USDA Forest Service – National Agroforestry Center
USDA Forest Service / UNL Faculty
Publications
University of Nebraska - Lincoln
Year
5 Fagaceae Trees
Antoine Kremer, UMR Biodiversité Gènes & Communautés, INRA
Manuela Casasoli, Dipartimento di Biologia Vegetale, Università
“La Sapienza”, Piazza A. Moro 5, 00185 Rome, Italy
Teresa Barreneche, Unité de Recherche sur les Espèces Fruitières
et la Vigne, INRA, 71 Avenue Edouard Bourlaux, 33883 Villenave
d’Ornon, France
Catherine Bodénès, UMR Biodiversité Gènes & Communautés, INRA
Paul Sisco, The American Chestnut Foundation, One Oak Plaza,
Suite 308 Asheville, NC 28801, USA
Thomas Kubisiak, Southern Institute of Forest Genetics, USDAForest Service, 23332 Highway 67, Saucier, MS 39574-9344, USA
Marta Scalfi, Dipartimento di Scienze Ambientali, Università di
Parma, Parco Area delle Scienze 11/A, 43100 Parma, Italy
Stefano Leonardi, Dipartimento di Scienze Ambientali, Università
di Parma, Parco Area delle Scienze 11/A, 43100 Parma, Italy
Erica Bakker, Department of Ecology and Evolution, University of
Chicago, 5801 South Ellis Avenue, Chicago, IL 60637, USA
Joukje Buiteveld, AlterraWageningen UR, Centre for Ecosystem
Studies, P.O. Box 47, 6700 AAWageningen, The Netherlands
Jeanne Romero-Severson, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556, USA
Kathiravetpillai Arumuganathan, Flow Cytometry and Imaging Core
Laboratory, Benaroya Research Institute at Virginia Mason, 1201
Ninth Avenue, Seattle, WA 98101, USA
Jeremy Derory, UMR Biodiversité Gènes & Communautés, INRA
Caroline Scotti-Saintagne, UMR Ecologie des Forêts de Guyane,
INRA, Campus agronomique BP 709, Avenue de France, 97387
Kourou, French Guyana
Guy Roussel, UMR Biodiversité Gènes & Communautés, INRA
Maria Evangelista Bertocchi, UMR Biodiversité Gènes & Communautés, INRA
Christian Lexer, Jodrell Laboratory, Royal Botanic Gardens, Kew,
Richmond, Surrey TW9 3DS, UK
Ilga Porth, Austrian Research Centre, 2444 Seibersdorf, Austria
Fred Hebard, The American Chestnut Foundation Research Farms,
14005 Glenbrook Avenue, Meadowview, VA 24361, USA
Catherine Clark, Department of Forestry, North Carolina State University, Box 8008, Raleigh, NC 27695-8008, USA
John Carlson, The School of Forest Resources and Huck Institutes
for Life Sciences, Pennsylvania State University, 323 Forest Resources Building, University Park, PA 16802, USA
Christophe Plomion, UMR Biodiversité Gènes & Communautés,
INRA
Hans-Peter Koelewijn, AlterraWageningen UR, Centre for Ecosystem Studies, P.O. Box 47, 6700 AAWageningen, The Netherlands
Fiorella Villani, Istituto per l’Agroselvicoltura, CNR, V.le G. Marconi, 2 - 05010, Porano, Italy
This paper is posted at DigitalCommons@University of Nebraska - Lincoln.
http://digitalcommons.unl.edu/usdafsfacpub/46
CHAPTER 5
5 Fagaceae Trees
Antoine Kremer1 , Manuela Casasoli2 , Teresa Barreneche3 , Catherine Bodénès1 , Paul Sisco4 , Thomas
Kubisiak5 , Marta Scalfi6 , Stefano Leonardi6 , Erica Bakker7 , Joukje Buiteveld8 , Jeanne Romero-Severson9 ,
Kathiravetpillai Arumuganathan10 , Jeremy Derory1 , Caroline Scotti-Saintagne11 , Guy Roussel1 , Maria
Evangelista Bertocchi1 , Christian Lexer12 , Ilga Porth13 , Fred Hebard14 , Catherine Clark15 , John Carlson16 ,
Christophe Plomion1 , Hans-Peter Koelewijn8 , and Fiorella Villani17
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
UMR Biodiversité Gènes & Communautés, INRA, 69 Route d’Arcachon, 33612 Cestas, France, e-mail:
antoine.kremer@pierroton.inra.fr
Dipartimento di Biologia Vegetale, Università “La Sapienza”, Piazza A. Moro 5, 00185 Rome, Italy
Unité de Recherche sur les Espèces Fruitières et la Vigne, INRA, 71 Avenue Edouard Bourlaux, 33883 Villenave d’Ornon,
France
The American Chestnut Foundation, One Oak Plaza, Suite 308 Asheville, NC 28801, USA
Southern Institute of Forest Genetics, USDA-Forest Service, 23332 Highway 67, Saucier, MS 39574-9344, USA
Dipartimento di Scienze Ambientali, Università di Parma, Parco Area delle Scienze 11/A, 43100 Parma, Italy
Department of Ecology and Evolution, University of Chicago, 5801 South Ellis Avenue, Chicago, IL 60637, USA
Alterra Wageningen UR, Centre for Ecosystem Studies, P.O. Box 47, 6700 AA Wageningen, The Netherlands
Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556, USA
Flow Cytometry and Imaging Core Laboratory, Benaroya Research Institute at Virginia Mason, 1201 Ninth Avenue, Seattle,
WA 98101, USA
UMR Ecologie des Forêts de Guyane, INRA, Campus agronomique BP 709, Avenue de France, 97387 Kourou, French Guyana
Jodrell Laboratory, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3DS, UK
Austrian Research Centre, 2444 Seibersdorf, Austria
The American Chestnut Foundation Research Farms, 14005 Glenbrook Avenue, Meadowview, VA 24361, USA
Department of Forestry, North Carolina State University, Box 8008, Raleigh, NC 27695-8008, USA
The School of Forest Resources and Huck Institutes for Life Sciences, Pennsylvania State University, 323 Forest Resources
Building, University Park, PA 16802, USA
Istituto per l’Agroselvicoltura, CNR, V.le G. Marconi, 2 - 05010, Porano, Italy
5.1
Introduction
Worldwide, there are more than 1,000 species belonging to the Fagaceae. All Fagaceae species are woody
plants and are spread throughout the northern hemisphere, from the tropical to the boreal regions. The
family comprises seven genera (Govaerts and Frodin
1998), and the number of species is extremely variable among genera: Castanea (12), Castanopsis (100
to 200), Chrysolepis (2), Fagus (11), Lithocarpus (300),
Quercus (450 to 600), Trigonobalanus (3). Oaks (Quercus), chestnuts (Castanea), and beeches (Fagus) are
widely used in forestry for wood products over the
three continents (Asia, Europe, and America) and
are important economic species. Consequently, they
have received more attention in forest genetic research
than other genera. In addition to their cultivation in
forestry, chestnuts are also used for their fruit production and have been partially domesticated for that
purpose. Castanopsis and Lithocarpus are important
ecological components of the Asian flora and have
recently been investigated for their biological diversity (Cannon and Manos 2003). The remaining genera
comprise only a very few species and for the time being
have been studied mainly in botany and taxonomy.
Genetic research in Fagaceae has been restricted
to the three genera of economic importance (Castanea, Fagus, and Quercus), although activities in
phylogeny and evolutionary genetics have recently
encompassed the whole family (Manos and Stanford
2001; Manos et al. 2001). Because of their long rotation times, breeding activities in the three main genera have been limited (Kremer et al. 2004). However,
population differentiation has been investigated in
a very large number of species, with the main aim
of identifying geographic patterns or historical foot-
Genome Mapping and Molecular Breeding in Plants, Volume 7
Forest Trees
C. Kole (Ed.)
© Springer-Verlag Berlin Heidelberg 2007
162
A. Kremer et al.
prints for molecular markers and phenotypic traits
of forestry relevance. Population genetics has driven
most of the research activities in molecular genetics
and also genetic mapping, in contrast to other forest tree species where tree improvement has been the
main goal. Genetic maps have been constructed in
at least one species of Quercus, Castanea, and Fagus.
Because of their low genetic divergence, it quickly
became obvious that molecular markers could be easily transferred from Quercus to Castanea (and vice
versa) but less easily to Fagus. These earlier findings led to further activities on comparative mapping
across genera, especially between Quercus and Castanea.
similarity between Quercus, Castanea, Lithocarpus,
and Castanopsis. Paleontological records suggest that
Quercus and Castanea separated 60 million years ago
(Crepet 1989). Interspecific separation within the genera Quercus, Fagus, and Castanea occurred between
22 and 3 million years ago as inferred from a molecular
clock based on cpDNA divergence (Manos and Stanford 2001). The reduced genetic divergence among the
different genera was recently confirmed by the results
obtained in transferring molecular tools and markers
among genera, as it is much more difficult to transfer
microsatellite markers from Quercus to Fagus than it
is from Quercus to Castanea (Steinkellner et al. 1997;
Barreneche et al. 2004).
