster cDNA-AFLP, reverse Northern Drought response Roots Dubos and Plomion 2003
P. pinaster cDNA-AFLP, reverse Northern Drought response Needles Dubos et al. 2003
P. pinaster cDNA library Bud phenology Buds at different developmental Collada, Guevara, and Cerera,
stages unpublished results
P. pinaster Six SSH cDNA libraries Wood formation Early vs. late, juvenile vs. mature Pacheco et al. 2005
wood-forming tissues.
P. pinaster cDNA-AFLP; reverse Northern; Wood formation Early vs. late, opposite vs. Le Provost et al. 2003
RT-PCR compression wood-forming tissues
P. pinaster cDNA library Wood formation Composite xylem tissues Cantón et al. 2003
P. pinaster EST sequencing, microarrays, Wood formation Early vs. late, juvenile vs. mature Paiva 2006
qPCR wood-forming tissues and
8 different organs
P. pinaster EST sequencing, macroarrays, Drought response in two ecotypes Roots Chaumeil 2006
qPCR, 2DE MS/MS
P. pinaster 2DE, tandem MS Wood formation Wood-forming tissues Gion et al. 2005
P. pinaster 2DE, micro-sequencing Drought response Needles Costa et al. 1998
P. pinaster 2DE, micro-sequencing Wood formation compression and opposite wood Plomion et al. 2000
forming tissues
P. pinea Two SSH cDNA libraries Cytokinine induction of Cotyledons Alonso, Cantón, Ordás,
adventitious buds unpublished results
P. sylvestris cDNA library Nitrogen and carbon assimilation Cotyledons Avila et al. 2000
P. radiata CAGE Wood formation Wood-forming tissues Cato et al. 2006a
P. radiata Three SSH cDNA libraries Adventitious rooting capacity Rooting competent tissues Sánchez et al. 2005
P. taeda Differential screening, Drought response Needles, stems, roots Chang et al. 1996
northern blot
P. taeda Microarrays Wood formation Wood-forming tissues Yang and Loopstra 2005
Chapter 2 Pines

P. taeda Microarrays Drought response and recovery Needles Watkinson et al. 2003
P. taeda Microarrays Drought response Needles, rooted cuttings, Heath et al. 2002
2 unrelated genotypes
75
 76

Table 7. (continued)

Species Methodology Study Tissue Authors

P. taeda Differential display Response to galled tissues Healthy and galled stem Warren and Covert 2004
P. taeda 0.35k microarrays; qRT-PCR Wood formation Unlignified early and latewood xylem: Egertsdotter et al. 2004
cambial region and developing
C. Plomion et al.

secondary nonlignified xylem

P. taeda cDNA and subtractive libraries; Wood formation Compression and opposite wood Allona et al. 1998
EST sequencing; Northern blot
P. taeda EST sequencing, microarray Wood formation Different tissues and organs, Whetten et al. 2001
including several types
of differentiating xylem samples
P. taeda SAGE profiling Wood formation Juvenile and mature wood formatting Lorenz and Dean 2002
tissues
P. taeda cDNA library; EST sequencing, Wood formation Six types of wood forming tissues Kirst et al. 2003
comparative sequence analysis
P. taeda Microarray, qRT-PCR Identification of genes preferentially Differentiating xylem, needles, Yang et al. 2004
expressed in P. taeda megagametophytes, embryo
(growing radicule)
P. taeda Comparative sequence analysis, Computational analysis of transcript Differentt typoes of wood-forming Pavy et al. 2005
digital profiling accumulation in xylem. tissues
P. taeda EST sequencing, in silico detection Drought response Roots Lorenz et al. 2005
of differentially expressed genes
P. taeda, P. sylvestris, Macroarrays Heterologous arrays Needles van Zyl et al. 2002
P. abies
P. taeda Microarrays Disease resistance Stem (infection court) Myburg et al. 2006
P. elliottii Microarrays Disease resistance Stem (infection court) Morse et al. 2004
var. eliottii
 Chapter 2 Pines 77

2.6 gions of the genome in LD, allowing for the tracking

Future Perspective: of most regions with mapped genetic markers (Lander
Challenges for Molecular Breeding and Botstein 1989). Once certain regions are identified
as causing (or at least being predictive of) significant
of Pines variation in the phenotype, marker alleles for these
regions can be selected for two or more successive
Pines offer a great opportunity for molecular breed- generations, thereby increasing the frequency of fa-
ing given their large size and long generation interval. vorable alleles in the population (Edwards and Page
The large size causes inefficiencies in progeny testing 1994; Hospital et al. 2000). One such view of this for
and concomitant decreases in heritability of traits. pine breeding is termed marker-directed population
Molecular-marker information can reduce these inef- improvement (MDPI) (Nelson and Echt 2003, 2004).
ficiencies and effectively increase heritability, leading Nonpedigreed populations provide an opportu-
to more accurate selections and commensurate in- nity for tracking very small regions across the genome
creases in genetic gain (Fernando and Grossman 1989; (Hirschhorn and Daly 2005). These are the regions
Lande and Thompson 1990; Hospital et al. 1997). The that have remained in LD for a very long time, pos-
long generation intervals cause delays in obtaining sibly since the creation of the DNA polymorphism
genetic gain. Savings of many years in the breeding used as the genetic marker or the gene mutation it-
cycle can easily be envisioned if selections can be self. Extremely dense genetic maps are required for
based on DNA marker information collected at a very mapping QTLs in nonpedigreed populations through
early age as opposed to phenotypes measured much a genome-scan analysis (Kruglyak 1999). Testing spe-
later (Williams and Neale 1992). However, these gains cific genes known as candidate genes for association
can only be achieved if breeding can commence very to phenotype does not require a genetic map (Pflieger
soon after the selections are made. This has tradi- et al. 2001; Neale and Savolainen 2004). Instead, a hy-
tionally been known as accelerated breeding, and it is pothesis about what gene(s) might affect a trait’s phe-
a critical, complementary technology (e.g., Bramlett notype is needed as well as markers that allow detec-
and Burris 1995). tion of alternative alleles of the candidate gene(s) in
While the opportunity for increased genetic gain the population (Krutovsky and Neale 2005).
per year is great, major challenges to implement The greatest opportunities for implementing
molecular breeding include pines’ highly heterozy- molecular breeding exist in breeding programs
gous, outbreeding nature and the presence of low that utilize best testing practices, including good
levels of LD in natural populations (Strauss et al. 1992; experimental design, careful field site selection and
Brown et al. 2004). Each of these factors limits the op- maintenance, and high-quality measurements on all
portunity for finding QTLs by association (Kruglyak important traits (e.g., Stuber et al. 1999; Tanksley and
1999; Hirschhorn and Daly 2005) or candidate-gene Nelson 1996). Most of these programs will contain
(Pflieger et al. 2001; Neale and Savolainen 2004) map- populations with pedigree structure, allowing for
ping approaches, and for extending QTL results from moderately dense genetic maps (10 to 20 cM spacing)
one pedigree to another within a breeding population. to be used in various linkage-mapping approaches
However, given very dense genetic maps or good leads (Darvasi et al. 1993) including MDPI. In addition, all
about possible candidate genes, these low LD levels trees are drawn from a source population where tight
allow for the possibility of discovering tight linkages linkages will be maintained over short recombination
to important genes. These close linkages should be intervals, allowing for mapping by association
generally useful across the population for predicting through a genome-scan or candidate-gene approach
genetic value and implementing molecular breeding. (Wu et al. 2002; Lund et al. 2003).
The challenge in moving forward lies in the devel- Where no breeding program exists, large ran-
opment of powerful genetic markers and maps and dom mating populations could be established specifi-
populations providing genetic information on traits cally for association mapping and subsequent marker-
of interest (e.g., Georges et al. 1995; Farnir et al. 2002; assisted selection and breeding. These would most
Laurie et al. 2004). These populations can include both likely use candidate-gene methods, as candidate genes
pedigreed and nonpedigreed populations, and both could be selected from first principles or knowledge
are valid for the application of molecular breeding. of related species and then tested for association with
Pedigreed populations place and maintain large re- phenotype in the target species and population. It
78 C. Plomion et al.

would seem that genome-scan approaches will only be Analysis of xylem formation in pine by cDNA sequencing.
viable in very intensively studied species where dense Proc Natl Acad Sci USA 95:9693–9698
genetic maps will be developed. In all cases, highly ef- Andrew RL, Peakall R, Wallis IR, Wood JT, Knight EJ, Foley WJ
ficient DNA isolation and genotyping methods (e.g., (2005) Marker-based quantitative genetics in the wild?: The
Darvasi and Soller 1994; Mosig et al. 2001) and effec- heritability and genetic correlation of chemical defenses in
Eucalyptus. Genetics 171:1989–1998
tive database management and bioinformatic tools
Apiolaza LA, Greaves BL (2001) Why are most breeders not
will be required for cost-effective implementation
using economic breeding objectives. In: IUFRO Symp “De-
(Nelson 1997). Expected results include increased ge- veloping the Eucalypt for the Future”, 10-14 Sept 2001,
netic gain per generation due to increased effective Valdivia, Chile
heritability and with accelerated breeding, increas- Arcade A, Faivre-Rampant P, Le Guerroue B, Paques LE, Prat D
ing the number of generations per unit of time. In (1996) Quantitative traits and genetic markers: analysis of
addition, traits that are difficult to measure can be a factorial mating design in larch. In: Ahuja MR, Boerjan
improved more readily assuming they are measured W, Neale D (eds) Somatic Cell Genetics and Molecular
and mapped in an early generation. Genetics of Trees. Kluwer, Dordrecht, pp 211–216
Arumuganathan K, Earle ED (1991) Nuclear DNA content of
some important plant species. Plant Mol Biol Rep 9:208–218
Acknowledgement. This work was funded in part by (A) ANR
Atwood RA, White TL, Huber DA (2002) Genetic parameters
awards GENOQB (GNP05013) and DIGENFOR (ANR-05-
and gains for growth and wood properties in Florida source
GPLA-006-01) to C Plomion, (B) NSF award DBI-0421717 to
loblolly pine in the southeastern United States. Can J For
DG Peterson, CD Nelson, and MN Islam-Faridi, (C) award
Res 32:1025–1038
RTA03-213 from the Programa Nacional de Recursos y
Auckland L, Bui T, Zhou Y, Shepherd M, Williams CG (2002)
Tecnologías Agroalimentarias (Ministerio de Educación y
Conifer Microsatellite Handbook. Corporate Press, College
Ciencia, MEC) to MT Cervera, (D) National Research Council
Station, TX
of Italy (Commessa: “Evoluzione e analisi della diversità
Avila C, Muñoz-Chapuli R, Plomion C, Frigerio J-M, Cánovas
genetica in piante forestali”) to GG Vendramin, (E) New
FM (2000) Two genes encoding distinct cytosolic glutamine
Zealand Foundation for Research Science and Technology
synthetases are closely located in the pine genome. FEBS
(FRST) contract CO4X0207 to Scion (P.L. Wilcox and R.D.
Lett 477:237–243
Burdon), (F) the Finnish Funding Agency for Technology and
Bahrman N, Damerval C (1989) Linkage relationships of
Innovation and by the Biosciences and Environment Research
loci controlling protein amounts in maritime pine (Pinus
Council of Finland, to O Savolainen.
pinaster Ait). Heredity 63:267–274
Ball RD (2001) Bayesian methods for quantitative trait loci
References mapping based on model selection: approximate anal-
ysis using the Bayesian Information Criterion. Genetics
159:1351–1364
Adams MD, Kelley JM, Gocayne JD, Dubnick M, Polymeropou-
Ball RD (2005) Experimental designs for reliable detection of
los MH, Xiao H, Merril CR, Wu A, Olde B, Moreno RF,
linkage disequilibrium in unstructured random population
Kerlavage AR, McCombie WR, Venter JC (1991) Comple-
association studies. Genetics 170:859–873
mentary DNA sequencing: expressed sequence tags and
Barker CL, Donald T, Pauquet J, Ratnaparkhe MB, Bouquet A,
human genome project. Science 252:1651–1656
Adam-Blondon AF, Thomas MR, Dry I (2005) Genetic and
Aimers-Halliday J, Shelbourne CJA, Hong SO (1997) Issues in
physical mapping of the grapevine powdery mildew resis-
developing clonal forestry with P. radiata. In: Burdon RD,
tance gene, Run1, using a bacterial artificial chromosome
Moore JM (eds) “IUFRO ’97 Genetics of Radiata Pine”, Proc
library. Theor Appl Genet 111:370–377
IUFRO Conf 1-4 Dec and Workshop 5 Dec, Rotorua, New
Beaulieu J, Plourde A, Daoust G, Lamontagne L (1996) Ge-
Zealand. FRI Bull 203:264–272
netic variation in juvenile growth of Pinus strobus in
Alazard P (2001) Genetic gain in seeds from first generation
replicated Quebec provenance-progeny tests. For Genet
seed orchards of maritime pine. Informations Forests, Afo-
3:103–112
cel, France, No 628
Beavis WD (1994) The power and deceit of QTL experiments:
Alía R, Gil L, Pardos JA (1995) Performance of 43 Pinus pinaster
lessons from comparative QTL studies. In: Proc 49th Annu
provenances on 5 locations in Central Spain. Silvae Genet
Corn and Sorghum Industry Res Conf, ASTA, Washington,
44:75–81
DC, pp 250–266
Alía R, Moro J, Denis JB (1997) Performance of Pinus pinaster
Bedell JA, Budiman MA, Nunberg A, Citek RW, Robbins D, Jones
Ait. provenances in Spain: interpretation of the genotype-
J, Flick E, Rholfing T, Fries J, Bradford K, McMenamy J,
environment interaction. Can J For Res 27:1548–1559
Smith M, Holeman H, Roe BA, Wiley G, Korf IF, Rabinowicz
Allona I, Quinn M, Shoop I, Swope K, ST-Cyr S, Carlis J, Rield
PD, Lakey N, McCombie WR, Jeddeloh JA, Martienssen
J, Retzel E, Campbell M, Sederoff R, Whetten RW (1998)
 Chapter 2 Pines 79