5.1.1
Evolutionary Biology and Phylogeny
of the Fagaceae
5.1.2
Ploidy, Karyotype, and Genome Size in Fagaceae
Fossil remains indicate that the Fagaceae appeared
at the transition between the secondary and tertiary
era. Remains of Dryophyllum, which is a fossil genus
belonging to the Fagaceae, were discovered in layers belonging to the early Cretaceous (Jones 1986).
Fossil remains that were unequivocally assigned to
Fagaceae and dated to the Upper Eocene and Late
Oligocene were found in North America (Herendeen
et al. 1995) and Europe (Kvacek and Walther 1989).
Differentiation of the various genera occurred during the mid Tertiary, and reported species of Fagaceae
at the late Tertiary resemble already extant species.
The oldest reported genera belonging to the Fagaceae
occurred in Southeast Asia, where the extant species
diversity is also the highest. The family originated
from Southeast Asia and radiated toward Europe and
America (Wen 1999; Xiang et al. 2000). Migration and
major continental rearrangements contributed to disjunction and vicariance within the family, especially
within Quercus (Manos and Stanford 2001). It is generally accepted that most major oak groups essentially
evolved in the areas where they reside today (Axelrod 1983).
Phylogenetic investigations based on chloroplast
or nuclear DNA data are poorly resolutive, suggesting that the differentiation into different genera was
extremely rapid during the mid Tertiary (Manos and
Steele 1997; Manos et al. 2001). All genera are usually
clustered into a “starlike” dendrogram (polytomy),
except Fagus, which diverged earlier from the common ancestor. However, there is a strong genomic
Reported karyotype studies in Quercus, Lithocarpus,
Castanopsis, and Castanea (Mehra et al. 1972), in
Quercus (D’Emerico et al. 1995), and Fagus (Ohri
and Ahuja 1991) indicate that the number of chromosomes within the family is remarkably stable (2n
= 24). Naturally occurring triploids have been mentioned occasionally in oaks (Butorina 1993; Naujoks
et al. 1995). Extra chromosomes (2n = 24+1, 2 or 3)
have been reported as consequences of irregular segregation in mitoses (Zoldos et al. 1998). Otherwise, Cbanding comparisons have shown that the morphology of the chromosomes of Fagus (Ohri and Ahuja
1991) and Quercus (Ohri and Ahuja 1990) are quite
similar.
The DNA content is variable across genera in the
Fagaceae: the 2C DNA values varying from a low
of 1.11 pg in Fagus to a high of 2.0 pg in Quercus
species (Table 1). GC content on the other hand appears constant across genera (40%) and is similar
to most higher plants (Table 1). All values reported
in Table 1 were obtained by flow cytometric analysis of interphasic nuclei and are slightly higher than
earlier assessments made with the Feulgen microdensitometry method (Ohri and Ahuja 1990). The
31 species in Table 1 represent a cross-section of
the Fagaceae across the northern hemisphere. The
two Fagus species, Fagus grandifolia and F. sylvatica, were quite similar in genome size (1.27 and
1.11 pg per 2C, respectively) and are at the lower
range of genome sizes among the Fagaceae, suggesting that Fagus has either the most rudimentary genome or the most greatly reduced genome
Chapter 5 Fagaceae Trees
163
Table 1. DNA content in 31 Fagaceae species determined by flow cytometric analysis
Species
2C nuclear DNA 1C nuclear DNA GC content
pg (mean value) Mbp
(%)
Reference
Genus Castanea
C. seguinii
C. sativa (1)
C. sativa (2)
1.57
1.61
1.62
755
777
–
–
–
–
C. crenata
C. mollissima
C. dentata
1.65
1.65
1.67
794
794
803
–
–
–
Arumuganathan et al.∗
Arumuganathan et al.
Brown and Siljak-Yakovlev
(pers comm)
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Genus Fagus
F. grandifolia
F. sylvatica
1.27
1.11
610
535
–
40
Arumuganathan et al.
Gallois et al. 1999
Genus Quercus Subgen Erythrobalanus∗∗
Q. velutina
Q. nuttallii
Q. shumardii
Q. nigra
Q. rubra
Q. palustris
Q. coccinea
Q. phellos
Q. falcata
Q. pagoda
Q. imbricaria
1.17
1.39
1.47
1.52
1.58
1.60
1.64
1.66
1.72
1.75
1.81
565
672
709
735
762
774
791
799
832
843
871
–
–
–
–
–
–
–
–
–
–
–
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Genus Quercus Subgen Lepidobalanus∗∗
Q. bicolor
Q. montana
Q. robur
Q. stellata
Q. alba
Q. macrocarpa
Q. robur
Q. pubescens
Q. petraea
Q. robur
Q. petraea
Q. pubescens
1.35
1.49
1.53
1.55
1.59
1.62
1.84
1.86
1.87
1.88
1.90
1.91
651
719
740
745
766
780
885
882
901
–
–
–
–
–
–
–
–
–
42
42.1
41.7
39.4
39.8
39.7
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Arumuganathan et al.
Favre and Brown 1996
Favre and Brown 1996
Favre and Brown 1996
Zoldos et al. 1998
Zoldos et al. 1998
Zoldos et al. 1998
Genus Quercus Subgen Cerris∗∗
Q. acutissima
Q. cerris
Q. suber
1.42
1.91
1.91
684
–
–
–
40.2
39.7
Arumuganathan et al.
Zoldos et al. 1998
Zoldos et al. 1998
Genus Quercus Subgen Sclerophyllodrys
Q. coccifera
Q. ilex
2.00
2.00
–
–
40.4
39.8
Zoldos et al. 1998
Zoldos et al. 1998
∗ Arumuganathan K, Schlarbaum SE, Carlson JE previously unpublished data (genome sizes are an average of three determinations
of 2 to 4 individuals per species)
∗∗ According to Flora Europea (http://rbg-web2.rbge.org.uk/FE/fe.html)
164
A. Kremer et al.
among the Fagaceae. In addition, the small genome
of Quercus velutina at 1.17 pg per 2C is essentially
the same as that of the Fagus species, again suggesting a basal genome size of about 1.2 pg per 2C
for the Fagaceae. Among the 24 Quercus species presented, the range of genome sizes is essentially continuous up to a maximum of 2.0 pg per 2C in Q. coccifera and Q. ilex. We looked for any indication that
the interspecific variation in the observed genome
sizes followed the botanical subdivisions within Quercus. We used here the classification into four distinct
botanical subgenera from Flora Europaea (http://rbgweb2.rbge.org.uk/FE/fe.html). This classification corresponds to earlier botanical descriptions of Schwarz
(1964) and Camus (1936-1954) and recent molecular analyses (Manos et al. 1999; Xu et al. 2005). The
species that were investigated include representatives
of all four subgenera: 12 species in Erythrobalanus
(red oaks), seven species in Lepidobalanus (white
oaks), three species in Cerris, and two in Sclerophyllodrys. The average 2C DNA contents were 2.0 pg for
subgenus Sclerophyllodris, 1.75 pg for subgenus Cerris, 1.73 pg for subgenus Lepidobalanus, and 1.56 pg
for subgenus Erythrobalanus. The two oak species
with the largest genomes, Q. coccifera and Q. ilex
(2.0 pg), are both evergreen species and are part of
a disputed botanical subgenus (named Sclerophyllodrys, according to Schwarz 1964). This is intriguing,
given that molecular phylogenetic analysis separates
the evergreen species from the two sections of deciduous oaks (Manos and Steele 1997 and Xu et al.
2005), confirming their earlier subdivision in Sclerophyllodrys by Schwarz (1964).
Among the five Castanea species studied, genome
sizes varied much less than among oaks, ranging only
from 1.57 pg in C. seguinii to 1.67 pg per 2C in C. dentata. In fact DNA content varied as much within the
chestnut species as between. For example, unrelated
C. seguinii trees varied from 1.5 to 1.63 pg per 2C,
while C. sativa varied from 1.57 to 1.65 pg per 2C
(Arumuganathan et al. this study). Thus there may
not be significant differences in average DNA content between Castanea species, and the range of average DNA content reported among species in Table 1 may just represent the natural variation in DNA
content among Castanea individuals. The intraspecific variation in DNA content in Quercus was also
as extensive as the amount of variation among the
species. For example, the 2C DNA content varied between 1.88 pg and 2.0 pg among Q. petraea trees of
the same populations (Zoldos et al. 1998), between
1.45 and 1.96 in Q. pagoda, and between 1.34 and 1.78
in Q. macrocarpa (Arumuganathan et al. this study).
The intraspecific variation may be due in part to the
occurrence of extra B chromosomes (Ohri and Ahuja
1990; Zoldos et al. 1998). While the range of DNA
content among oak species appears to be greater than
among chestnut and beech species, the magnitude
of the differences among oak species may be related
to experimental issues as well as biological ones. For
example, the size estimates by Arumuganathan et al.
(this study) were consistently smaller than those by
Favre and Brown (1996) and Zoldos et al. (1998). One
could speculate that the differences relate to the fact
that Arumuganathan et al. (this study) studied primarily New World species, while the other two studies dealt exclusively with Old World species. However,
the three groups report different genome sizes for
Q. robur (1.53, 1.84, and 1.88 pg per 2C, respectively).
Whether this discrepancy has a biological basis (the
Arumuganathan et al. study sampled three trees of
Q. robur “fastigiata,” the “upright” horticultural variety) or resulted from experimental differences in
sampling, internal size standards, and other methodologies is not clear.
In general, the genome sizes in the Fagaceae are
only 3.5 to 6 times larger than the genome of Arabidopsis (0.32 pg; Bennett et al. 2003) and are within the
size range of the sequenced rice and poplar genomes
(both 1.0 pg; Brunner et al. 2004). Comparative genomics should thus be relatively efficient within the
Fagaceae. Comparative genomics will lead to a better understanding of the extent to which the continuous range of DNA content is related to adaptive radiation of the species during evolution or is
the result of overlapping ranges and interspecies hybridizations. Knowledge of the genome sizes reveals
that genome-level comparisons between Fagus sylvatica, Q. velutina, Q. coccifera, and Q. ilex would
be particularly informative and could illuminate the
role of genome duplication in the evolution of the
Fagaceae. When comparative genomics studies are
extended to more species within the Fagaceae, it
will be interesting to determine whether or not the
broader range of genome sizes observed in Quercus relates to more extensive adaptations and specializations than exist among Fagus and Castanea
species. Given the extensive natural populations of
Fagaceae species that still exist across the northern
hemisphere, such information will certainly provide
insights into the ecology of temperate forest ecosystems.
Chapter 5 Fagaceae Trees
5.2
Construction of Genetic
Linkage Maps
165
5.2.2
Genetic Mapping Initiatives in Fagaceae
Oak Mapping Initiatives
5.2.1
Genetic Mapping in Forest Trees
PCR-based molecular markers and the two-way pseudotestcross strategy are useful tools for constructing genetic maps in forest trees (Grattapaglia and
Sederoff 1994). These outbred species are characterized by long generation times, long life spans, and
a high genetic load that often leads to significant
inbreeding depression. Although all these elements
hinder the development of the type of mapping populations normally used for genetic linkage mapping
(for instance inbred lines and backcrosses), the high
level of heterozygosity in forest species made twogeneration full-sib pedigrees suitable populations for
marker segregation analysis. Full-sib and half-sib
crosses can, therefore, be used to construct singletree genetic linkage maps thanks to dominant PCRbased molecular markers. Following this approach,
called the two-way pseudotestcross strategy (Grattapaglia and Sederoff 1994), three types of segregation
configurations can be obtained for dominant molecular markers in the mapping population: (1) male
testcross markers, segregating in a 1:1 ratio and inherited from the male parent; (2) female testcross
markers, segregating in a 1:1 ratio and inherited from
the female parent; and (3) intercross markers, segregating in a 3:1 ratio and inherited from both parental
trees. Male and female testcross markers are used to
construct two independent single-tree genetic maps
that are then aligned thanks to the intercross markers. RAPD (Williams et al. 1990) and AFLP (Vos et al.
1995) dominant molecular markers have been used
most commonly to construct genetic linkage maps in
forest tree species (Verhaegen and Plomion 1996; Marques et al. 1998; Arcade et al. 2000; Costa et al. 2000;
Cervera et al. 2001), as their random distribution in
the genome allows all chromosomes to be covered
most efficiently.
The two-way pseudotestcross strategy was first applied in forest trees by Grattapaglia et al. (1995) to
identify loci controlling quantitative trait loci (QTLs).
In forest genetics, QTL analysis has been one of the
most important applications of linkage mapping, and
several studies reported successful QTL detections
(Sewell and Neale 2000; van Buijtenen 2001).
European white oaks Starting in 1995, activities
in genetic mapping were implemented in European
white oaks at the INRA Research Centre in BordeauxCestas (France). Motivations for genetic mapping in
oaks were threefold: (1) the detection of genomic regions involved in species differentiation, (2) the detection of QTLs controlling traits of adaptive significance, and (3) the comparative analysis of genomic
evolution in the Fagaceae. The whole mapping project
is based on three pedigrees: one full-sib family of
Quercus robur (3P × A4), one full-sib family of Q. petraea (QS28 × QS21), and one interspecific F1 fullsib family Q. robur × Q. petraea (11P × QS29). An
interspecific F2 cross is planned as well. Given the
objectives of the mapping experiments, the parents
of the pedigrees were not selected for any particular
criteria. The Q. robur parent trees originated from
the southwest of France (INRA research station of
Bordeaux-Cestas, and Arcachon) and the Q. petraea
parents were from the central part of France (INRA
research station of Orléans-Ardon). The controlled
crosses were repeatedly done over successive years
until 2004. From 200 to 1,000 seeds were obtained
for each cross. The young seedlings were installed
in a seedbed in a nursery, where they are raised as
stool beds. Starting at age 5, the full-sibs were hedged
every year at the ground level at the end of winter.
Following the hedging, stump sprouts developing in
spring were harvested and cut in 15- to 20-cm-long
cuttings. These cuttings were then transplanted in
field tests for phenotypic observations and further
QTL detection. For the time being, only the Q. robur
intraspecific cross has been fully exploited for genetic
mapping and QTL detection. The clonal test of the
full-sibs has now been planted in three different sites
(two near Bordeaux, southwest France, and one near
Nancy, northeastern France). The genetic mapping of
Q. robur mapping was done on a sample of 94 offspring (pedigree 3P × A4), and the QTL detection on
a sample of 278 offspring (replicated on average in five
vegetative propagules).
Another mapping initiative for Q. robur was implemented in the Netherlands (ALTERRA, Wageningen). This mapping pedigree consists of 101 full-sibs
(Bakker 2001). The sibs were screened by paternity
analysis within an open-pollinated progeny set of 397
166
A. Kremer et al.
sibs collected on a single tree located in an urban
area (Amsterdam). This tree was surrounded by three
other oak trees within a radius of 10 m. One of these
oak trees was selected to be the paternal tree. Paternity analysis revealed that 26% of the collected seeds
were sired by this male parent. The selected seeds were
germinated, grown individually in pots in a nursery
(ALTERRA research station, Wageningen, the Netherlands), and measured for several morphological and
physiological traits during the next 2 years (1999,
2000). The objective of this work was quite similar
to the French initiative: detection of QTL controlling
morphological and adaptive traits involved in species
differentiation.
tained from a controlled cross performed between two
highly differentiated trees originating from Turkey.
Anatolia Peninsula was shown to be an important
region for chestnut genetic diversity (Villani et al.
1991, 1992). As illustrated by these studies, a remarkably high level of genetic, morphological, and
physiological differentiation was observed between
two groups of chestnut populations coming from
two phytogeographic regions, characterized by striking climatic differences: the Eurosiberian part of the
peninsula in northeastern Anatolia (humid) and the
Mediterranean region in western Anatolia (xeric).
Common field experiments carried out at the experimental field site of Istituto di Biologia Agroambientale e Forestale, CNR (Porano, Italy), showed
significant differences between these populations in
growth rate, bud flush, and physiological parameters, related to the water use efficiency, allowing
“drought-adapted” and “wet-adapted” ecotypes to be
identified (Lauteri et al. 1997, 1999). Differences observed in the ecophysiological behavior suggested that
Turkish chestnut populations are genetically adapted
to contrasting environments, making them a suitable material to study the adaptive potential of this
species.
The controlled cross was performed in 1998
between a female parent (Bursa) belonging to the
“drought-adapted” type from western Turkey and
a male parent (Hopa) belonging to the “wet-adapted”
type from eastern Turkey. The parental individuals
were 9 years old and were chosen according to their
heterozygosity level at isozymes and high degree of
variation in physiological traits. An F1 full-sib family
of 186 offsprings was obtained, and 96 F1 individuals
were used to construct two separate genetic linkage
maps: a female or Bursa map and a male or Hopa
map. The main objective of the project was to exploit
the peculiar genetic and adaptive variation observed
in these populations in order to identify the genomic
regions affecting carbon isotope discrimination
(related to the water use efficiency), bud phenology,
and growth by means of QTL analysis.
American red oaks Genetic mapping in northern
red oak (Quercus rubra L.) was initiated at Purdue University (http://www.genomics.purdue.edu/forestry/;
Romero-Severson 2003) and has continued at the
University of Notre Dame. Using exclusion methods
based on microsatellite polymorphisms (Aldrich
et al. 2002, 2003a), a preliminary mapping population
of 97 full-sibs was identified from the open-pollinated
progeny of a single tree. The most likely male parent
male was the closest conspecific. Recombination
patterns revealed (Romero-Severson et al. 2003) six
linkage groups (LGs) of three or more markers.
A second acorn harvest from the same female parent
yielded 462 full-sibs. The genetic map under construction now includes 15 microsatellites, 66 AFLP
markers from the first set of progeny, and several
hundred new AFLP markers from the second set of
progeny. All of the potential pollen parents within
200 m of the female parent are being genotyped
with all 15 microsatellite markers to eliminate any
doubt over the full-sib status of the mapping population. The microsatellite markers used for genetic
mapping are the same as those used for studies
on interspecific gene flow (Aldrich et al. 2003b)
and in northern red oak genetic diversity studies.
No map has yet been published for Q. rubra. The
long-term goal of the red oak mapping project is the
detection of QTLs and genes controlling heartwood
color and resistance to specific pests, specifically American and Chinese chestnuts During the
Phytophthora ramorum, the agent of sudden oak last century, American Chestnut, Castanea dentata
death.