RA (2005) Sorghum genome sequencing by methylation Burdon RD (2002) An introduction to pines. In: Pines of Silvi-
filtration. PLoS Biol 3:13 cultural Importance. CABI, Wallingford, UK pp x–xxi
Bell JC, Powell M, Devey ME, Moran GF (2004) DNA profiling, Burdon RD (2004) Breeding goals: Issues of goal-setting and
pedigree lineage analysis and monitoring in the Australian applications. In: Walter C, Carson M (eds) Plantation For-
breeding program of radiata pine. Silvae Genet 53:130–134 est Biotechnology for the 21st Century, Research Signpost,
Blada I (1994) Inter-specific hybridization of Swiss Stone Pine Kerala, India, pp 101–118
Pinus cembra L. Silvae Genet 43:14–20 Burdon RD, Bannister MH, Low CB (1992) Genetic Survey of
Borzan Z, Papes D (1978) Karyotype analysis in Pinus: contri- Pinus radiata 4: variance structures and heritabilities in
bution to the standardization of the karyotype analysis and juvenile clones. NZ J For Sci 22:187–210
review of some applied techniques. Silvae Genet 27:144–150 Burdon RD, Firth A, Low CB, Miller MA (1998) Multi-site prove-
Bowe LM, Coat G, dePamphilis CW (2000) Phylogeny of seed nance trials of Pinus radiata in New Zealand. In: Forest
plants based on all three genomic compartments: Extant Genetic Resources 26:3–8, FAO, Rome
gymnosperms are monophyletic and Gnetales’ closest rel- Burdon RD, Gea LD (2006) Pursuit of genetic gain and biotic
atives are conifers. Proc Natl Acad Sci USA 97:4092–4097 risk management for Pinus radiata in New Zealand. In:
Bradshaw HD, Foster GS (1992) Marker-aided selection and Proc 13th Australasian Plant Breed Conf, Christchurch,
propagation systems in trees: advantages of cloning New Zealand pp 75–85, [Link]
for studying quantitative inheritance. Can J For Res Burdon RD, Kumar S (2003) Stochastic modelling of the impact
22:1044–1049 of four generations of pollen contamination in unpedigreed
Bramlett DL, Burris LC (1995) Topworking young scions into gene resources. Silvae Genet 52:1–7
reproductively-mature loblolly pine. In: Weir RJ, Hatcher Burdon RD, Kumar S (2004) Forwards versus backwards se-
AV (eds) Proc 23rd South For Tree Improv Conf, Asheville, lection: trade-offs between expected genetic gain and risk
NC, pp 234–241 avoidance. NZ J For Sci 34:3–21
Brendel O, Pot D, Plomion C, Rozenberg P, Guehl JM (2002) Burdon RD, Low CB (1992) Genetic survey of Pinus radiata.
Genetic parameters and QTL analysis of ä13C and ring 6: Wood properties: variation, heritabilites, and interrela-
width in maritime pine. Plant Cell Environ 25:945–953 tionship with other traits. NZ J For Sci 22:228–245
Brettschneider R (1998) RFLP analysis. In: Karp A, Isaac PG, In- Burdon RD, Miller JT (1992) Introduced forest trees of New
gram DS (eds) Molecular Tools for Screening Biodiversity Zealand: Recognition, role and seed source. 12. Radiata
Plants and Animals. Chapman & Hall, London, pp 83–95 pine (Pinus radiata D. Don). NZ For Res Inst Bull 124/12
Brinker M, van Zyl L, Liu WB, Craig D, Sederoff RR, Clapham pp 59
DH, von Arnold S (2004) Microarray analyses of gene ex- Byun KO, Kim MZS, Shim SY, Hong SH, Sohn SI (1989) Review
pression during adventitious root development in Pinus of pitch- loblolly hybrid pine (Pinus rigida and P. taeda)
contorta. Plant Physiol 135(3):1526-1539 breeding researches in Korea and future strategy. Res Rep,
Brown GR, Kadel III EE, Bassoni DL, Kiehne KL, Temesgen B, For Genet Res Inst, Korea 25:204–211
van Buijtenen JP, Sewell MM, Marshall KA, Neale DB (2001) Cahalan CM (1981) Provenance and clonal variation in growth,
Anchored reference loci in loblolly pine (Pinus taeda L.) branching and phenology in Picea sitchensis and Pinus
for integrating pine genomics. Genetics 159:799–809 contorta. Silvae Genet 30:40–46
Brown GR, Bassoni DL, Gill GP, Fontana JR, Wheeler NC, Cai Q, Zhang D, Liu ZL, Wang XR (2006) Chromosomal localiza-
Megraw RA, Davis MF, Sewell MM, Tuskan GA, Neale DB tion of 5S and 18S rDNA in five species of subgenus Strobus
(2003) Identification of quantitative trait loci influencing and their implications for genome evolution of Pinus. Ann
wood property traits in loblolly pine. III. QTL verification Bot 97:715–722
and candidate gene mapping. Genetics 164:1537–1546 Cánovas FM, Dumas-Gaudot E, Recorbet G, Jorrin J, Hans-Peter
Brown GR, Gill GP, Kuntz RJ, Langley CH, Neale DB (2004) Mock HP, Rossignol M (2004) Plant proteome analysis.
Nucleotide diversity and linkage disequilibrium in loblolly Proteomics 4:285–298
pine. Proc Natl Acad Sci USA 101:15255–15260 Cantón FR, García-Gutiérrez A, Gallardo F, de Vicente A, Cáno-
Burban C, Petit RJ (2003) Phylogeography of maritime pine vas FM (1993) Molecular characterization of a cDNA clone
inferred with organelle markers having contrasted inheri- encoding glutamine synthetase from a gymnosperm: Pinus
tance. Mol Ecol 12:1487–1495 sylvestris. Plant Mol Biol 22:819–828
Burdon RD (1997) Genetic diversity for the future: conservation Cantón FR, Le Provost G, Garcia V, Barré A, Frigerio J-M, Paiva
or creation and capture? In: Burdon RD, Moore JM (eds) J, Fevereiro P, Avila C, Mouret J-F, de Daruvar A, Cánovas
“IUFRO ’97 Genetics of Tadiata Pine”, Proc IUFRO Conf FM, Plomion C (2003) Transcriptome analysis of wood for-
1–4 Dec and workshop 5 Dec, Rotorua, New Zealand. FRI mation in maritime pine In: Espinel S, Barredo Y, Ritter E
Bull 203:263–274 (eds) Sustainable Forestry, Wood Products and Biotechnol-
Burdon RD (2001) Genetic diversity and disease resistance: ogy. DFA-AFA Press, Vitoria-Gasteiz, Spain, pp 333–347
some considerations for research, breeding and deploy- Cato S, Mc Millan L, Donaldson L, Richardson T, Echt C, Gard-
ment. Can J For Res 32:596–606 ner R (2006a) Wood formation from the base to the crown in
80 C. Plomion et al.

Pinus Radiata: gradients of tracheid wall thickness, wood Colbert T, Till BJ, Tompa R, Reynolds S, Steine MN, Yeung
density, radial growth rate and gene expression. Plant Mol AT, McCallum CM, Comai L, Henikoff S (2001) High-
Biol 60:565–581 throughput screening for induced point mutations. Plant
Cato SA, Pot D, Kumar S, Douglas J, Gardner RC, Wilcox PL Physiol 126:480–484
(2006b) Balancing selection in a dehydrin gene associated Conkle MT (1981) Isozyme variation and linkage in six conifer
with increased wood density and decreased radial growth species. In: Proc Isozymes of North American Forest Trees
in Pinus radiata (Abstr). In: Plant & Animal Genome XIV and Insects, USD Forest Service, General Technical Report
Conf, San Diego PSW-48, pp 11–17
Cato SA, Richardson TE (1996) Inter- and intra-specific poly- Costa P (1999) Réponse moléculaire, physiologique et géné-
morphism at chloroplast SSR loci and the inheritance tique du pin maritime à une contrainte hydrique. PhD the-
of plastids in Pinus radiata D. Don. Theor Appl Genet sis. Univ of Nancy
93:587–592 Costa P, Plomion C (1999) Genetic analysis of needle protein
Cervera MT, Plomion C, Malpica CA (2000a) Molecular mark- in Maritime pine. 2. Quantitative variation of protein ac-
ers and genome mapping in woody plants. In: Jain SM, cumulation. Silvae Genet 48:146–150
Minocha SC (eds) Molecular Biology of Woody Plants. Costa P, Pot D, Dubos C, Frigerio JM, Pionneau C, Bodenes C,
Kluwer, Dordrecht, pp 375–394 Bertocchi E, Cervera MT, Remington DL, Plomion C (2000)
Cervera MT, Remington D, Frigerio JM, Storme V, Ivens B, A genetic map of Maritime pine based on AFLP, RAPD and
Boerjan W, Plomion C (2000b) Improved AFLP analysis of protein markers. Theor Appl Genet 100:39–48
tree species. Can J For Res 30:1608–1616 Costa, P, Bahrman, N, Frigerio, J-M, Kremer A, Plomion C
Chagné D, Brown G, Lalanne C, Madur D, Pot D, Neale DB, (1998) Water-deficit-responsive proteins in maritime pine.
Plomion C (2003) Comparative genome and QTL map- Plant Mol Biol 38:587–596
ping between maritime and loblolly pines. Mol Breed Cornelius J (1994) Heritabilities and additive genetic coeffi-
12:185–195 cients of variation in forest trees. Can J For Res 24:372–379
Chagné D, Chaumeil P, Ramboer A, Collada C, Guevara A, Danjon F (1995) Observed selection effects on height growth,
Cervera M-T, Vendramin GG, Garcia V, Frigerio JM, Echt diameter and stem form in maritime pine. Silvae Genet
C, Richardson T, Plomion C (2004) Cross species transfer- 44(1):10-19
ability and mapping of genomic and cDNA SSRs in pines. Darvasi A, Soller M (1994) Selective DNA pooling for determi-
Theor Appl Genet 109:1204–12 nation of linkage between a molecular marker and a quan-
Chagné D, Lalanne C, Madur D, Kumar S, Frigerio JM, Krier C, titative trait locus. Genetics 138:1365–1373
Decroocq S, Savouré A, Bou-Dagher-Kharrat M, Bertocchi Darvasi A, Weinreb A, Minke V, Weller JI, Soller M (1993)
E, Brach J, Plomion C (2002) A high density genetic map of Detecting marker-QTL linkage and estimating QTL gene
Maritime pine based on AFLPs. Ann For Sci 59:627–636 effect and map location using a saturated genetic map.
Chambers PGS, Borralho NMG (1999) A simple model to ex- Genetics 134:943–951
amine the impact of changes in wood traits on the cost of Devey ME, Fiddler TA, Liu BH, Knapp SJ, Neale DB (1994)
thermomechanical pulping and high-brightness newsprint An RFLP linkage map for loblolly pine based on
production with radiata pine. Can J For Res 29:1615–1626 a three-generation outbred pedigree. Theor Appl Genet
Chang S, Puryear JD, Dias MADL, Funkhouser EA, Newton RJ, 88:273–278
Cairney J (1996) Gene expression under water deficit in Devey ME, Fino-Mix A, Kinloch BB Jr, Neale DB (1995) Ran-
loblolly pine (Pinus taeda): isolation and characterization dom amplified polymorphic DNA markers tightly linked
of cDNA clones. Physiol Planta 97:139–148 to a gene for resistance to white pine blister rust in sugar
Chaumeil P (2006) Plasticité moléculaire de deux écotypes de pine. Proc Natl Acad Sci USA 92:2066–2070
pin maritime soumis à un stress osmotique. Thèse de Doc- Devey ME, Bell JC, Smith DN, Neale DB, Moran GF (1996) A ge-
torat. Université Henri Poincaré, Nancy I, Nancy netic linkage map for Pinus radiata based on RFLP, RAPD,
Ciavatta VT, Morillon R, Pullman GS, Chrispeels MJ, Cairney J and microsatellite markers. Theor Appl Genet 6:673–679
(2001) An aquaglyceroporin is abundantly expressed early Devey ME, Sewell MM, Uren TL, Neale DB (1999) Compara-
in the development of the suspensor and the embryo proper tive mapping in loblolly and radiata pine using RFLP and
of loblolly [Link] Physiol 127:1556–1567 microsatellite markers. Theor Appl Genet 99:656–662
Claros MG, Cantón FR, Cánovas FM (2004) Construction Devey ME, Bell JC, Uren TL, Moran GF (2003) A set of mi-
of a pine BAC genomic library and bioinformatics plat- crosatellite markers for fingerprinting and breeding appli-
form for automated high-throughput sequence analy- cations in Pinus radiata. Genome 45:984–989
sis (Abstr). 2nd Annual Meeting of the Spanish For- Devey ME, Carson SD, Nolan MF, Matheson AC, Te Riini C,
est Functional Genomics Network. Pontevedra, Spain, Hohepa J (2004a) QTL associations for density and diam-
29 Nov 2004. [Link] eter in Pinus radiata and the potential for marker-aided
pontevedra_04/Gonzalo%[Link] selection. Theor Appl Genet 108:516–524
 Chapter 2 Pines 81