(Marsch) Borkh, one of the most important timber
and nut-producing tree species in eastern North
Chestnut Mapping Initiatives
America, was dramatically affected by a canker
European chestnut Starting in 1998, a genetic map- disease (chestnut blight) caused by Cryphonectria
ping project for European chestnut (Castanea sativa parasitica. American chestnut showed low levels
Mill.) was implemented using a full-sib family ob- of resistance to blight, whereas Asian chestnut
Chapter 5 Fagaceae Trees
species (Castanea crenata (Japanese chestnut) and
C. mollissima (Chinese chestnut) exhibited higher
levels of resistance to the disease. During the 1980s
an important backcross breeding program was
undertaken in the USA in order to obtain selected
material combining the blight resistance of Asian
chestnut and good timber qualities of American
chestnut (Burnham et al. 1986).
In this context, a genetic map for chestnut was constructed. The main objective of this mapping project
was to identify genomic regions involved in blight resistance. In addition, the map was also used to locate loci controlling morphological traits that differentiated both species. The mapping population
was F2 progeny derived from a cross between two
C. mollissima × C. dentata F1 hybrids. The female
parent was the C. mollissima cultivar “Mahogany”
and two different American chestnut trees from Roxbury, CT were used as male to create the F1 hybrids. One hundred and two F2 individuals were used
for the map construction, and 185 individuals were
assessed for resistance to Cryphonectria parasitica.
Beech Mapping Initiative
A genetic mapping project for European beech (Fagus sylvatica) has been implemented at the University
of Parma (Italy) during the last 10 years (Scalfi et al.
2004). The objective was to dissect important adaptive
traits and to identify their underlying QTLs to detect
genomic regions involved in important quantitative
traits such as growth, phenology, and water-use efficiency. The mapping pedigree consisted of a full-sib
family comprising 143 offsprings. The family was the
largest in a 4 × 4 diallel controlled cross performed
in 1995 (Ceroni et al. 1997). The parents originated
from a natural population located at high altitude in
northern Italy (1,650 m altitude, just below the tree
line).
5.2.3
Genetic Linkage Maps
for Quercus, Castanea, and Fagus
Genetic Map of Q. robur
The first Quercus map was published in 1998 on
Q. robur (Barreneche et al. 1998) (pedigree 3P ×
A4). Using the pseudotestcross mapping strategy, two
maps were constructed comprising 307 markers (271
RAPD, 10 SCARs, 18 SSRs, 1 minisatellite, 6 isozymes,
and 1 ribosomal DNA marker). Both maps provided
167
85 to 90% coverage of the Q. robur genome. Segregating markers could be aligned in 12 LGs, and the
map size amounted to 893.2 cM for the paternal and
921.7 cM for the female map. These maps were further
upgraded by the inclusion of new SSRs (Barreneche
et al. 2004) and additional AFLP and STS. The upgrading is still ongoing and to date 854 markers (271
RAPD, 457 AFLP, 10 SCAR, 59 SSR, 49 EST, 1 minisatellite, 6 isozymes, and 1 ribosomal DNA marker)
have been located (Table 2).
The Dutch Q. robur map (pedigree A1 × A2) was
also constructed using the two-way pseudotestcross
strategy (Bakker 2001). Two parental maps were first
established comprising 18 SSR and 343 AFLP markers.
The total lengths of the maternal and paternal maps
were respectively 496 and 566 cM. Thirteen LGs were
obtained (for 12 chromosomes) and the two maps
could be partially merged using 58 “bridge” markers (2 LGs could not be aligned). One of the paternal
LGs (LG 13, 27 cM) was highly dissimilar to the other
LGs in terms of marker density. This LG contained
almost half (48%) of all paternal markers and 22% of
the segregating (heterozygote) markers. This markerdense LG was homologous to one of the maternal LGs
that remarkably was composed exclusively of 13 segregating markers. Congruence of LGs with the French
map was based on the location of SSR markers (Sect.
5.3.1). The total map length of the integrated map was
659 cM, map density being one marker per 2.4 cM for
the map without taking the exceptionally dense LG 13
into account.
Genetic Map of Castanea sativa
A first framework of the chestnut genetic linkage map
was obtained with RAPD and ISSR markers (Casasoli
et al. 2001). Few isozyme loci were integrated in this
first version of the map. A total of 381 molecular markers segregated in the chestnut full-sib family covering
a good portion of the chestnut genome (more than
70%). Intercross segregating markers allowed 11 of
the 12 LGs identified to be aligned between the female
and male maps. This original framework was then
used to map AFLP markers and codominant locusspecific markers such as SSR- and EST-derived markers. Table 2 shows the number and type of molecular markers contained in the chestnut genetic linkage
map. At present, 517 molecular markers have been
mapped in chestnut covering 80% of its genome. The
12 LGs were aligned to obtain 12 consensus female
and male LGs (chestnut linkage consensus groups are
available at the Web site www.pierroton.inra.fr).
168
Table 2. Summary of genetic linkage maps of Quercus robur and Castanea sativa
Number
of LGs
Total number
of marker
loci
RAPD,
ISSR,
AFLP
SSR
STS
Isozymes
% of
distorted
markers
Total
genetic
distance
(cM)
Genome
saturation
(%)
Total
size
(cM)
Ref.
Q. robur
(3P × A4)
12
854
728
59
61
6
18
950
80
1200
Barreneche et al. 2004;
and this study
Q. robur
(A1 × A2)
13
361
343
18
–
–
17.5
659
64
1035
Bakker (2001);
and this study
C. sativa
(Bursa × Hopa)
12
517
427
39
46
5
10
865
82
1050
Casasoli et al. 2001;
and this study
C. mollissima ×
C. dentata
(Mahogany)
12
559
521
29
1
8
25
524
–
–
Kubisiak et al 1997;
Sisco et al. 2005
F. sylvaticaa)
(44 × 45)
12/11
138/124
128/113
10/6
9
–
23
844/971
a)
The two numbers indicated correspond to the map of the female and male parent
78/82
1081/1185
Scalfi et al. 2004;
Scalfi 2005
A. Kremer et al.
Species
(pedigree)
Chapter 5 Fagaceae Trees
Genetic Map of Castanea mollissima/
Castanea dentata
The C. mollissima/C. dentata map was the first to be
published in the Fagaceae (Kubisiak et al. 1997). At
first a total of 241 markers, including 8 isozymes, 17
RFLPs, 216 RAPDs, were mapped in the F2 family.
Twelve LGs were identified, covering a genetic distance of 530.1 cM (corresponding to 75% of the chestnut genome). To saturate the map, additional markers were recently added to the initial map: 275 AFLP
(Clark et al. 2001) and 30 STS (29 SSR and the 5SrDNA
locus) (Sisco et al. 2005). To date, a total of 559 markers
have been located. Relatively high levels of segregation distortion (more than 25%) have been reported
in the C. mollissima/C. dentata family. Skewed segregation is a common feature in progenies resulting
from interspecific crosses.
Genetic Map of Fagus sylvatica
The Fagus genetic linkage map was based on a total of
312 markers: 28 RAPDs, 274 AFLPs, and 10 SSRs. Two
maps were constructed using the “double testcross”
strategy. In the female map 132 markers were distributed in 12 LGs covering 844 cM. In the male parent
only 11 LGs were identified, resulting in linkage relationships between 119 markers spanning over 971 cM
(Table 2). The two maps cover about 78% and 82% of
the Fagus genome. Using intercross markers (15 AFLP
and 2 SSR) seven homologous LGs could be identified
(Scalfi et al. 2004). Ten additional EST markers were
then added to the map since its publication (Scalfi
2005).
5.3
Comparative Mapping
between Quercus, Castanea,
and Fagus
5.3.1
Mapping of Microsatellites in Quercus robur,
Castanea sativa, C. mollissima, and C. dentata
Microsatellite markers, which are tandemly repeated
units of 2 to 6 nucleotides evenly dispersed throughout plant genomes, have been sometimes used for
comparative mapping studies (Marques et al. 2002).
Amplification of orthologous SSR markers across phylogenetically related species depends largely on evolutionary distance and genome complexity of com-
169
pared species (Powell et al. 1996). Usually, SSR crossamplification is more efficient between closely related
species with a low proportion of highly repeated sequences in their genome. Steinkellner et al. (1997)
showed that microsatellite markers specifically developed in Quercus species were cross-amplified in
chestnut. For these reasons microsatellites were supposed to be useful molecular markers for comparing
Q. robur and C. sativa genetic linkage maps. To obtain
orthologous markers for comparative mapping, SSR
markers developed both in Quercus species and in
C. sativa were therefore tested for cross-amplification
and transferability between these two genera (Barreneche et al. 2004 and references therein) in a reciprocal way. We tested a total of 83 primer pairs:
53 developed in Quercus species and 30 in C. sativa.