Devey ME, Groom KA, Nolan MF, Bell JC, Dudzinski MJ, Old Egertsdotter U, van Zyl LM, Mackay J, Peter G, Kirst M, Clark
KM, Matheson AC, Moran GF (2004b) Detection and ver- C, Whetten R, Sederoff R (2004) Gene expression during
ification of quantitative trait loci for resistance to Doth- formation of earlywood and latewood in loblolly pine: ex-
istroma needle blight in Pinus radiata. Theor Appl Genet pression profiles of 350 genes. Plant Biol 6:654–663
108:1056–1063 Elsik CG, Williams CG (2000) Retroelements contribute to the
Dhakal LP, White TL, Hodge GR (1996) Realized genetic gains excess low-copy-number DNA in pine. Mol Gen Genet
from slash pine (Pinus elliottii) tree improvement. Silvae 264:47–55
Genet 45:190–197 Elsik CG, Williams CG (2001) Low copy microsatellite recovery
Dilbirligi M, Erayman M, Sandhu D, Sidhu D, Gill KS (2004) form a conifergenome. Theor Appl Genet 103:1189–1195
Identification of wheat chromosomal regions containing Emebiri LC, Devey ME, Matheson AC, Slee MU (1997) Linkage
expressed resistance genes. Genetics 166:461–481 of RAPD markers to NESTUR, a stem growth index in
Doudrick RL, Heslop-Harrison JS, Nelson CD, Schmidt T, Nance radiata pine seedlings. Theor Appl Genet 95:119–124
WL, Schwarzacher T (1995) Karyotype of slash pine (Pi- Emebiri LC, Devey ME, Matheson AC, Slee MU (1998) Age-
nus elliottii var elliottii) using patterns of fluorescence related changes in the expression of QTLs for growth in
in situ hybridization and fluorochrome banding. J Hered radiata pine seedlings. Theor Appl Genet 97:1053–1061
86:289–296 Epperson BK, Allard RW (1989) Spatial autocorrelation anal-
Drewry A (1982) G-banded chromosomes in Pinus resinosa. J ysis of the distribution of genotypes within population of
Hered 73:305–306 lodgepole pine. Genetics 121:369–377
Du Toit B, Smith C, Carlson C, Esprey L, Allen R, Little K (1998) Falkenhagen E (1991) Provenanace variation in Pinus radiata
Eucalypt and pine plantations in South Africa. In: Nambiar at six sites in South Africa. Silvae Genet 40:41–50
EKS, Cossalter C, Triarks A (eds) Workshop Proceedings. Farjon A (1984) Pines. Drawings and description of the genus
CIFOR, 16-20 Feb 1998, South Africa, pp 23–30 Pinus. Brill, Leiden
Dubos C, Plomion C (2003) Identification of water-deficit re- Farjon A (2001) World checklist and bibliography of conifers,
sponsive genes in maritime pine (Pinus pinaster Ait.) roots. 2nd edn. The Royal Botanic Gardens, Kew, UK
Plant Mol Biol 51(2):249-262 Farjon A, Styles BT (1997) Pinus (Pinaceae). Monograph 75,
Dubos C, Provost G, Pot D, Salin F, Lalane C, Madur D, Frige- Flora Neotropica. New York Botanical Gardens, New York
rio JM, Plomion C (2003) Identification and characteri- Farnir F, Grisart B, Coppieters W, Riquet J, Berzi P, Cambisano
zation of water-stress-responsive genes in hydroponically N, Karim L, Mni M, Moisio S, Simon P, Wagenaar D, Vilkki
grown maritime pine (Pinus pinaster) seedlings. Tree Phys- J, Georges M (2002) Simultaneous mining of linkage and
iol 23:169–179 linkage disequilibrium to fine map quantitative trait loci
Dvornyk V, Sirviö A, Mikkonen M, Savolainen O (2002) Low nu- in outbred half-sib pedigrees: revisiting the location of
cleotide diversity at the pal1 locus in the widely distributed a quantitative trait locus with major effect on milk pro-
Pinus sylvestris. Mol Biol Evol 19:179–188 duction on bovine chromosome 14. Genetics 161:275–287
Echt CS, DeVerno LL, Anzidei M, Vendramin GG (1998) Chloro- Fernando RL, Grossman M (1989) Marker-assisted selection
plast microsatellites reveal population genetic diversity in using best linear unbiased prediction. Genet Sel Evol
red pine, Pinus resinosa Ait. Mol Ecol 7:307–317 21:246–477
Echt CS, Burns R (1999) PCR primer pair sequences and SSRs for Fisher PJ, Gardner RC, Richardson TE (1996) Single locus mi-
various P. taeda ESTs. Available on the dendrome Web site crosatellites isolated using 5’ anchored PCR. Nucleic Acids
at [Link] accessed 20 Res 24:4369–4371
March 2006 Fisher PJ, Gardner RC, Richardson TE (1998) Characteristics of
Echt CS, Nelson CD (1997) Linkage mapping and genome length single and multi-copy microsatellities from Pinus radiata.
in eastern white pine (Pinus strobus L.). Theor Appl Genet Theor Appl Genet 96:969–979
94:1031–1037 Flint-Garcia SA, Thornsberry JM, Buckler ES (2003) Structure
Echt CS, Saha S, Deemer D, Nelson CD (2006) Evalua- of linkage disequilibrium in plants. Annu Rev Plant Biol
tion of pine EST-SSR markers. In: Plant & Animal 54:357–374
Genome XIV Conf, San Diego, W-117 ([Link] Folkertsma RT, Spassova MI, Prins M, Stevens MR, Hille J,
[Link]/14/abstracts/PAG14_W117.html, accessed 20 Goldbach RW (1999) Construction of a bacterial artificial
March 2006) chromosome (BAC) library of Lycopersicon esculentum cv.
Edwards MD, Page NJ (1994) Evaluation of marker-assisted Stevens and its application to physically map the Sw-5 locus.
selection through computer simulation. Theor Appl Genet Mol Breed 5:197–207
88:376–382 Friesen N, Brandes A, Heslop-Harrison JS (2001) Diversity, ori-
Edwards MD, Stuber CW, Wendel JF (1987) Molecular-marker- gin, and distribution of retrotransposons (gypsy and copia)
facilitated investigations of quantitative-trait loci in maize. in conifers. Mol Biol Evol 18:1176–1188
I. Numbers, genomic distribution and types of gene action. Fulton TM, Van der Hoeven R, Eannetta NT, Tanksley SD (2002)
Genetics 116:113–125 Identification, analysis, and utilization of conserved or-
82 C. Plomion et al.

tholog set markers for comparative genomics in higher González-Martínez SC, Ersoz E, Brown GR, Wheeler NC, Neale
plants. Plant Cell 14:1457–1467 DB (2006a) DNA sequence variation and selection of tag
García-Gil MR, Mikkonen M, Savolainen O (2003) Nucleotide SNPs at candidate genes for drought-stress response in
diversity at two phytochrome loci along a latitudinal cline Pinus taeda. Genetics 172:1915–1926
in Pinus sylvestris. Mol Ecol 12:1195–1206 González-Martínez SC, Krutovsky KV, Neale DB (2006b) Forest-
García-Gutiérrez A, Cantón FR, Gallardo F, Sánchez-Jiménez tree population genomics and adaptive evolution. New
F, Cánovas FM (1995) Expression of ferredoxin-dependent Phytol 170:227–238
glutamate synthase in dark-grown pine seedlings. Plant Govindaraju D, Lewis P, Cullis C (1992) Phylogenetic analysis
Mol Biol 27:115–128 of pines using ribosomal DNA restriction fragment length
Gaussen H (1960) Les Gymnospermes actuelles et fossils. Fas- polymorphisms. Plant Syst Evol 179:141–153
cicule VI. Généralités, genre Pinus; Faculté des Sciences, Grace SL, Hamrick JL, Platt WJ (2004) Estimation of seed dis-
Toulouse persal in an old-growth population of Longleaf Pine (Pi-
Geada López G, Kamiya K, Harada K (2002) Phylogenetic re- nus palustris) using maternity exclusion analysis. Castanea
lationships of Diploxylon pines (subgenus Pinus) based on 69:207–215
plastid sequence [Link] J Plant Sci 163:737–747 Grattapaglia D, Sederoff R (1994) Genetic linkage maps of Eu-
Geldermann H (1975) Investigations on inheritance of quanti- calyptus grandis and Eucalyptus urophylla using a pseudo-
tative characters in animals by gene markers. I. Methods. testcross mapping strategy and RAPD markers. Genetics
Theor Appl Genet 46:319–330 137:1121–1137
Georges M, Nielsen D, Mackinnon M, Mishra A, Okimoto R, Greaves BL (1999) The value of tree improvement: A case study
Pasquino AT, Sargeant LS, Sorensen A, Steele MR, Zhao in radiata pine grown for structural sawn timber and liner-
X, Womack JE, Hoeschele I (1995) Mapping quantitative board. In: Nepveu G (ed) Connection between silviculture
trait loci controlling milk production in dairy cattle by and wood quality through modelling approaches and sim-
exploiting progeny testing. Genetics 139:907–920 ulation software. Proc IUFRO Workshop, 5-12 Dec 1999,
Gerber S, Rodolphe F (1994) An estimation of the genome France, pp 448–459
length of maritime pine (Pinus pinaster Ait). Theor Appl Groover AT, Devey ME, Lee JM, Megraw R, Mitchell-Olds T,
Genet 88:289–292 Sherman B, Vujcic S, Williams C, Neale DB (1994) Identi-
Gerber S, Rodolphe F, Bahrman N, Baradat P (1993) Seed- fication of quantitative trait loci influencing wood specific
protein variation in maritime pine (Pinus pinaster Ait.) gravity in an outbred pedigree of loblolly pine. Genetics
revealed by two-dimensional electrophoresis: genetic de- 138:1293–1300
terminism and construction of a linkage map. Theor Appl Groover AT, Williams CG, Devey ME, Lee JM, Neale DB
Genet 85:521–528 (1995) Sex-related differences in meiotic recombination
Gernandt DS, Geada Lopez G, Ortiz Garcia S, Liston A (2005) frequency in Pinus taeda. J Hered 86:157–158
Phylogeny and classification of Pinus. Taxon 54:29–42 Grotkopp E, Rejmanek M, Sanderson MJ, Rost TL (2004) Evo-
Gernandt DS, Liston A, Pinero D (2003) Phylogenetics of Pi- lution of genome size in pines (Pinus) and its life-history
nus subsections Cembroides and Nelsoniae inferred from correlates: supertree analyses. Evolution 58:1705–0729
cpDNA sequences. Syst Bot 28:657–673 Guevara MA, Chagné D, Almeida MH, Byrne M, Collada C,
Gion JM, Lalanne C, Le Provost G, Ferry-Dumazet H, Paiva J, Favre JM, Harvengt L, Jeandroz S, Orazio C, Plomion C,
Chaumeil P, Frigerio JM, Brach J, Barré A, de Daruvar A, Ramboer A, Rocheta M, Sebastiani F, Soto A, Vendramin
Claverol S, Bonneu M, Sommerer N, Negroni L, Plomion GG, Cervera MT (2005a) Isolation and characterization of
C, Gion J-M, Lalanne C, et al. (2005) The proteome of mar- nuclear microsatellite loci in P. pinaster. Mol Ecol Notes
itime pine wood forming tissue. Proteomics 5:3731–3751 5:57–59
Gómez MA, González-Martínez SC, Collada C, Climent J, Gil Guevara MA, Soto A, Collada C, Plomion C, Savolainen O,
L (2003) Complex population genetic structure in the en- Neale DB, González-Martínez SC, Cerevera MT (2005b) Ge-
demic Canary Island pine revealed using chloroplast mi- nomics applied to the study of adaptation in pine species.
crosatellite markers. Theor Appl Genet 107:1123–1131 Invest Agrar Syst Recur For 14:292–306
González-Martínez SC, Gerber S, Cervera MT, Martínez- Gugerli F, Senn J, Anzidei M, Madaghiele A, Buchler U, Sperisen
Zapater J, Gil L, Alía R (2002) Seed gene flow and fine-scale C, Vendramin GG (2001) Chloroplast microsatellites and
structure in a Mediterranean pine (Pinus pinaster Ait.) mitochondrial nad1 intron 2 sequences indicate congruent
using nuclear microsatellite markers. Theor Appl Genet phylogenetic relationships of Swiss stone pine (Pinus cem-
104:1290–1297 bra), Siberian stone pine (P. sibirica) and Siberian dwarf
González-Martínez SC, Robledo-Arnuncio JJ, Collada C, Díaz pine (P. pumila). Mol Ecol 10:1489–1497
A, Williams CG, Alía R, Cervera MT (2004) Cross- Gupta PK, Rustgi S, Kulwal PL (2005) Linkage disequilibrium
amplification and sequence variation of microsatellite loci and association studies in higher plants: Present status and
in Eurasian hard pines. Theor Appl Genet 109:103–11 future prospects. Plant Mol Biol 57:461–485
 Chapter 2 Pines 83

Gwaze DP, Harding KJ, Purnell RC, Bridgwater FE (2002) Opti- Hizume M, Shibata F, Matsusaki Y, Garajova Z (2002) Chromo-
mum selection age for wood density in loblolly Pine. Can J some identification and comparative karyotypic analyses
For Res 32:1393–1399 of four Pinus species. Theor Appl Genet 105:491–497
Gwaze DP, Zhou Y, Reyes-Valdés MH, Al-Rababah MA, Williams Hodge GR, Dvorak WS (1999) Genetic parameters and prove-
CG (2003) Haplotypic QTL mapping in an outbred pedi- nance variation of Pinus tecunumanii in 78 international
gree. Genet Res 81:43–50 trials. For Genet 6:157–180
Hamrick JL, Godt JW (1990) Allozyme diversity in plant species. Hodge GR, Volker PW, Potts BM, Owen JV (1996) A comparison
In: Brown A, Clegg MT, Kahler AL, Weir BS (eds) Plant Pop- of genetic information from open-pollinated and control-
ulation Genetics, Breeding and Genetic Resources. Sinauer, pollinated progeny tests in two Eucalypt species. Theor
Sunderland, MA, pp 43–63 Appl Genet 92:53–63
Hamrick JL, Godt MJW (1996) Effects of life history traits on Hospital F, Goldringer I, Openshaw S (2000) Efficient marker-
genetic diversity in plant species. Phil Trans R Soc Lond B based recurrent selection for multiple quantitative trait
Biol Sci 351:1291–1298 loci. Genet Res 75:357–368
Hannrup B, Ekberg I (1998) Age-age correlations for tracheid Hospital F, Moreau L, Lacoudre F, Charcosset A, Gallais A (1997)
length and wood density in Pinus sylvestris. Can J For Res More on the efficiency of marker-assisted selection. Theor
28:1373–1379 Appl Genet 95:1181–1189
Hannrup B, Ekberg I, Persson A (2000) Genetic correlations Hurme P, Sillanpää M, Repo T, Arjas E, Savolainen O (2000) Ge-
among wood, growth capacity and stem traits in Pinus netic basis of climatic adaptation in Scots pine by Bayesian
sylvestris. Can J For Res 15:161–170 QTL analysis. Genetics 156:1309–1322
Harkins DM, Johnson GN, Skaggs PA, Mix AD, Dupper GE, De- Hyun SK (1976) Inter-specific hybridization in pines with the
vey ME, Kinloch Jr BB, Neale DB (1998) Saturation mapping special reference to Pinus rigida and P. taeda. Silvae Genet
of a major gene for resistance to white pine blister rust in 25:188–191
sugar pine. Theor Appl Genet 97:1355–1360 Ingvarsson PK (2005) Nucleotide polymorphism and linkage
Harry DE, Temesgen B, Neale DB (1998) Codominant PCR- disequilibrium within and among natural populations of
based markers for Pinus taeda developed from mapped European Aspen (Populus tremula L., Salicaceae). Genetics
cDNA clones. Theor Appl Genet 97:327–336 169:945–953
Hartmann HT, Kester DE, Davies FT Jr (1990) Plant Propagation Isik F, Li B, Frampton J, Goldfarb B (2004) Efficiency of seedlings
Principles and Practices, 5th edn. Prentice Hall, Englewood and rooted cuttings for testing and selection in Pinus taeda.
Cliffs, NJ For Sci 50:44–53
Hayashi E, Kondo T, Terada K, Kuramoto N, Goto Y, Oka- Islam-Faridi MN, Chang Y-L, Zhang HB, Kinlaw C, Doudrick
mura M, Kawasaki H (2001) Linkage map of Japanese black RL, Neale DB, Echt CS, Price HJ, Stelly DM (1998) Construc-
pine based on AFLP and RAPD markers including mark- tion of a pine BAC library. In: Plant & Animal Genome VI
ers linked to resistance against the pine needle gall midge. Conf, San Diego
Theor Appl Genet 102:871–875 Islam-Faridi MN, Mujeeb-Kazi A (1995) Visualization of Secale
Hayashi E, Kondo T, Terada K, Kuramoto N, Kawasaki S (2004) cereale DNA in wheat germplasm by FISH. Theor Appl
Identification of AFLP markers linked to a resistance gene Genet 90:595–600
against pine needle gall midge in Japanese balck pine. Islam-Faridi MN, Nelson CD, Kubisiak TL (2007) Reference
Theor Appl Genet 108:1177–1181 karyotype and cytomolecular map for loblolly pine (Pinus
Heath LS, Ramakrishnan N, Sederoff RR, Whetten RW, Chevone taeda L.). Genome (in press)
BI, Struble CA, Jouenne VY, Chen D, van Zyl LM, Grene Islam-Faridi MN, Nelson CD, Kubisiak TL, Gullirmo MV, Mc-
R (2002) Studying the functional genomics of stress re- Namara VH, Price HJ, Stelly DM (2003) Loblolly pine kary-
sponses in loblolly pine with the “Expresso microarray otype using FISH and DAPI positive banding. In: Proc 27th
experiment management system”. Comp Funct Genom Southern For Tree Improv Conf, 25-27 June 2003, Stillwater,
3:226–243 OK, pp 184–188
Hirschhorn JN, Daly MJ (2005) Genome-wide association stud- Ivkovic M, Kumar S, Wu H (2006) Adding value to the end-
ies for common diseases and complex traits. Nat Rev Genet products of radiata pine: A review of breeding for structural
6:95–108 timber production. In: Proc 13th Australasian Plant Breed
Hizume M, Arai M, Tanaka A (1990) Chromosome banding Conf, 18-22 April 2006, Christchurch, NZ, pp 273–278
in the genus Pinus. III. Fluorescent banding pattern of Jackson RB, Schlesinger WH (2004) Curbing the U.S. carbon
P. luchuensis and its relationships among the Japanese deficit. Proc Natl Acad Sci USA 101:15827–15829
diploxylon pines. Bot Mag Tokyo 103:103–111 Jacobs MD, Gardner RC, Murray BG (2000) Cytological charac-
Hizume M, Ohgiku A, Tanaka A (1989) Chromosome banding terization of heterochromatin and rDNA in Pinus radiata
in the genus Pinus. II. Interspecific variation of fluorescent and P. taeda. Plant Syst Evo1 223:71–79
banding patterns in P. densiflora and P. thunbergii. Bot Mag Jansen RC (1993) Interval mapping of multiple quantitative trait
Tokyo 102:25–36 loci. Genetics 135:205–211
84 C. Plomion et al.

Jewell DC, Islam-Faridi MN (1994) Details of a technique Kirst M, Myburg AA, De Leon JP, Kirst ME, Scott J, Sederoff
for somatic chromosome preparation and C-banding R (2004) Coordinated genetic regulation of growth and
of maize. In: Freeling M, Walbot V (eds) The Maize lignin revealed by quantitative trait locus analysis of cDNA
Hand Book. Springer, Berlin Heidelberg New York, microarray data in an interspecific backcross of Eucalyptus.
pp 484–493 Plant Physiol 135:2368–2378
Jiang C, Zeng ZB (1995) Multiple trait analysis of genetic map- Kirst M, Cordeiro CM, Rezende GDSP, Grattapaglia D (2005)
ping for quantitative trait loci. Genetics 140:1111–1127 Power of microsatellite markers for fingerprinting and
Jiang JM, Liu ZX, Lu BS, Jiang JM, Liu ZX, Lu BS (1999) Prove- parentage analysis in Eucalyptus grandis breeding popu-
nance genetic variation analysis of loblolly pine and deter- lations. J Hered 96:161–166
mination of suitable provenances (areas). For Res Beijing Knott S, Neale DN, Sewell MM, Haley C (1997) Multiple marker
12:485–492 mapping of quantitative trait loci in an outbred pedigree
Johnson GR, Wheeler NC, Strauss SH (2000) Financial feasibil- of loblolly pine. Theor Appl Genet 94:810–820
ity of marker-aided selection in Douglas-fir. Can J For Res Komulainen P, Brown GR, Mikkonen M, Karhu A, Garcia-
30:1942–1952 Gil MR, O’Malley DM, Lee B, Neale DB, Savolainen O
Kamm A, Doudrick RL, Heslop-Harrison JS, Schmidt T (1996) (2003) Comparing EST-based genetic maps between Pinus
The genomic and physical organization of Ty1-copia-like sylvestris and Pinus taeda. Theor Appl Genet 107:667–678
sequences as a component of large genomes in Pinus elliot- Kondo T, Terada K, Hayashi E, Kuramoto N, Okamura M,
tii var. elliottii and other gymnosperms. Proc Natl Acad Sci Kawasaki H (2000) RAPD markers linked to a gene for
USA 93:2708–2713 resistance to pine needle gall midge in Japanese black pine
Kao C-H, Zeng J, Teasdale RD (1999) Multiple interval mapping (Pinus thunbergii). Theor Appl Genet 100:391–395
for quantitative trait loci. Genetics 152:1203–1216 Korol AB, Ronin YI, Kirzhner VM (1995) Interval mapping
Karhu A, Hurme P, Karjalainen M, Karvonen P, Kärkkäinen K, of quantitative trait loci employing correlated traits com-
Neal DB, Savolainen O (1996) Do molecular markers reflect plexes. Genetics 140:1137–1147
patterns of differentiation in adaptive traits of conifers? Korol AB, Ronin YI, Nevo E, Hayes PM (1998) Multi-
Theor Appl Genet 93:215–221 interval mapping of correlated trait complexes. Heredity
Kärkkäinen K, Savolainen O (1993) The degree of early in- 80:273–284
breeding depression determines the selfing rate at the seed Kossack DS, Kinlaw CS (1999) IFG, a gypsy-like retrotranspo-
stage: model and results from Pinus sylvestris (Scots pine). son in Pinus (Pinaceae), has an extensive history in pines.
Heredity 71:160–166 Plant Mol Biol 39:417–426
Kaya Z, Neale DB (1995) Utility of random amplified polymor- Kremer A (1992) Predictions of age-age correlations of total
phic DNA (RAPD) markers for linkage mapping in Turkish height based on serial correlations between height incre-
red pine (Pinus brutia Ten). Silvae Genet 44:110–116 ments in Maritime pine (Pinus pinaster Ait.). Theor Appl
Kaya Z, Sewell MM, Neale DB (1999) Identification of quanti- Genet 85:152–158
tative trait loci influencing annual height- and diameter- Kriebel HB (1985) DNA sequence components of the Pinus
increment growth in loblolly pine (Pinus taeda L.). Theor strobus nuclear genome. Can J For Res 15:1–4
Appl Genet 98:586–592 Kruglyak L (1999) Prospects for whole-genome linkage dise-
Kayihan GC, Huber DA, Morse AM, White TL, Davis JM (2005) quilibrium mapping of common disease genes. Nat Genet
Genetic dissection of fusiform rust and pitch canker disease 22:139–144
traits in loblolly pine. Theor Appl Genet 110:948–958 Krupkin AB, Liston A, Strauss SH (1996) Phylogenetic anal-
Kent J, Richardson TE (1997) Fluorescently labelled, mul- ysis of the hard pines (Pinus subgenus Pinus, Pinaceae)
tiplexed chloroplast microsatellites for high-throughput from chloroplast DNA restriction site analysis. Am J Bot
paternity analysis in Pinus radiata. NZ J For Sci 83:489–498
27:305–312 Krutovsky KV, Neale DB (2005) Nucleotide diversity and
Khoshoo TN (1961) Chromosome numbers in gymnosperms. linkage disequilibrium in cold hardiness and wood
Silvae Genet 10:1–9 quality related candidate genes in Douglas-fir. Genetics
Kinlaw CS, Neale DB (1997) Complex gene families in pine 171:2029–2041
genomes. Trends Plant Sci 2:356–359 Krutovsky KV, Troggio M, Brown GR, Jermstad KD, Neale DB
Kinloch BB, Parks GK, Flower CW (1970) White pine blister (2004) Comparative mapping in the Pinaceae. Genetics
rust: Simply inherited resistance in sugar pine. Science 168:447–461
167:193–195 Kuang H, Richardson TE, Carson SD, Bongarten B (1999b) Ge-
Kirst M, Johnson AF, Baucom C, Ulrich E, Hubbard K, Staggs R, netic analysis of inbreeding depression in plus tree 850.55
Paule C, Retzel E, Whetten R, Sederoff R (2003) Apparent of Pinus radiata D. Don. II. Genetics of viability genes.
homology of expressed genes from wood forming tissues Theor Appl Genet 99:140–146
of loblolly pine (Pinus taeda L.) with Arabidopsis thaliana. Kuang H, Richardson TE, Carson SD, Wilcox PL, Bon-
Proc Natl Acad Sci USA 100:7383–7388 garten B (1999a) Genetic analysis of inbreeding depres-
 Chapter 2 Pines 85