Primer pairs giving a strong amplification product
were selected for mapping. Nineteen loci, 15 from oak
and 4 from chestnut, were integrated into the two previously established genetic maps, allowing the first
comparative mapping between LGs of the two species
(Barreneche et al. 2004). Figure 1 shows the seven
homeologous LGs identified by orthologous SSR and
all microsatellite loci mapped in Q. robur and C. sativa
genetic maps. These same SSR loci were used to align
the European chestnut genetic linkage map with the
C. mollissima × C. dentata interspecific map. Eleven
of the 12 LGs of the two maps could be associated,
nine LGs were aligned on the basis of pairs, triplets,
or quadruplets of common markers, while three additional groups were matched using a single SSR marker
(Sisco et al. 2005).
Overall, these findings showed that microsatellite
markers could be cross-transferred between Quercus and Castanea genera and be used to recover orthologous markers for comparative mapping. Nevertheless, cross-transferability efficiency was low and
the number of cross-transferred loci was not sufficient to link the 12 LGs of the two species. As expected, SSR loci were extremely useful for comparative mapping within the same genus (Castanea),
but their cross-transferability efficiency decreased
between different genera. SSR loci mapped both in
Q. robur and C. sativa were sequenced in order to
definitely demonstrate their orthology. Sequencing
results clearly showed that both orthologous and paralogous loci could be recovered among the SSR crosstransferred between the two genera. Moreover, indels
were sometimes observed within the flanking regions
of the repeated motif. Therefore, although SSR loci
can be cross-transferred between Quercus and Cas-
170
A. Kremer et al.
Fig. 1. Assignment between Q. robur and C. sativa based on orthologous microsatellites. Oak (Q, green on left, pedigree 3P × A4) and chestnut (C, light blue on right, pedigree Bursa
× Hopa) LGs aligned using microsatellite markers. LGs are named as in Barreneche et al. (1998) and in Casasoli et al. (2001). Oak LGs are taken as reference and arranged in sequence
from Q1 to Q12. Nine chestnut LGs, aligned with the corresponding oak LGs, are given on the right. The three remaining chestnut LGs are reported according to the oak LGs. Common
orthologous SSR markers are shown in red. The EMCs1 marker was later shown to be a paralogous locus, and LG Q7 was not homeologous to LG C11 (Fig. 2). The figure, modified from
Barreneche et al. (2004), was drawn using MapChart software (Voorrips 2002)
Chapter 5 Fagaceae Trees
171
tanea genera, a sequence analysis is needed to demon- molecular markers mapped in the two species, no
strate orthology and to avoid the risk of paralogy. major chromosomal rearrangements have been identified, suggesting that oak and chestnut genomes are
quite stable. Thus it appears likely that the “single
genetic system” model of the grass genomes (Gale
and Devos 1998) can also be applied to Q. robur and
5.3.2
C. sativa. EST-derived markers were very easily transMapping of EST-Derived Markers in Q. robur
ferred from oak to chestnut. About 50% of them conand C. sativa: Alignment of the 12 Linkage
tained intron-derived sequences. This increased the
Groups between the Two Species
probability of detecting segregating polymorphisms
Several factors make EST (expressed sequence tag)- useful for mapping in both oak and chestnut full-sib
derived markers very useful for comparative map- families. These markers proved to be ideal markers
ping studies (Brown et al. 2001). First, ESTs are se- for comparative mapping within the Fagaceae family.
quence fragments of coding regions; therefore sequence conservation among species is expected to
be higher than that observed, for instance, in SSR 5.3.3
loci. Second, ESTs correspond very often to genes Mapping of Microsatellites and EST-Derived
of known function. This is of great interest because Markers in Fagus sylvatica, Quercus robur,
some ESTs colocalized with QTLs in a genetic link- and Castanea sativa
age map may be putative positional candidate genes
for a given trait. Finally, transcriptome analyses give Success of transferability between Fagus sylvatica,
rise to a high number of EST sequences that are the Quercus robur, and Castanea sativa was lower. Alsource of numerous EST-derived markers distributed though 86 SSR markers originally developed in other
throughout plant genomes. In oak, ESTs were devel- Fagaceae species were tested in Fagus (66 from Queroped by Derory et al. (2006) and Porth et al. (2005a). cus, 20 from Castanea), only seven produced an interThis gave the opportunity to exploit EST sequence in- pretable banding pattern and only one marker from
formation for marker design in order to complete the Q. rubra and one from C. sativa could be placed on
comparative mapping between Q. robur and C. sativa the beech map (Scalfi 2005). One marker originally
(Casasoli et al. 2006). About 100 EST sequences were developed in Fagus gave good amplification also in
selected from oak databases. Oak sequences were Quercus and Castanea but was monomorphic in the
aligned with homologous sequences obtained from crosses used for these species.
GenBank in order to design primer pairs for amSimilarly, 86 EST markers originally developed in
plification in the most conserved regions of the se- Quercus were tested in beech, 46 coming from a budquence and assure a good cross-amplification effi- burst c-DNA library (Derory et al. 2006), 22 from osciency in chestnut. A total of 82 primer pairs were de- motic stress response (Porth et al. 2005a), and 17 from
signed. A proportion of about 70% produced by PCR hypoxia response cDNA-AFLP markers (C. Bodénès
a single and strong band both in oak and chestnut unpublished results). The success rate was higher than
and 51 and 45 ESTs were mapped in oak and chest- for microsatellites. In total 16 were polymorphic using
nut, respectively, using single strand conformation various techniques (SSCP, DGGE, sequencing, CAPS,
polymorphism (SSCP) and denaturing gradient gel dCAPS), and 10 were finally mapped onto the beech
electrophoresis (DGGE) approaches (Casasoli et al. map (Scalfi 2005).
2006). These EST-derived markers, together with SSR
Two markers (1T11 and 1T62) that mapped on
markers previously mapped, provided 55 orthologous Quercus and Castanea on LG 10 (Table 4) were mapped
molecular markers that allowed the 12 LGs of Q. robur also on group 4 in Fagus with the help of a “bridge”
and C. sativa to be aligned. As shown in Fig. 2, from marker (1T41): this can be considered as evidence
2 to 7 common orthologous markers were mapped of synteny between LG10 of Q. robur (3P*A4) and
in the 12 homeologous pairs of LGs. Macrosynteny C. sativa with LG 4 of Fagus. For the two markers
and macrocollinearity were well conserved between the sequence homology of Fagus with Quercus was
the two species. Few inversions, probably due to map- 82% and 43%, respectively; the lower value is due to
ping errors, were observed. Although these data are a large insertion in the beech sequence that was not
still preliminary given the low number of common present in the cDNA of Quercus. Eliminating the gap,
172
A. Kremer et al.
Fig. 2. Comparative mapping between Q. robur and C. sativa. The 12 homeologous LGs between Q. robur (Q, green, pedigree 3P × A4) and C. sativa (C, light blue, pedigree Bursa
× Hopa). The orthologous molecular markers mapped in both species are shown in red (SSRs and EST-derived markers). A subsample of molecular markers of the oak and chestnut
consensus genetic linkage maps (available at www.pierroton.inra.fr) is shown in this figure. Orthologous molecular markers mapped in a different oak cross (or showing a low mapping
statistical support, Cons 75 in the oak LG Q3) are marked in blue below the LGs
Chapter 5 Fagaceae Trees
the homology increased to 92%. Synteny could not
be assessed for any other group since none had more
than one marker mapped on it. For example, marker
2T32 mapped on LG 2 in Quercus was found linked
to markers on LG 7-F in beech, but more than one
comapping marker is needed to establish synteny.
5.3.4
Assignment of Linkage Groups
Between Quercus and Castanea
Most genetic maps constructed within Fagaceae
species comprised 12 LGs (Table 2). Crosstransferable SSR- and EST-derived markers made
it possible to assign LGs among the four species
Q. robur, C. sativa, C. mollissima, and C. dentata
(Table 4). However, the assignment is still based on
a limited number of markers per LG. Assignment was
done by pairwise comparisons:
– Between the two Q. robur maps (3P × A4 and A1 ×
A2): 11 out of 12 LGs had at least two orthologous
SSRs in common; the remaining LG was assigned
by default.
– Between Q. robur (3P × A4) and C. sativa (Bursa
× Hopa): between two and seven pairs of common
markers (either SSR, isozymes, or EST) allowed the
LGs to be assigned.
– Between the two chestnut maps (Bursa × Hopa
and Mahogany): the assignment is still incomplete
as only nine LGs could be assigned so far by at least
two pairs of SSRs.
The results obtained so far need to be confirmed by
further mapping experiments, based mainly on EST
markers. They are also encouraging as suggested by
the conservation of the macrosynteny and macrocollinearity that have so far been observed between
the two most intensively studied species: Q. robur and
C. sativa.
5.4
Genes Mapped in Oaks and Chestnut
Transcriptomic investigations and differential gene
expression studies were implemented recently with
the main aim of identifying genes that are involved in
the adaptation of oak or chestnut trees to their environment. Gene expression was monitored for different
traits, or tissues:
–
–
–
–
–
173
Bud burst in oaks
Hypoxia in oaks
Osmotic stress in oaks
Juvenile and mature shoots in oaks
Blight infection in chestnuts
Various techniques were implemented for constructing expression profiles: cDNA-AFLP, SSH, and
Quantitative RT-PCR. We will briefly summarize the
experiments conducted and the functions of genes
that were identified. Table 3 provides a list of ESTderived markers mapped in Q. robur and C. sativa.