sion in plus tree 850.055 of Pinus radiata D. Don. 1. Lande R, Thompson R (1990) Efficiency of marker-assisted se-
Genetic map with distorted markers. Theor Appl Genet lection in the improvement of quantitative traits. Genetics
98:697–703 124:743–756
Kubisiak TL, Nelson CD, Nance WL, Stine M (1995) RAPD Lander ES, Botstein D (1989) Mapping Mendelian factors un-
linkage mapping in a longleaf pine × slash pine F1 family. derlying quantitative traits using RFLP linkage maps. Ge-
Theor Appl Genet 90:1119–1127 netics 121:185–199
Kubisiak TL, Nelson CD, Nance WL, Stine M (1996) Comparison Laurie CC, Chasalow SD, LeDeaux JR, McCarroll R, Bush D,
of RAPD linkage maps constructed for a single longleaf Hauge B, Lai C, Clark D, Rocheford TR, Dudley JW (2004)
pine from both haploid and diploid mapping populations. The genetic architecture of response to long-term artificial
For Genet 3:230–211 selection for oil concentration in the maize kernel. Genetics
Kubisiak TL, Nelson CD, Nowak J, Friend AL (2000) Genetic 168:2141–2155
linkage mapping of genomic regions conferring tolerance Le Dantec L, Chagné D, Pot D, Cantin O, Garnier-Géré P, Bedon
to high aluminum in slash pine. J Sust For 10:69–78 F, Frigerio JM, Chaumeil P, Léger P, García V, Laigret F, de
Kubisiak TL, Nelson, CD, Stine M (1997) RAPD mapping of Daruvar A, Plomion C (2004) Automated SNP detection
genomic regions influencing early height growth in lon- in expressed sequence tags: statistical considerations and
gleaf pine × slash pine F1 hybrids. In: Proc 24th South- application to maritime pine sequences. Plant Mol Biol
ern For Tree Improv Conf, Orlando, FL, 9-12 June 1997, 54:461–470
pp 198–206 Le Provost G, Paiva J, Pot D, Brach J, Plomion C (2003) Seasonal
Kumar S (2004) Genetic parameter estimates for wood stiffness, variation in transcript accumulation in wood forming tis-
strength, internal checking and resin bleeding for radiata sues of maritime pine (Pinus pinaster Ait.) with emphasis
pine. Can J For Res 34:2601–2610 on a cell wall Glycine Rich Protein. Planta 217:820–830
Kumar S, Garrick DJ (2001) Genetic response to within-family Ledig FT (1998) Genetic variation in Pinus. In: Richardson DM
selection using molecular markers in some radiata pine (ed) Ecology and Biogeography of Pinus. Cambridge Uni-
breeding schemes. Can J For Res 31:779–785 versity Press, Cambridge, UK, pp 251–280
Kumar S, Richardson TE (2005) Inferring heritability and relat- Ledig FT, Conkle MT (1983) Gene diversity and genetic struc-
edness using molecular markers in radiata pine. Mol Breed ture in a narrow endemic Torrey pine (Pinus torreyana
15:55–64 Parry ex Carr). Evolution 37:79–85
Kumar S, Spelman R, Garrick D, Richardson TE, Wilcox PL Leitch IJ, Hanson L, Winfield M, Parker J, Bennett MD (2001)
(2000) Multiple marker mapping of quantitative trait loci Nuclear DNA C-values complete familial representation in
on chromosome three in an outbred pedigree of radiata gymnosperms. Ann Bot 88:843–849
pine. Theor Appl Genet 100:926–933 Lerceteau EC, Plomion C, Andersson B (2000) AFLP mapping
Kumar S, Echt C, Wilcox PL, Richardson TE (2004) Testing for and detection of quantitative trait loci (QTLs) for econom-
linkage disequilibrium in the New Zealand radiata pine ically important traits in Pinus sylvestris: a preliminary
breeding population. Theor Appl Genet 108:292–298 study. Mol Breed 6:451–458
Kumar S, Gerber S, Richardson TE (2006) Pedigree recon- Li C, Yeh FC (2001) Construction of a framework map in Pinus
struction using SSR markers in a radiata pine breed- contorta subsp. latifolia using random amplified polymor-
ing programme. In: Proc 13th Australasian Plant Breed phic DNA markers. Genome 44:147–153
Conf, 17-21 April, Christchurch, New Zealand pp 578–583, Liston A, Robinson WA, Pinero D, Alvarez-Buylla ER (1999)
[Link] Phylogenetics of Pinus (Pinaceae) based on nuclear ribo-
Kutil BL, Williams CG (2001) Triplet-repeat microsatellites somal DNA internal transcribed spacer region sequences.
shared among hard and soft pines. J Hered 92:327–32 Mol Phylogenet Evol 11:95–109
Lambeth C (2000) Realized genetic gains for first generation Liu ZL, Zhang D, Hong DY, Wang XR (2003) Chromosomal
improved loblolly pine in 45 tests in coastal North Carolina. localization of 5S and 18S-5.8S-25S ribosomal DNA sites in
J Appl For 24:140–144 five Asian pines using fluorescence in situ hybridization.
Lambeth C, Lee BC, O’Malley D, Wheeler N (2001) Polymix Theor Appl Genet 106:198–204
breeding with parental analysis of progeny: an alterna- Lopez-Upton J, White TL, Huber DA (2000) Species differences
tive to full-sib breeding and testing. Theor Appl Genet in early growth and rust incidence of loblolly and slash
103:930–943 pine. For Ecol Mng 132:211–222
Lamoureux D, Peterson DG, Li W, Fellers JP, Gill BS (2005) The Lorenz WW, Dean JFD (2002) SAGE profiling and demonstra-
efficacy of Cot-based gene enrichment in wheat (Triticum tion of differential gene expression along the axial develop-
aestivum L.). Genome 48:1120–1126 mental gradient of lignifying xylem in loblolly pine (Pinus
Lamprecht H (1990) Silviculture in the Tropics: tropical forest taeda). Tree Physiol 22:301–310
ecosystems and their tree species-possibilities and meth- Lorenz WW, Sun F, Liang C, Kolychev D, Wang H, Zhao X,
ods for their long-term utilization. GTZ, Eschborn, Ger- Cordonnier-Pratt MM, Pratt LH, Dean JFD (2005) Water
many stress-responsive genes in loblolly pine (Pinus taeda) roots
86 C. Plomion et al.

identified by analyses of expressed sequence tag libraries. Matziris DI (1995) Provenance variation of Pinus radiata grown
Tree Physiol 26:1–16 in Greece. Silvae Genet 44:88–96
Lorenz WW, Sun F, Liang C, Kolychev D, Wang H, Zhao X, McKeand S, Mullin T, Byram T, White TL (2003) Deployment of
Cordonnier-Pratt MM, Pratt LH, Dean JF (2006) Water genetically improved loblolly and slash pines in the South
stress-responsive genes in loblolly pine (Pinus taeda) roots USA. J For 101:32–37
identified by analyses of expressed sequence tag libraries. Mergen F (1958) Natural polyploidy in slash pine. For Sci
Tree Physiol 26:1–16 4:283–295
Lowe WJ, Byram TD, Bridgwater FE (1999) Selecting loblolly Miksche JP, Hotta Y (19931973) DNA base composition and
pine parents for seed orchards to minimise the costs of repetitious DNA in several conifers. Chromosoma 41:29–36
producing pulp. For Sci 45:213–216 Millar CI (1998) Early evolution of pines. In: Richardson DM
Lubaretz O, Fuchs J, Ahne R, Meister A, Schubert I (1996) Kary- (ed) Ecology and Biogeography of Pinus. Cambridge Uni-
otyping of three Pinaceae species via fluorescent in situ versity Press, Cambridge, UK, pp 69–91
hybridization and computer-aided chromosome analysis. Mitton JB (1992) The dynamic mating system of conifers. New
Theor Appl Genet 92:411–416 For 6:187–216
Luikart G, England P, Tallmon D, Jordan S, Taberlet P (2003) The Moore G, Devos KM, Wang Z, Gale MD (1995) Grasses, line up
power and promise of population genomics: from genotyp- and form a circle. Curr Biol 5(7):737-739
ing to genome typing. Nat Rev Genet 4:981–994 Moran GF, Bell JC, Hilliker AJ (1983) Greater meiotic recombi-
Lund MS, Sorensen P, Guldbrantsen B, Sorensen DA (2003) nation in male vs. female gametes in Pinus radiata. J Hered
Multitrait fine mapping of quantitative trait loci using com- 74:62
bined linkage disequilibrium and linkage analysis. Genet- Morgante M, Salamini F (2003) From plant genomics to breed-
ics 163:405–410 ing practice. Curr Opin Biotechnol 14:214–219
Ma XF, Szmidt A, Wang WR (2006) Genetic structure and evo- Morgante M, Vendramin GG, Rossi P, Olivieri AM (1993) Se-
lutionary history of a diploid hybrid pine Pinus densata lection against inbreds in early life-cycle phases in Pinus
inferred from the nucleotide variation at seven gene loci. leucodermis. Heredity 70:622–627
Mol Biol Evol 23:807–816 Morin PA, Luikart G, Wayne RK, and the SNP workshop group
MacPherson P, Filion WG (1981) Karyotype analysis and the (2004) SNPs in ecology, evolution and conservation. Trends
distribution of constitutive heterochromatin in five species Ecol Evol 19:208–2169
of Pinus. J Hered 72:193–198 Morse AM, Nelson CD, Covert SF, Holliday AG, Smith
Margulies M, Egholm M, Altman WE, Attiya S, Bader JS, Bem- KE, Davis JM (2004) Pine genes regulated by the
ben LA, Berka J, Braverman MS, Chen YJ, Chen Z, Dewell necrotrophic pathogen Fusarium circinatum. Theor Appl
SB, Du L, Fierro JM, Gomes XV, Godwin BC, He W, Helge- Genet 109:922–932
sen S, Ho CH, Irzyk GP, Jando SC, Alenquer ML, Jarvie Morse AM, Nelson CD, Covert SF, Smith KE, Davis JM
TP, Jirage KB, Kim JB, Knight JR, Lanza JR, Leamon JH, (2004) Pine genes regulated by the necrotrophic pathogen,
Lefkowitz SM, Lei M, Li J, Lohman KL, Lu H, Makhijani Fusarium circinatum. Theoretical and Applied Genetics
VB, McDade KE, McKenna MP, Myers EW, Nickerson E, 109:922-932
Nobile JR, Plant R, Puc BP, Ronan MT, Roth GT, Sarkis GJ, Morton NE (1991) Parameters of the human genome. Proc Natl
Simons JF, Simpson JW, Srinivasan M, Tartaro KR, Tomasz Acad Sci USA 88:7474–7476
A, Vogt KA, Volkmer GA, Wang SH, Wang Y, Weiner MP, Mosig MO, Lipkin E, Khutoreskaya G, Tchourzyna E, Soller M,
Yu P, Begley RF, Rothberg JM (2005) Genome sequencing Friedmann A (2001) A whole genome scan for quantita-
in microfabricated high-density picolitre [Link] tive trait loci affecting milk protein percentage in Israeli-
437:376–380 Holstein cattle, by means of selective milk DNA pooling in
Mariette S, Chagné D, Decroocq S, Vendramin GG, Lalanne a Daughter design, using an adjusted false discovery rate
C, Madur D, Plomion C (2001) Microsatellite markers for criterion. Genetics 157:1683–1698
Pinus pinaster Ait. Ann For Sci 58:203–206 Moura VPG, Dvorak WS, Hodge GR (1998) Provenance and
Markussen T, Fladung M, Achere V, Favre JM, Faivre-Rampant family variation of Pinus oocarpa grown in the Brazilian
P, Aragones A, Da Silva Perez, Havengt L, Ritter E (2003) cerrado. For Ecol Manage 109:315–322
Identification of QTLs controlling growth, chemical and Müller-Starck G (1998) Isozymes. In: Karp A, Isaac PG, Ingram
physical wood property traits in Pinus pinaster (Ait.). Silvae DS (eds) Molecular Tools for Screening Biodiversity: Plants
Genet 52:8–15 and Animals. Chapman and Hall, London, UK, pp 75–81
Marquardt PE, Epperson BK (2004) Spatial and population ge- Muona O, Harju A (1989) Effective population sizes, genetic
netic structure of microsatellites in white pine. Mol Ecol variability and mating system in natural stands and seed
13:3305–3315 orchards of Pinus sylvestris. Silvae Genet 38:221–228
Matheson AC, Lindgren D (1985) Gain from the clonal and the Muona O, Yazdani R, Rudin R (1987) Genetic change between
clonal seed orchard options compared for tree breeding life stages in Pinus sylvestris: allozymes variation in seeds
programs. Theor Appl Genet 71:242–249 and planted seedlings. Silvae Genet 35:39–42
 Chapter 2 Pines 87

Murray B (1998) Nuclear DNA amounts in gymnosperms. Ann O’Malley D, Grattaplagia D. Chaparro JX, Wilcox PL, Amerson
Bot 82(Suppl A):3-15 HV, Liu B-H, Whetton R, McKeand S, Kuhlman EG, McCord
Myburg H, Morse AM, Amerson HV, Kubisiak TL, Huber DA, S, Crane B, Sederoff RR (1996) Molecular markers, forest
Osborne JA, Garcia SA, Nelson CD, Davis JM, Covert SF, van genetics and tree breeding. In: Gustafson JP, Flavell RB
Zyl LM (2006) Differential gene expression in loblolly pine (eds) Genomes of Plants and Animals: 21st Stadler Genet
(Pinus taeda L.) challenged with the fusiform rust fungus, Symp. Plenum, New York
Cronartium quercuum f. sp. fusiforme. Physiological and Ohri D, Khoshoo TN (1986) Genome size in gymnosperms.
Molecular Plant Pathology 69:79-101 Plant Syst Evol 153:119–131
Myers RM, Maniatis T, Lerman LS (1987) Detection and local- OlssonT, Lindgren D, Li B (2001) Balancing genetic gain and
ization of single base changes by denaturing gradient gel relatedness in seed orchards. Silvae Genet 50:222–227
electrophoresis. Methods Enzymol 155:501–527 Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T
Neale DB, Savolainen O (2004) Association genetics of complex (1989) Detection of polymorphisms of human DNA by gel
traits in conifers. Trends Plant Sci 9:325–330 electrophoresis as single-strand conformation polymor-
Neale DB, Sederoff RR (1989) Paternal inheritance of chloro- phisms. Proc Natl Acad Sci USA 86:2766–2770
plast DNA and maternal inheritance of mitochondrial DNA Ott J (1991) Analysis of Human Genetic Linkage. Johns Hopkins
in loblolly pine. Theor Appl Genet 77:212–216 University Press, Baltimore, MD
Neale DB, Williams CG (1991) Restriction fragment length poly- Pacheco D, Díaz S, Osuna D, Bautista R, Claros MG, Cánovas
morphism mapping in conifers and applications to forest FM, Cantón FR (2005) Identification of candidate genes
genetics and tree improvement. Can J For Res 21:545–554 conferring specific properties to distinct types of Pinus
Nelson CD, Echt CS (2003) New models for marker-assisted pinaster Ait. wood. IUFRO Tree Biotechnol, Pretoria, South
selection in tree breeding. In: McKinley C (ed) Proc 27th Africa
South For Tree Improv Conf, Stillwater, OK, p 114 Paiva J (2006) Phenotypic and molecular plasticity of wood
Nelson CD, Echt CS (2004) Marker-directed population im- forming tissues in maritime pine (Pinus pinaster Ait.). PhD
provement. In: Li B, McKeand S (eds) Proc IUFRO Joint thesis of Universidad Nova de Lisboa and Bordeaux 1
Conf of Div 2, For Genet Tree Breed in the Age of Genomics: Paran I, Zamir D (2003) Quantitative traits in plants: beyond
Progress and Future, Charleston, SC, p 255 the QTL. Trends Genet 19:303–306
Nelson CD, Kubisiak TL, Johnson G, Burdine C, Bridgwater Park YS (2002) Implementation of conifer somatic embryogen-
FE (2003) Microsatellite analysis of loblolly pine. In: esis in clonal forestry: technical requirements and deploy-
Plant & Animal Genome XI Conf, San Diego, p 560 ment considerations. Ann For Sci 59:651–656
([Link] Paterson AH, Freeling M, Sasaki T (2005) Grains of knowledge:
accessed 20 March 2006) genomics of model cereals. Genome Res 15:1643–1650
Nelson CD, Kubisiak, TL, Stine M, Nance WL (1994) A genetic Paul AD, Foster GS, Caldwell T, McRae J (1997) Parameters
linkage map of longleaf pine (Pinus palustris Mill.) based on for height, diameter, and volume in a multilocation clonal
random amplified polymorphic DNAs. J Hered 85:433–439 study with loblolly pine. For Sci 43:87–98
Nelson CD, Nance WL, Doudrick RL (1993) A partial genetic Pavy N, Laroche J, Bousquet J, MacKay J (2005) Large-scale
linkage map of slash pine (Pinus elliottii Engelm. var. elliot- analysis of secondary xylem ESTs in pine. Plant Mol Biol
tii) based on random amplified polymorphic DNAs. Theor 57:203224
Appl Genet 8:145–151 Perlack RD, Wright LL, Turhollow AF, Graham RL,
Nelson JC (1997) QGENE: software for marker-based genomic Stokes BJ, Erbach DC (2005) Biomass as a feed-
analysis and breeding. Mol Breed 3:239–245 stock for a bioenergy and bioproducts industry: the
Nikles DG (2000) Experience with some Pinus hybrids in technical feasibility of a billion-ton annual supply.
Queensland, Australia. In: Dungey HS, Dieters MJ, Nikles Report No DOE/GO-102005-2135. US Dept of En-
DG (Compl) Hybrid Breeding and Genetics of Forest ergy/US Dept of Agri. [Link]
Trees. Proc of QFRI/CRC-SPF Symp, 9-14 April 2000, biomass/pdfs/final_billionton_vision_report2.pdf
Noosa, Queensland, Australia. Dept of Pri Indus, Brisbane, Peterson DG (2005) Reduced representation strategies
pp 14–26 and their application to plant genomes. In: Meksem
No E, Zhou Y, Loopstra CA (2000) Sequences upstream and K, Kahl G (eds) The Handbook of Genome Map-
downstream of two xylem-specific pine genes influence ping: Genetic and Physical Mapping. Wiley, Weinheim,
their expression. Plant Sci 160:77–86 pp 307–335
NSF (2005) NSF, USDA and DOE Award $32 Million to Se- Peterson DG, Chouvarine P, Thummasuwan S, Saha S, Mukher-
quence Corn Genome. Press Release 05-197. National jee D, Carlson JE (2006) Exploring the pine genome us-
Science Foundation. 11-15-2005. [Link] ing Cot filtration and 454 Life Sciences massively par-
news/news_summ.jsp?cntn_id=104608 allel shotgun sequencing. In: Plant & Animal Genomes
O’Farrell PH (1975) High resolution two-dimensional elec- XIV Conf, San Diego. [Link]
trophoresis of proteins J Biol Chem 250:4007–4021 abstracts/PAG14_W376.html
88 C. Plomion et al.

Peterson DG, Nelson CD, Islam-Faridi MN, Main DS, xylem of maritime pine (Pinus pinaster Ait.). Plant Physiol
Tomkins JP (2005) Accelerating pine genomics through 123:959–969
development and utilization of molecular and cy- Plomion C, LeProvost G, Pot D, Vendramin G, Gerber S, De-
togenetic resources. In: Plant & Animal Genomes croocq S, Brach J, Raffin A, Pastuszka P (2001) Pollen
XIII Conf, San Diego. [Link] contamination in a maritime pine polycross seed orchard
abstracts/PAG13_P515.html and certification of improved seeds using chloroplast mi-
Peterson DG, Schulze SR, Sciara EB, Lee SA, Bowers JE, Nagel A, crosatellites. Can J For Res 31:1816–1825
Jiang N, Tibbitts DC, Wessler SR, Paterson AH (2002) Inte- Plomion C, Bahrman N, Costa P, Dubos C, Frigério J-M, Ger-
gration of Cot analysis, DNA cloning, and high-throughput ber S, Gion J-M, Lalanne C, Madur D, Pionneau C (2004)
sequencing facilitates genome characterization and gene Proteomics for genetics and physiological studies in forest
discovery. Genome Res 12:795–807 trees: application in maritime pine. In: Kumar S, Fladung
Petit RJ, Duminil J, Fineschi S, Hampe A, Salvini D, Vendramin M (eds) Molecular Genetics and Breeding of Forest Trees.
GG (2005) Comparative organisation of chloroplast, mito- Haworth, Binghamton, NY, pp 53–80
chondrial and nuclear diversity in plant populations. Mol Pot D (2004) Déterminisme génétique de la qualité du bois
Ecol 14:689–701 chez le pin maritime: du phénotype aux gènes. PhD thesis,
Petit RJ, Vendramin GG (2007) Phylogeography of organelle University of Rennes I, ENSAR, France
DNA in plants: an introduction. In: Weiss S, Ferrand N (eds) Pot D, Chantre G, Rozenberg P, Rodrigues JC, Jones GL, Pereira
Phylogeography of Southern European Refugia. Kluwer, H, Hannrup B, Cahalan C, Plomion C (2002) Genetic con-
Amsterdam, (in press) trol of pulp and timber properties in maritime pine (Pinus
Pflieger S, Lefebvre V, Causse M (2001) The candidate gene pinaster Ait.). Ann For Sci 59:563–575
approach in plant genetics: a review. Mol Breed 7:275–291 Pot D, McMillan L, Echt C, Le Provost G, Garnier-Géré P, Cato
Plomion C, Durel C-E (1996) Estimation of the average effects S, Plomion C (2005a) Nucleotide variation in genes in-
of specific alleles detected by the pseudo-testcross QTL volved in wood formation in two pine species. New Phytol
mapping strategy. Gen Sel Evol 28:223–235 167:101–112
Plomion C, O’Malley DM (1996) Recombination rate differ- Pot D, Rodrigues J-C, Rozenberg P, Chantre G, Tibbits J, Cahalan
ences for pollen parents and seed parents in pine. Heredity C, Pichavant F, Plomion C (2005b) QTLs and candidate
77:341–350 genes for wood properties in maritime pine (Pinus pinaster
Plomion C, Bahrman N, Durel CE, O’Malley DM (1995a) Ge- Ait.). Tree Genet Genom 2:10–26
nomic analysis in Pinus pinaster (Maritime pine) using Price RA, Liston A, Strauss SH (1998) Phylogeny and systemat-
RAPD and protein markers. Heredity 74:661–668 ics of Pinus. In: Richardson DM (ed) Ecology and Biogeog-
Plomion C, O’Malley DM, Durel CE (1995b) Genomic analysis in raphy of Pinus. Cambridge University Press, Cambridge,
Maritime pine (Pinus pinaster). Comparison of two RAPD UK, pp 49–68
maps using selfed and open-pollinated seeds of the same Provan J, Soranzo N, Wilson NJ, Goldstein DB, Powell W (1999)
individual. Theor Appl Genet 90:1028–1034 A low mutation rate for chloroplast microsatellites. Genet-
Plomion C, Durel C-E, O’Malley D (1996a) Genetic dissection of ics 153:943–947
height in maritime pine seedlings raised under accelerated Quackenbush J, Cho J, Lee D, Liang F, Holt I, Karamycheva
growth condition. Theor Appl Genet 93:849–858 S, Parvizi B, Pertea G, Sultana R, White J (2001) The
Plomion C, Yani A, Marpeau A (1996b) Genetic determinism of TIGR Gene Indices: analysis of gene transcript sequences
?3-carene in maritime pine using random amplified poly- in highly sampled eukaryotic species. Nucleic Acids Res
morphic DNA (RAPD) markers. Genome 39:1123–1127 29:159–164
Plomion C, Liu BH, O’Malley DM (1996c) Genetic analysis using Quencez C, Bastien C (2001) Genetic variation within and be-
trans dominant linked markers in an F2 family. Theor Appl tween populations of Pinus sylvestris L. (Scots pine) for
Genet 93:1083–1089 susceptibility to Melampsora pinitorqua Rostr. (pine twist
Plomion C, Costa P, Bahrman N, Frigerio JM (1997) Genetic rust). Heredity 86:36–44
Analysis of needle proteins in maritime pine. 1. Mapping Rabinowicz PD, Citek R, Budiman MA, Nunberg A, Bedell JA,
dominant and codominant protein markers assayed on Lakey N, O’Shaughnessy AL, Nascimento LU, McCombie
diploid tissue, in a haploid-based genetic map. Silvae Genet WR, Martienssen RA (2005) Differential methylation of
46:161–165 genes and repeats in land plants. Genome Res 15:1431–1440
Plomion C, Hurme P, Frigerio J-M, Ridolphi M, Pot D, Pionneau Rabinowicz PD, Schutz K, Dedhia N, Yordan C, Parnell LD,
C, Avila C, Gallardo F, David H, Neutlings G, Campbell M, Stein L, McCombie WR, Martienssen RA (1999) Differ-
Canovas FM, Savolainen O, Bodénès C, Kremer A (1999) ential methylation of genes and retrotransposons facili-
Developing SSCP markers in two Pinus species. Mol Breed tates shotgun sequencing of the maize genome. Nat Genet
5:21–31 23:305–308
Plomion C, Pionneau C, Brach J, Costa P, Baillères H (2000) Rafalski A (1998) Randomly amplified polymorphic DNA
Compression wood-responsive proteins in developing (RAPD) analysis. In: Caetano-Anollés G, Gresshoff
 Chapter 2 Pines 89