For each EST the accession number, amplification, sequencing, and mapping results are reported.
5.4.1
Bud Burst
Candidate genes for bud burst were identified in Q. petraea using SSH libraries, macroarray experiments,
and RT-PCR. Three subtracted libraries (SSH method)
were constructed, generating 801 ESTs derived from
six developmental stages of bud burst. Expression patterns of these transcripts were monitored in apical
buds during bud flushing in order to identify genes
differentially expressed between the quiescent and active stage of bud development. After bioinformatic
processing of the ESTs, macroarray experiments revealed a total of 233 unique transcripts exhibiting differential expression during the process, and a putative function was assigned to 65% of them (Derory
et al. 2006). Cell rescue/defense-, metabolism-, protein synthesis-, cell cycle-, and transcription-related
transcripts were among the most regulated genes. Reverse northern and RT-PCR showed that several genes
exhibited contrasting expression between quiescent
and swelling buds. Among this set of 233 unique transcripts, ca. 100 were selected and tentatively amplified
and mapped in oak and chestnut, as previously described. In oak and chestnut, 51 and 45 ESTs were successfully mapped, respectively, using SSCP and DGGE
approaches (Casasoli et al. 2006).
5.4.2
Hypoxia
Q. robur and Q. petraea exhibit different responses
to hypoxia, the first one being more tolerant to waterlogged conditions. Hypoxia-induced genes were
identified from vegetative copies of the two species
174
Table 3. List of genes mapped in both Quercus robur and Castanea sativa
Accession
number
Reference for primer
sequences and
PCR protocols
Expected
size (bp)b
Observed
size
Identityc
Linkage
group
Q–Cd
Functional
category
1T11
1T12
1T21
1T25
1T57
1T62
CF369263
CF369264
CF369266
CF369268
CF369273
CF369274
Porth et al. 2005b
Porth et al. 2005b
Porth et al. 2005b
Porth et al. 2005b
Porth et al. 2005b
Porth et al. 2005b
555
522
338
187
282
346
99
93.5∗
94
97.2
93.5
91.9
10 – 10
nm – 3
1–6
6 – 11
4–2
10 – 10
Unknown
Metabolism
Protein synthesis
Unknown
Transcription
Metabolism
2T11
CF369278
Porth et al. 2005b
397
89∗
nm – 8
Cell rescue, defense and virulence
2T3
2T13
2T32
CF369283
CF369280
CF369284
Porth et al. 2005b
Porth et al. 2005b
Porth et al. 2005b
284
334
386
94.3
94.5
93
10 – 10
11 – 3
2–1
Unclassified (plasma membrane related?)
Metabolism
Protein synthesis
01A03
1,00E+07
CR627501
CR627526
Casasoli et al. 2006
Casasoli et al. 2006
382
145
94.3
86
7–5
5d–4
Protein synthesis
Transcription
02F02
02G03
06B07
6,00E+10
CR627566
CR627575
CR627724
CR627745
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
164
206
307
333
555
522
770
187
282
705Q
600C
397Q
500C
630
334
608Q
500C
500
400Q
145C
164
700
307
700
86.3
90.3
93.3
94.5
1–6
3–8
8–7
12 – 12
07°08
07°09
07B10
07C03
CR627771
CR926157
CR627781
CR627785
Casasoli et al. 2006
Casasoli et al. 2006
Derory et al. 2006
Casasoli et al. 2006
341
252
360
285
700
300
360
285
86.1∗
95.7
2 – ni
5–4
8 – na
ni – 4
Transcription
Hypothetical protein
Hypothetical protein
Protein with binding function
or cofactor requirement
Hypothetical protein
Cellular transport
Transcription
Hypothetical protein
a
96.6∗
STSs 08A01, 07B10, 08B04, Cons 86, and 08G04 have not been sequenced.
The expected sizes were based on the knowledge of the EST sequence and primer design. The observed sizes were approximate because based on an electrophoresis on agarose gel. The
unmapped amplified ESTs were either noninformative or mapping methods (SSCP or DGGE) have not been successfully optimized.
c Except for 01E07 and Cons 129, all STS sequences matched the same gene in both species using a BLASTX procedure. If STS was mapped or sequence was available for only one species,
alignment has been done with the original oak EST (∗ ). In 5 cases, sequence reaction did not work (-).
d We used a LOD threshold ³6.0 to map STS, except for those marked with e, for which 4.0 < LOD score < 6.0. ni: noninformative; nm: nonmapped. Q-C: Quercus robur (3P × A4)-Castanea
sativa (Bursa × Hopa)
b
A. Kremer et al.
EST
Namea
Table 3. (continued)
Reference for primer
sequences and
PCR protocols
Expected
size (bp)b
08°01
08°03
CR627918
CR627920
Derory et al. 2006
Casasoli et al. 2006
210
454
500
454
08B04
08C05
08C11
08D11
08G04
Cons 13
Cons 14
Cons 19
Cons 21
Cons 30
CR627933
CR627943
CR627947
CR627958
CR627986
CR627506
CR627508
CR627517
CR627523
CR627541
Derory et al. 2006
Casasoli et al. 2006
Derory et al. 2006
Casasoli et al. 2006
Derory et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
327
213
316
343
393
301
243
178
333
424
Cons 33
CR627568
Casasoli et al. 2006
153
Cons 38
Cons 41
Cons 46
Cons 48
CR627606
CR627646
CR627952
CR627721
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
123
443
215
191
Cons 58
Cons 61
Cons 68
Cons 72
CR627732
CR627776
CR627777
CR627907
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
255
260
244
312
Cons 74
Cons 75
Cons 86
Cons 90
CR627801
CR627924
CR627976
CR628018
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
137
257
270
188
Cons 104
CR627823
Casasoli et al. 2006
250
327
213
316
700
1000
301
1200
300
550
1400Q
1500C
200Q
250C
123
500
800C
191Q
400C
500
1600Q
500Q
1000Q
800C
137
600
600
300Q
1200C
250
Observed
size
Identityc
Linkage
group
Q–Cd
Functional
category
3 d – nm
12d – ni
89.5–93.1
–
81.3
89.3
93.1
9 – nm
2 – ni
2–1
11 – 3
11 – na
1–6
5–4
9 – 2/4
2–1
4–2
Metabolism
Protein with binding function
or cofactor requirement
Metabolism
Hypothetical protein
Hypothetical protein
Metabolism
Hypothetical protein
Transcription
Protein synthesis
Protein synthesis
Protein synthesis
Hypothetical protein
95.3
12 d – 12
Hypothetical protein
91.7
90.7∗
–
–
2–1d
ni – 1
na – 9
6 – ni
Energy
Cell rescue, defense, and virulence
Cell rescue, defense, and virulence
Unknown
92.7
95.7∗
92.9∗
90.9∗
5–4
6 – na
1 – na
10 – ni
Hypothetical protein
Cell rescue, defense, and virulence
Metabolism
Cell cycle and DNA processing
86.7
88∗
9–9
ni – 8
8 – nm
2–7
Cell rescue, defense, and virulence
Metabolism
Unknown
Cell rescue, defense, and virulence
3–8
Hypothetical protein
94.1∗
95.5∗
94.4
88.4∗
–
95.4
175
Accession
number
Chapter 5 Fagaceae Trees
EST
Namea
176
Table 3. (continued)
Accession
number
Reference for primer
sequences and
PCR protocols
Expected
size (bp)b
Observed
size
Identityc
Linkage
group
Q–Cd
Functional
category
Cons 105
Cons 106
Cons 107
Cons 109
Cons 110
Cons 111
CR627826
CR627828
CR627830
CR627834
CR627835
CR627837
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
185
326
272
194
219
219
95.4
91.2∗
92.6
100∗
92.6
89.5∗
12 – 12
ni – 1
11 – 3
7–5
9–9
12 – ni
Metabolism
Energy
Cell-type differentiation
Cell rescue, defense, and virulence
Metabolism
Hypothetical protein
Cons 112
Cons 126
Cons 127
CR627839
CR628009
CR628014
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
171
238
289
600
326
900
1200
219
219Q
600C
171
400
289
93.4∗
94.7
94.3
5–4
7–5
6 – 11
Cons 128
Cons 129
Cons 130
Cons 135
CR628019
CR628021
CR628241
CR628167
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
Casasoli et al. 2006
120
210
190
115
120
500
190
200
–
79.6
91.9∗
100∗
6 – ni
9d –9
2 – ni
ni – 1
Transcription
Protein synthesis
Protein with binding function
or cofactor requirement
Energy
Cell rescue, defense, and virulence
Energy
Hypothetical protein
A. Kremer et al.
EST
Namea
Chapter 5 Fagaceae Trees
177
Table 4. Homologous linkage groups (LGs) in genetic maps of Quercus robur, Castanea sativa, and C. mollissima/C. dentata
LG in Quercus robur a
Pedigree 3P × A4
LG in Quercus robur b
Pedigree A1 × A2
LG in Castanea sativa c
Pedigree Bursa × Hopa
LG in C. mollissima/C. dentata d
Pedigree Mahogany
1
2
3∗
4
5
6
7
8
9
10
11
12
1
2
11∗
4
5
6
7
10
8
3
9
12
6
1
8
2
4
11∗∗
5
7
9
10
3
12
H
A
C
K
E
B∗∗
I
F
L
D
G
J
Assignment of linkage groups was made by comparison within the following pairs: 3P × A4 and A1 × A2, 3P × A4 and Bursa ×
Hopa, Bursa × Hopa and Mahogany.