PM (eds) DNA Markers Protocols, Applications, and Rudin D, Ekberg I (1978) Linkage studies in Pinus sylvestris L.
Overviews. Wiley-Liss, New York, pp 75–83 using macrogametophyte allozymes. Silvae Genet 27:1–12
Rake AV, Miksche JP, Hall RB, Hansen KM (1980) DNA reassoci- Sanchez AC, Ilag LL, Yang D, Brar DS, Ausubel F, Khush GS,
ation kinetics of four conifers. Can J Genet Cytol 22:69–79 Yano M, Sasaki T, Li Z, Huang N (1999) Genetic and physical
Rat Genome Sequencing Project Consortium (2004) Rat mapping of xa13, a recessive bacterial blight resistance gene
Genome Sequencing Project Consortium, Genome se- in rice. Theor Appl Genet 98:1022–1028
quence of the Brown Norway rat yields insights into mam- Sánchez C, Vielva MJ, Vieitez AM, de Mier B, Abarca D, Díaz-
malian evolution. Nature 428:493–521 Sala C (2005) Identification of genes related to adventitious
Remington DL, O’Malley DM (2000a) Evaluation of major ge- rooting capacity in pine and chestnut. IUFRO Tree Biotech-
netic loci contributing to inbreeding depression for sur- nol, Pretoria, South Africa
vival and early growth in a selfed family of Pinus taeda. Sathyan P, Newton RJ, Loopstra CA (2005) Genes induced
Evolution 54:1580–1589 by WDS are differentially expressed in two populations
Remington DL, O’Malley DM (2000b) Whole genome character- of aleppo pine (Pinus halepensis). Tree Genet Genom
ization of embryonic state inbreeding depression in a selfed 1:166–173
loblolly pine family. Genetics 155:337–348 Sax K (1923) The association of size differences with seed-coat
Remington DL, Whetten RW, Liu BH, O’Malley DM (1999) pattern and pigmentation in Phaseolus vulgaris. Genetics
Construction of an AFLP genetic map with nearly com- 8:552–560
plete genome coverage in Pinus taeda. Theor Appl Genet Sax K, Sax HJ (1933) Chromosome numbers and morphology
98:1279–1292 in the conifers. J Arnold Arbor 14:356–375
Ribeiro MM, Plomion C, Petit RJ, Vendramin GG, Szmidt AE Saylor LC (1961) A karyotype analysis of selected species of
(2001) Variation of chloroplast single-sequence repeats in Pinus. Silvae Genet 10:77–85
Portuguese maritime pine (Pinus pinaster Ait.). Theor Appl Saylor LC (1964) Karyotype analysis of Pinus –group Lariciones.
Genet 102:97–103 Silvae Genet 13:165–170
Richardson BA., Brunsfeld SJ, Klopfenstein NB (2002) DNA Saylor LC (1972) Karyotype analysis of Pinus –subgenus Pinus.
from bird-dispersed seed and wind-disseminated pollen Silvae Genet 21:155–163
provides insights into postglacial colonization of whitebark Saylor LC (1983) Karyotype analysis of the genus Pinus –
pine (Pinus albicaulis). Mol Ecol 11:215–227 subgenus Strobus. Silvae Genet 32:119–121
Richardson DM, Rundel PW (1998) Ecology and biogeography Schmidt A, Doudrick RL, Heslop-Harrison JS, Schmidt T (2000)
of Pinus: an introduction. In: Richardson DM (ed) Ecology The contribution of short repeats of low sequence complex-
and Biogeography of Pinus. Cambridge University Press, ity to large conifer genomes. Theor Appl Genet 101:7–14
Cambridge, UK, pp 3–46 Schmidt R (2002) Plant genome evolution: lessons from com-
Ritland K, Zhuang J, Ralph S, Ritland C, Bohlmann J (2006) parative genomics at the DNA level. Plant Mol Biol 48:21–37
Comparative conifer genomics and the new “Treenomix” Scotti I, Burelli A, Cattonaro F, Chagné D, Fuller J, Hedley PE,
program. In: Plant & Animal Genomes XIV Conf, San Jansson G, Lalanne C, Madur D, Neale D, Plomion C, Powell
Diego, W124 W, Troggio M, Morgante M (2005) Analysis of the distri-
Ritter E, Aragonés A, Markussen T, Acheré V, Espinel S, Fladung bution of marker classes in a genetic linkage map: a case
M, Wrobel S, Faivre-Rampant P, Jeandroz S, Favre JM (2002) study in Norway spruce (P. abies karst). Tree Genet Genom
Toward construction of an ultra high density linkage map 1:93–102
for Pinus pinaster. Ann For Sci 59:637–643 Scotti-Saintagne C, Bodénès C, Barreneche T, Bertocchi E,
Robledo-Arnuncio JJ, Collada C, Alía R, Gil L (2005) Genetic Plomion C, Kremer A (2004) Detection of quantitative trait
structure of montane isolates of Pinus sylvestris L. in loci controlling bud burst and height growth in Quercus
a Mediterranean refugial area. J Biogeogr 32:595–605 robur L. Theor Appl Genet 109:1648–1659
Rosvall O, Lindgren D, Mullen TJ (1998) Sustainability, robust- Sewell MM, Bassoni DL, Megraw RA, Wheeler NC, Neale DB
ness and efficiency of a multi-generation breeding strat- (2000) Identification of QTLs influencing wood property
egy based on within-family clonal selection. Silvae Genet traits in loblolly pine (Pinus taeda L.). I. Physical wood
47:307- 321 properties. Theor Appl Genet 101:1273–1281
Rounsley S, Xiaoying L, Ketchum KA (1998) Large-scale se- Sewell MM, Davis MF, Tuskan GA, Wheeler NC, Elam CC, Bas-
quencing of plant genomes. Curr Opin Plant Biol 1:136–141 soni DL, Neale DB (2002). Identification of QTLs influenc-
Rozenberg P, Van Loo J, Hannrup B, Grabner M. 2002. Clonal ing wood property traits in loblolly pine (Pinus taeda L.). II.
variation of wood density record of cambium reaction Chemical wood properties. Theor Appl Genet 104:214–222
to water deficit in Picea abies (L.) Karst. Ann For Sci Sewell MM, Neale DB (2000) Mapping quantitative traits in for-
59:533–540 est trees. In: Jain SM, Minocha SC (eds) Molecular Biology
Rudd S (2003) Expressed sequence tags: alternative or com- of Woody Plants. Kluwer, Dordrecht, pp 407–433
plement to whole genome sequences? Trends Plant Sci Sewell MM, Sherman BK, Neale DB (1999) A consensus map
8:321–329 for loblolly pine (Pinus taeda L.). I. Construction and in-
90 C. Plomion et al.

etegration if individual linkage maps from two outbred Strauss SH, Doerksen AH (1990) Restriction fragment analysis
three-generation pedigrees. Genetics 151:321–330 of pine phylogeny. Evolution 44:1081–1096
Shelbourne CJA (1992) Genetic gains from different kinds of Strauss SH, Lande R, Namkoong G (1992) Limitations of molec-
breeding population and seed or plant production popula- ular marker-aided selection in forest tree breeding. Can J
tions. S Afr For J 160:49–65 For Res 22:1050–1061
Shelbourne CJA (2000) Some insights on hybrids in forest Stromberg LD, Dudley JD, Rufener GK (1994) Comparing con-
tree improvement. In: Dungey HS, Dieters MJ, Nikles ventional early generation selection with molecular marker
DG (Compl) Hybrid Breeding and Genetics of For- assisted selection in maize. Crop Sci 34:1221–1225
est Trees. Proc QFRI/CRC-SPF Symp, 9-14 April 2000, Stuber CW, Polacco M, Senior ML (1999) Synergy of empiri-
Noosa, Queensland, Australia. Dept of Pri Indus, Brisbane, cal breeding, marker-assisted selection, and genomics to
pp 53–62 increase crop yield. Crop Sci 39:1571–1583
Shelbourne CJA, Apiolaza LA, Jayawickrama KJS, Sorensson Syring J, Willyard A, Cronn R, Liston A (2005) Evolution-
CT (1997) Developing breeding objectives for radiata pine ary relationships among Pinus (Pinaceae) subsections in-
in New Zealand. In: Burdon R, Moore JM (eds) “IUFRO ferred from multiple low-copy nuclear loci. Am J Bot
’97 Genetics of Radiata Pine”, Proc IUFRO Conf 1–4 Dec 92:2086–2100
and workshop 5 Dec, Rotorua, New Zealand. FRI Bull Syvanen AC (2001) Accessing genetic variation. Genotyping
203:60–168 single nucleotide polymorphisms. Nat Rev Genet 2:930–942
Shepherd M, Cross M, Dieters MJ, Henry R (2003) Genetic Szmidt AE, Muona O (1989) Linkage relationships of allozyme
maps for Pinus elliottii var. elliottii and P. caribaea var. loci in Pinus sylvestris. Heredity 111:91–97
hondurensis using AFLP and microsatellite markers. Theor Tanksley SD, Nelson JC (1996) Advanced backcross QTL analy-
App Genet 106:1409–1419 sis: a method for the simultaneous discovery and transfer
Shepherd M, Cross M, Dieters MJ, Henry R (2002b) Branch of valuable QTLs from unadapted germplasm into elite
architecture QTL for Pinus elliottii var. elliottii × Pinus breeding lines. Theor Appl Genet 92:191–203
caribea var. hondurensis hybrids. Ann For Sci 59:617–625 Telfer EJ, Echt CS, Nelson CD, Wilcox PL (2006) Comparative
Shepherd M, Cross M, Maguire TL, Dieters MJ, Williams CG, mapping in Pinus radiata and P. taeda reveals co-location
Henry RJ (2002a) Transpecific microsatellites for hard of wood density-related QTL. In: Plant & Animal Genome
pines. Theor Appl Genet 104:819–827 XIV Conf, San Diego
Shoulders E (1984) The case for planting longleaf pine. Proc of Temesgen B, Brown GR, Harry DE, Kinlaw CS, Sewell MM,
the Southern Silvicultural Res Conf, 7-8 Nov 1984, Atlanta, Neale DB (2001) Genetic mapping of expressed sequence
GA, pp 255–260 tag polymorphism (ESTP) markers in loblolly pine (Pinus
Smith DN, Devey M (1994) Occurrence and inheritance of mi- taeda L.). Theor Appl Genet 102:664–675
crosatellites in Pinus radiata. Genome 37:977–983 The Arabidopsis Genome Initiative (2000) Analysis of the
Soranzo N, Alia R, Provan J, Powell W (2000) Patterns of vari- genome sequence of the flowering plant Arabidopsis
ation at a mitochondrial sequence-tagged-site locus pro- thaliana. Nature 408:796–815
vides new insights into the postglacial history of European Thiellement H, Plomion C, Zivy M (2001) Proteomics as a tool
Pinus sylvestris populations. Mol Ecol 9:1205–1211 for plant genetics and breeding. In: Dunn MJ, Pennington S
Sorensson CT, Shelbourne CJA (2005) Clonal forestry. In: Col- (eds) Proteomics: from protein sequence to function. BIOS,
ley M (ed) Forestry Handbook. NZ Inst of Foresters, Oxford, pp 289–309
pp 92–96 Thoday JM (1961) Location of polygenes. Nature 191:368–370
Springer NM, Xu X, Barbazuk WB (2004) Utility of dif- Thumma BR, Nolan MF, Evans R, Moran GF (2005) Polymor-
ferent gene enrichment approaches toward identifying phisms in Cinnamoyl CoA Reductase (CCR) are associated
and sequencing the maize gene space. Plant Physiol with variation in microfibril angle in Eucalyptus spp. Ge-
136:3023–3033 netics 171:1257–1265
Stelzer HE, Goldfarb B (1997) Implementing clonal forestry in Travis SE, Ritland K, Whitham TG, Keim P (1998) A genetic
the southeastern United States: SRIEG satellite workshop linkage map of pinyon pine (Pinus edulis) based on am-
summary. Can J For Res 27:442–446 plified fragment length polymorphisms. Theor Appl Genet
Stirling B, Yang ZK, Gunter LE, Tuskan GA, Bradshaw HD 97:871–880
(2003) Comparative sequence analysis between ortholo- Tulsieram LK, Glaubitz JC, Kiss G, Carlson JE (1992) Single tree
gous regions of the Arabidopsis and Populus genomes re- genetic linkage mapping in conifers using haploid DNA
veals substantial synteny and microcollinearity. Can J For from megagametophytes. Bio/Technology 10:686–690
Res 33:2245–2251 Tuskan GA, DiFazio SP, Teichmann T (2004) Poplar genomics is
Strabala TJ (2004) Expressed sequence tag databases from getting popular: the impact of the poplar genome project
forestry tree species. In: Kumar S, Fladung M (eds) Molec- on tree research. Plant Biol (Stuttg) 6:2–4
ular Genetics and Breeding of Forest Trees. Haworth, Bing- Van der Burgh J (1973) Hölzer der nierderrheinischen
hamton, NY, pp 19–52 Braunkohlenformation. 2. Hölzer der Braunkohlengruben
 Chapter 2 Pines 91