∗ ) LG 3 in (3P × A4) and 11 in (A1 × A2) assigned by “default” (all other 11 LGs being assigned by at least 2 markers present in
each species)
∗∗ ) LG 11 in (Bursa × Hopa) and B in (Mahogany) assigned by “default” (all other 11 linkage group being assigned by at least
two markers present in each species, except for pairs 7-F and 3-G where only one marker was common).)
The numbers or letters of linkage groups (LG) correspond to the following publications:
a Barreneche et al. (1998); Barreneche et al. (2004)
b Bakker (2001) and this study
c Casasoli et al. (2001)
d Kubisiak et al. (1997); Sisco et al. (2005)
grown in hydroponic conditions. Gene expression was
monitored in seedlings raised under reduced oxygen
(3%) applied for 24 h. RNA was extracted from root
tips before (0 h time stress) and after oxygen reduction, following the protocol of Chang et al. (1993).
Stress induction was validated by measuring alcohol
dehydrogenase activity. Differentially expressed fragments were obtained by cDNA-AFLP, and 170 were
sequenced and compared to databanks (C. Bodénès
unpublished results).
5.4.3
Osmotic Stress
Osmotic stress induced genes were identified in
a Q. petraea cell line grown under moderate stress
(Porth et al. 2005a). Two subtraction libraries (SSH
method) were established from callus cell cultures exposed to hyperosmotic stress for 1 h (indicated as 1T)
and 2 d (2T), respectively. The differentially expressed
ESTs were classified according to their putative functions. At least five of these gene products were
assumed to be targets for stress tolerance in oak, e.g.,
betaine aldehyde dehydrogenase, two trans-acting
transcription factors (one ABA-responsive, the other
ABA-independent), a glutathione-S-transferase, and
a heat shock cognate protein.
Seven genes were selected, based on their putative functions, to monitor their expression in vivo.
Leaf tissue from hyperosmotically grown Q. petraea
and Q. robur plantlets was harvested and investigated
by RT-PCR at time intervals of 1, 6, 24, and 72 h. Indications of stress adaptation were found in Q. petraea based on up-regulation of certain genes related
to protective functions, whereas in Q. robur downregulation of those genes was evident (Porth et al.
2005a).
Segregating osmo-regulated loci were mapped to
ten different LGs of Quercus (Porth et al. 2005b).
By using orthologous primers, ten of the loci, including the four putatively water-stress tolerance related genes (1T57, 1T62, 2T11, and 2T13), were successfully amplified in C. sativa. Sequence analysis
showed an identity of at least 90% (Table 3) with
Quercus.
178
A. Kremer et al.
5.4.4
Differential Expression in Juvenile
and Mature Oak Shoots
A gene named QRCPE (Quercus robur crown preferentially expressed) that is differentially expressed
between mature and juvenilelike shoots was recently
discovered in oaks (Gil et al. 2003). QRPCE accumulates in ontogenetically older organs of oak trees, although it is present in zygotic and somatic embryos
but absent in callus cells. The encoded protein is small,
contains a predicted N-terminal hydrophobic signal
peptide that targets the protein to the cell wall, and is
rich in glycine and histidine residues. In C. sativa, the
QRCPE homolog is also expressed at different levels
between adult and juvenilelike tissues.
5.4.5
Blight Infection in Chestnut
A cDNA clone showed similarity to a gene previously
identified as encoding a cystatin. A protein shown to
have antifungal activity in C. sativa (Pernas et al. 1998,
1999) was isolated from a cDNA library from stem tissues of C. dentata (Connors et al. 2001). The expression of this gene was verified by RT-PCR in healthy
and diseased tissues of American chestnut (Connors
et al. 2002). Amplification of a fragment of the gene
in American and Chinese chestnuts and comparison
of the sequences of the cloned amplification products
revealed differences within the intron (SNPs or deletion). These differences could be used to locate the
cystatin gene on the map of C. mollissima/C. dentata
and to verify its putative colocalization with QTLs
involved in blight resistance (Connors et al. 2002).
However, cystatin did not map to any region known
to be involved in resistance to chestnut blight.
5.5
QTL Detection
5.5.1
Phenotypic Traits Investigated
A common objective in genetic mapping in oak, chestnut, and beech is the detection of QTLs for adaptive
traits, e.g., phenotypic traits that respond strongly to
natural selection, and particularly to abiotic or biotic
stresses. The interest in these traits lies in the issues
raised by global change and the capacity of trees to respond to these challenges (Parmesan and Yohe 2003).
This capacity depends on the level of genetic diversity
for these traits and their underlying genes in natural
populations. Knowledge of the genetic architecture
of these traits (number and distribution of QTLs) is
therefore of primary importance and has motivated
research in QTL in conifers as well (Sewell and Neale
2000; van Buijtenen 2001).
In European oak, chestnut, and beech, the genetic
control of three different adaptive traits, bud phenology, growth, and carbon isotope discrimination, were
studied using a QTL approach (Casasoli et al. 2004;
Scalfi et al. 2004; Scotti-Saintagne et al. 2004; Brendel et al. 2007). Bud phenology, growth, and carbon
isotope discrimination (delta or ∆, which provides
an indirect measure of plant water-use efficiency)
are adaptive traits that show great phenotypic variation in natural populations of forest trees (Zhang
and Marshall 1995; Tognetti et al. 1997; Lauteri et al.
1999; Hurme et al. 2000; Jermstad et al. 2001). Initiation and cessation of the growing seasons, defined
through bud flush and bud set timing, have profound
implications for adaptation of perennial plants to cold
winter temperatures. Early flushing genotypes might
be susceptible to spring frost damage. Likewise, bud
set timing is related to the fall cold acclimation (Howe
et al. 2000). Growth traits, such as annual height and
diameter increments, are important components of
plant vigor and biomass production, and they are
profoundly influenced by abiotic and biotic stress occurrences during the growing season. In addition,
they are relevant characteristics from an economic
point of view and are often evaluated in breeding programs (Bradshaw and Stettler 1995). Carbon isotope
discrimination (∆) is a parameter related to the isotopic fractionation of carbon stable isotopes during
the photosynthetic process (for review see Farquhar
et al. 1989; Brugnoli and Farquhar 2000). Plant material is always enriched in 13 C with respect to the isotopic composition (δ13 C) of atmospheric CO2 . This is
particularly evident in C3 plants where the fractionation effect mostly occurs during CO2 diffusion from
outside the leaf to the carboxylation sites into the
chloroplasts, and during the carboxylation by ribulose 1,5-bisphosphate (RuBP) carboxylase. Due to its
relationships with the diffusional path of photosynthetic gas exchange (for both CO2 and water vapor
in reverse directions) and with the photosynthetic
substrate demand (CO2 fixation by RuBP carboxylation activity), ∆ has been theoretically predicted and
Chapter 5 Fagaceae Trees
empirically demonstrated to be inversely related to
plant water-use efficiency (roughly the ratio of carbon
gain to water losses; for deeper insights see Farquhar
et al. 1989; Brugnoli and Farquhar 2000). Despite the
complexity of this trait, significant heritabilities and
low genotype × environment interactions have been
found for ∆ in crop species (Hall et al. 1994) encouraging the use of this parameter for breeding purposes.
5.5.2
Strategies and Methods Used for QTL Detection
In forest trees, QTLs for several traits have been
detected, clearly showing the usefulness of this approach to dissect genomic regions controlling complex traits (Sewell and Neale 2000). With few exceptions (Brown et al. 2003; Jermstad et al. 2003), the
size of segregating populations used in these studies is often small (150 to 200 individuals). Among
factors influencing QTL detection power, small sample sizes and low trait heritability were shown to
cause an overestimation of QTL effects and the underestimation of QTL number and to hamper detection of QTLs with low effects (Beavis 1995). For
these reasons, a single QTL detection experiment does
not give an exhaustive idea of the genetic architecture of a quantitative trait. One possible strategy to
overcome these difficulties is to detect QTLs several
times across different environments and developmental stages. In this way, environmental and temporal
stability of QTLs can be verified and a more complete
picture of genetic architecture of the complex trait
can be drawn. Moreover, comparative QTL mapping
between phylogenetically related species offers an important tool to validate QTLs from the evolutionary
point of view. In oak and chestnut, a QTL-detection
strategy based on multiple experiments across different environments and years has been performed
to give an idea, as much as possible, of the complete genetic architecture of adaptive traits in both
species. Afterwards, comparative QTL mapping for
the three adaptive traits studied was carried out between the two species in order to identify genomic regions conserved through evolution controlling these
traits.