,Maria Theresia’ zu Herzogenrath, ,Zukunft West’ zu Es- patterns of gene expression in drought-stressed loblolly
chweiler und ,Victor’ (Zülpich Mitte) zu Zülpich. Nebst pine. Plant Physiol 133:1702–1716
einer systematisch-anatomischen Bearbeitung der Gattung Wei RP, Lindgren D, Yeh FC (1997) Expected gain and sta-
Pinus L. Rev Palaeobot Palynol 15:73–275 tus number following restricted individual and combined-
van Tienderen PH, de Haan AA, van der Linden CG, Vosman B index selection. Genome 40:1–8
(2002) Biodiversity assessment using markers for ecologi- Weng C, Kubisiak TL, Nelson CD, Stine M (2002) Mapping
cally important traits. Trends Ecol Evol 17:577–582 quantitative trait loci controlling early growth in a (longleaf
van Zyl L, von Arnold S, Bozhkov P, Chen Y, Egertsdotter U, pine × slash pine) × slash pine BC1 family. Theor Appl
MacKay J, Sederoff R, Weir B, Shen J, Sun Y-H, Whetten R, Genet 104:852–859
Zelena L, Clapham D (2002) Heterologous array analysis in Weng C, Kubisiak TL, Stine M (1998) SCAR markers in a longleaf
Pinaceae: hybridization of Pinus taeda cDNA arrays with pine × slash pine F1 family. For Genet 5:239–247
cDNA from needles and embryogenic cultures of P. taeda, Whetten R, Sun Y, Zhang Y Sederoff RR (2001) Functional
P. sylvestris or Picea abies. Comp Funct Genom 3:306–318 genomics and cell wall biosynthesis in loblolly pine. Plant
Verhaegen D, Plomion C, Poitel M, Costa P, Kremer A (1998) Mol Biol 47:275–291
Quantitative trait disscetion analysis in Eucalyptus using Whitelaw CA, Barbazuk WB, Pertea G, Chan AP, Cheung F, Lee Y,
RAPD markers: 2. linkage disequilibrium in a factorial de- van Heeringen S, Karamycheva S, Bennetzen JL, SanMiguel
sign between E. urophylla and E. Grandis. For Genet 5:61–69 P, Lakey N, Bedford J, Yuan Y, Budiman MA, Resnick A, van
Vos P, Hogers R, Bleeker M, Reijans M, van de Lee T, Hornes Aken S, Utterback T, Riedmuller S, Williams SM, Feldblyum
M, Friters A, Pot J, Paleman J, Kuiper M, Zabeau M (1995) T, Schubert K, Beachy R, Fraser CM, Quackenbush J (2003)
AFLP: a new technique for DNA fingerprinting. Nucleic Enrichment of gene-coding sequences in maize by genome
Acids Res 23:4407–4414 filtration. Science 302:2118–2120
Wahlenberg WG (1946) Longleaf Pine. Charles Lathrop Pack Wilcox PL (1995) Genetic dissection of fusiform rust resistance.
Forestry Foundation, Washington, DC PhD Thesis, Dept of Forestry, North Carolina State Univer-
Wakamiya I, Price HJ, Messina MG, Newton RJ (1996) Pine sity, Raleigh, NC
genome size diversity and water relations. Physiol Planta Wilcox PL (1997) Linkage groups, map length, and recombi-
96:13–20 nation in Pinus radiata. In: Burdon RD, Moore JM (eds)
Walker S, Haines R, Dieters M (1996) Beyond 2000. In: Di- “IUFRO ’97 Genetics of Radiata Pine”, Proc IUFRO Conf
eters MJ, Matheson AC, Nikles DG, Harwood CE (eds) 1–4 Dec and workshop 5 Dec, Rotorua, New Zealand. FRI
Clonal Forestry in Queensland. Queensland For Res Inst. Bull 203
Tree Improvement for Sustainable Tropical Forestry. QFRI- Wilcox PL, Burdon RD (2006) Application of association ge-
IUFRO Conf, Caloundra, Queensland, Australia, 27 Oct-1 netics to coniferous forest trees. In: Proc 13th Australasian
Nov 1996, vol 2, pp 351–354 Plant Breed Conf, Christchurch, New Zealand pp 651–659,
Walter R, Epperson BK (2001) Geographic pattern of genetic [Link]
variation in Pinus resinosa: area of greatest diversity is not Wilcox PL, Amerson HV, Kuhlman EG, Liu B-H, O’Malley DM,
the origin of postglacial populations. Mol Ecol 10:103–111 Sederoff RR (1996) Detection of a major gene for resis-
Wang XR, Szmidt AE, Nguyen HN (2000) The phylogenetic tance to fusiform rust disease in loblolly pine by genomic
position of the endemic flat-needle pine Pinus krempfii mapping. Proc Natl Acad Sci USA 93:3859–3864
(Pinaceae) from Vietnam based on PCR-RFLP analysis of Wilcox PL, Richardson TE, Carson SD (1997) Nature of quan-
chloroplast DNA. Plant Syst Evol 220:21–36 titative trait variation in Pinus radiata: insights from QTL
Wang XR, Szmidt AE, Savolainen O (2001) Genetic compo- detection experiments. In: Burdon RD, Moore JM (eds)
sition and diploid hybrid speciation of a high mountain “IUFRO ’97 Genetics of Radiata Pine”, Proc IUFRO Conf
pine, Pinus densata, native to the Tibetan plateau. Genetics 1–4 Dec and workshop 5 Dec, Rotorua, New Zealand. FRI
159:337–346 Bull 203:304–312
Wang XT, Tsumara Y, Yoshimaru H, Nagasaka K, Szmidt Wilcox PL, Richardson TE, Corbet GE, Ball RD, Lee JR, Djorovic
A (1999) Phylogenetic relationships of Eurasian pines (Pi- A, Carson SD (2001a) Framework Linkage Maps of Pinus ra-
nus, Pinaceae) based on chloroplast rbcL, matK, rpl20-rps18 diata D. Don based on pseudotestcross markers. For Genet
spacer, and trnV intron sequences. Am J Bot 86:1742–1753 8:109–117
Ware D, Jaiswal P, Ni J, Pan X, Chang K, Clark K, Teytelman Wilcox PL, Carson SD, Richardson TE, Ball RD, Horgan GP,
L, Schmidt S, Zhao W, Cartinhour S, McCouch S, Stein Carter P (2001b) Benefit-cost analysis of DNA marker-
L (2002) Gramene: a resource for comparative grass ge- based selection in progenies of Pinus radiata seed orchard
nomics. Nucleic Acids Res 30:103–105 parents. Can J For Res 31:2213–2224
Watkinson JI, Sioson AA, Vasquez-Robinet C, Shukla M, Kumar Wilcox PL, Echt CS, Cato SA, McMillan LK, Kumar S, Ball RD,
D, Ellis M, Heath LS, Ramakrishnan N, Chevone B, Watson Burdon RD, Pot D (2003) Gene assisted selection – a new
LT, van Zyl L, Egertsdotter U, Sederoff RR, Grene R, et al paradigm for selection in forest tree species? In: Plant &
(2003) Photosynthetic acclimation is reflected in specific Animal Genomes XI Conf, San Diego
 92 C. Plomion et al.

Wilcox PL, Cato S, McMillan L, Power M, Ball RD, Bur- Yang SH, Loopstra CA (2005) Seasonal variation in gene ex-
don RD, Echt CS (2004) Patterns of linkage dis- pression for loblolly pines (Pinus taeda) from different
equilibrium in Pinus radiata. In: Plant & Animal geographical regions. Tree Physiol 25:1063–107325(8):
Genome XII Conf, San Diego, W89. [Link] Yazdani R, Nilsson JE, Plomion C, Mathur G (2003) Marker
[Link]/12/abstracts/W22_PAG12_89.html trait association for autumn cold acclimatation and growth
Wilcox PL, Echt CS, Burdon RD (2007) Gene-assisted selection: rhythm in Pinus sylvestris. Scand J For Res 18:29–38
applications of association genetics for forest tree breeding. Yazdani R, Yeh FC, Rimsha J (1995) Genomic mapping of Pinus
In: Oraguzie N, Rikkerink EHA, Gardiner SA, Nihal De Silva sylvestris (L.) using random amplified polymorphic DNA
H (eds) Association Mapping in Plants. Springer, Berlin markers. For Genet 2:109–116
Heidelberg New York, pp 211–247 Yin TM, Huang MR, Wang MX, Zhu LH, Zhai WX (1997) Con-
Williams CG, Neale DB (1992) Conifer wood quality and struction of molecular linkage map in masson pine using
marker-aided selection: a case study. Can J For Res RAPD markers and megagametophytes from a single tree.
22:1009–1017 Acta Bot Sin 39:607–612
Williams C, Megraw RA (1993) Juvenile-mature relation- Yin TM, Wang XR, Andersson B, Lerceteau-Köhler E (2003)
ships for wood density in Pinus taeda. Can J For Res Nearly complete maps of Pinus sylvestris L. (Scots pine)
24:714–722 constructed by AFLP marker analysis in a full-sib family.
Williams JGK, Kubelik AR, Livak KJ, Rafalski JA (1990) DNA Theor Appl Genet 106:1075–1083
polymorphisms amplified by arbitrary primers are useful Yuan Y, SanMiguel PJ, Bennetzen JL (2003) High-Cot sequence
as genetic markers. Nucleic Acids Res 18:6531–6535 analysis of the maize genome. Plant J 34:249–255
Williams CG, Zhou Y, Hall SE (2001) A chromosomal Zeng ZB (1993a) Theoretical basis for separation of multiple
region promoting outcrossing in a conifer. Genetics linked gene effects in mapping quantitative trait loci. Proc
159:1283–1289 Natl Acad Sci USA 90:10972–10976
Wright SI, Gaut BS (2005) Molecular population genetics and Zeng ZB (1993b) Precision mapping of quantitative trait loci.
the search for adaptive evolution in plants. Mol Biol Evol Genetics 136:1457–1468
22:506–519 Zeng ZB, Liu J, Stam LF, Kao CH, Mercer JM, Cathy CL (2000)
Wu H (2002) Study of early selection in tree breeding. 4. Ef- Genetic architecture of a morphological shape difference
ficiency of marker aided early selection (MAES). Silvae between two drosophila species. Genetics 154:299–310
Genet 51:261–269 Zhang H-B, Wu C (2001) BACs as tools for genome sequencing.
Wu J, Krutovskii KV, Strauss SH (1999) Nuclear DNA diversity, Plant Physiol Biochem 39:195–209
population differentiation, and phylogenetic relationship Zheng Y, Ennos R, Wang HR, Zheng YQ, Wang HR (1994) Prove-
in the Californian closed-cone pines based on RAPD and nance variation and genetic parameters in a trial of Pinus
allozyme markers. Genome 42:893–908 caribaea Morelet var. bahamensis Barr and Golf. For Genet
Wu R, Ma C-X, Casella G (2002) Joint linkage and linkage dis- 1:165–174
equilibrium mapping of quantitative trait loci in natural Zhou Y, Bui T, Auckland L, Williams CG (2002) Undermethyla-
populations. Genetics 160:779–792. tion as a Source of microsatellites in large plant genomes.
Yan HH, Mudge J, Kim DJ, Shoemaker RC, Cook DR, Young ND Genome 34:91–99
(2004) Comparative physical mapping reveals features of Zhou Y, Gwaze DP, Reyes-Valdes MH, Biu T, Williams CG (2003)
microsynteny between Glycine max, Medicago truncatula, No clustering for linkage map based on low-copy and un-
and Arabidopsis thaliana. Genome 47:141–155 dermethylated microsatellites. Genome 46:809–816

View publication stats