In oak, QTL detection was done in both the French
(3P*A4) and Dutch (A1*A2) Q. robur mapping pedigrees. In the French studies, phenotypic assessments
were done over successive years using a clonal test
179
planted with the vegetative copies of the full-sibs belonging to the pedigree. The phenotypic data obtained so far all originate from the two plantations
installed in the southwest of France. The assessments
first addressed the same three major adaptive traits as
for chestnut: phenology, growth, and carbon isotope
discrimination (Scotti-Saintagne et al. 2004; Brendel
et al. 2007). In addition, leaf morphology characters
(Saintagne et al. 2004) and the ability for vegetative reproduction by cutting propagation (Scotti-Saintagne
et al. 2005) were assessed. The focus on leaf morphological traits is related to their use in species discrimination as shown by previous analyses (Kremer et al.
2002). The Dutch study focused on QTL detection for
morphological and growth characters in one specific
full-sib cross that was grown for two successive years
in a nursery (Bakker 2001).
In European chestnut (C. sativa), bud flush,
growth, and carbon isotope discrimination measurements were performed for 3 years: 2000, 2001,
and 2002, corresponding to the growing seasons
2, 3, and 4 since seed germination. Bud set timing
was scored only in 2002. During the three years,
plants were grown in central Italy (Istituto di
Biologia Agroambientale e Forestale, CNR, Porano,
central Italy, 42◦ 43′ latitude, 500 m elevation)
as previously reported. Details about phenotypic
measurements are reported in Casasoli et al. (2004)
and in Table 5.
In American (C. dentata) and Asian (C. mollissima) chestnut, the blight resistance response of F2
progeny was assessed by using the agar-disk corkborer method (Griffin et al. 1983). During the growing season, each F2 individual was inoculated with two
different strains of Cryphonectria parasitica. Canker
evaluations were made over two successive months.
The mean canker sizes in each month for each isolate were used as relative measures of resistance.
The degree of association between marker loci and
blight resistance trait was investigated using successively single-locus or nonsimultaneous analysis of
variance (ANOVA) models and multiple marker or
simultaneous analysis of variance (ANOVA) models (Kubisiak et al. 1997). In European beech (F. sylvatica), leaf traits (size and shape) were assessed
over 2 years, whereas growth and carbon isotope
discrimination measurements were done only one
year.
The MultiQTL software (Britvin et al. 2001,
http://esti.haifa.ac.il/∼poptheor) was used for QTL
detection both in oak and chestnut in the French
180
Table 5. QTL data in oak, chestnut, and beech
Species
(reference)
Pedigree
Number of
offsprings
phenotyped
Number of field
plantations
where the trait
was assessed
Growing
season(s) when
the trait
was assessed
Heritability
or repeatability
Number
of QTLs
detected (f )
Range of
variation of
PEV
(minimum –
maximum)
Height growth
Quercus robur (a)
Quercus robur (b)
Castanea sativa (c)
Fagus sylvatica (d)
Quercus robur (a)
Quercus robur(b)
3P × A4
A1 × A2
Bursa × Hopa
44 × 45
3P × A4
A1 × A2
207
101
135–153
118
174–278
101
1
1
1
1
3
1
4a
2
3
1
8, 4, 5
1
0.14–0.23 c
–
–
–
0.15–0.52 c
–
5
1
6
0
12
2
Castanea sativa (c)
Fagus sylvatica (d)
Quercus robur (e)
Castanea sativa (c)
Fagus sylvatica (d)
Castanea sativa (c)
Quercus robur (b)
Bursa × Hopa
44 × 45
3P × A4
Bursa × Hopa
44 × 45
Bursa × Hopa
A1 × A2
150–174
124
121–207
152–155
102
151
101
1
1
2
1
1
1
1
3
1
4, 5, 5
3
1
1
2
–
–
0.32–0.80 c
–
–
–
–
9
1
5
7
0
3
–
Castanea sativa (c)
Bursa × Hopa
136–153
1
3
–
4
9.5–18.7
31.2
7.0–17.0
–
3.1–10.7
Trait not normally
distributed.
Nominal scale
of 1–5
6.3–12.2
27.3
4.4–34.4
5.7–13.2
–
8.9–17.1
No significant
QTL detected
5.9–10.3
Bud burst
Delta
Bud set
Diameter
growth
(a)
Scotti-Saintagne et al. (2004)
Bakker (2001)
(c) Casasoli et al. (2004)
(d) Scalfi et al. (2004)
(e) Brendel et al. (2007)
(f) QTL detected at p < 0.05 at the genome level
(b)
A. Kremer et al.
Trait
Chapter 5 Fagaceae Trees
181
Fig. 3. Comparative QTL mapping between Q. robur and C.sativa (from Casasoli et al. 2006). Homeologous LGs between Q. robur
(Q) and C. sativa (C) are named and ordered as in Figs. 1 and 2. Orthologous markers are linked by dotted lines. Common
intervals between the two genomes identified by orthologous markers are filled with corresponding backgrounds in both oak and
chestnut LGs. The figure was drawn using MapChart software (Voorrips 2002). Each QTL is represented on the right of the LG
by its confidence interval (95% confidence intervals, black line) and the most probable position (Casasoli et al. 2006). QTLs were
detected for three different phenotypic traits on the male (m) and female map (f): bud burst (Bud), total height (H), and carbon
isotope discrimination (∆). The phenotypic traits were observed over three seasons (indicated by subscripts 1 to 3). In oak, the
date of bud burst was assessed as the date when the apical bud flushed. In chestnut, bud burst was assessed in two different ways:
A = date of first observed unfolded leaf of a tree; B = date when 70% of buds showed an unfolded leaf BudA2f: QTL for bud burst
assessed with method A during season 2 in female map. H2m: QTL for total height measured at season 2 and located on male
map. ∆3f: QTL for carbon isotope discrimination assessed during season 3 and located on female map
182
A. Kremer et al.
Fig. 3. (continued)
studies. This software was chosen for several
reasons. First, the composite interval mapping
was available (CIM, Jansen and Stam 1994; Zeng
1994); second, QTL significance thresholds could
be computed by permutation (Churchill and Doerge 1994); and finally, confidence intervals for
QTL position could be estimated by bootstrap
(Visscher et al. 1996). The same statistical analysis
was performed in oak and chestnut; the details are
reported in Casasoli et al. (2004) and Scotti-Saintagne
et al. (2004). The Dutch study used MapQTL 4.0
(http://www.kyazma.nl/index.php/mc.MapQTL) for
QTL detection. For beech, QTL Cartographer 1.12
(Basten et al. 1994, 2002) and MultiQTL (Britvin et al.
2001) softwares were used to detect QTLs (Scalfi et al.
2004)
Chapter 5 Fagaceae Trees
5.5.3
Number and Distribution
of QTLs and Their Effects
The results of QTL detection are extremely heterogeneous across pedigrees and species. The survey of the
results has been limited to traits that were assessed in
at least two species (Table 5). Heterogeneity among
the results is most likely related to the reduced size
of the mapping pedigree. As already mentioned, with
on average less than 200 sibs per pedigree, the contribution of a given QTL to the phenotypic variance of
the trait is usually overestimated and exhibits a large
sample variance (Beavis 1995). However, because the
sampling efforts were similar and the phenotypic assessments were the same, a comparative analysis of
QTL detection could be done between Q. robur and
C. sativa.
The alignment of the 12 Q. robur and C. sativa LGs
gives rise to a logical framework defined by common
orthologous markers for comparing QTL location between the two species. Figure 3 shows the alignment
of the 12 LGs, the common genomic regions identified
by orthologous markers, and QTLs compared in oak
and chestnut. Details about the definition of common
genomic intervals and corresponding unique QTLs
between the two species (i.e., more individual QTLs
detected several times in the same genomic region in
a single species) are reported in Casasoli et al. (2006).
A total number of 34 common intervals were identified between the oak and chestnut genetic linkage
maps thanks to the orthologous markers. Following
the previously described criteria to declare unique
QTLs, 13 and 10 unique QTLs were identified for timing of bud burst, 5 and 7 unique QTLs were identified
for carbon isotope discrimination, and, finally, 5 and
6 unique QTLs for height growth were identified in
oak and chestnut, respectively (Fig. 3). Among these
unique QTLs, nine controlling timing of bud burst and
two controlling height growth were colocated between
the two species. No QTL involved in carbon isotope
discrimination was colocated in the oak and chestnut
map. Following Lin et al. (1995), the probability of
obtaining these colocations by chance is p = 0.0002 in
the case of timing of bud burst and p = 0.20 in the case
of height growth. When QTL number and effects were
compared for the three traits between the two species,
a similar genetic architecture was observed for adaptive traits in oak and chestnut (Casasoli et al. 2006).
From this simple comparison it was clear that adaptive
traits are controlled by more loci of low and moderate
183
than large effect in both species. Timing of bud flush
was the trait showing the higher number of detected
and stable QTLs. Despite this similar genetic architecture, most of the QTLs for bud flush were conserved,
whereas only a few QTLs were conserved for height
growth, and none for carbon isotope discrimination.
The different conservation of QTLs may be explained
taking into account differences for the three adaptive
traits investigated in trait heritability values, QTL stability across experiments, and QTL-by-environment
interactions. The striking conservation of QTLs for
bud flush is very interesting from an evolutionary
point of view. Although correspondence of QTLs does
not imply correspondence of genes underlying the
QTLs, as already reported in other species (Doust
et al. 2004), these findings showed that loci controlling
bud flush have remained highly polymorphic in both
species. This high polymorphism of loci controlling
bud flush, despite strong natural selection acting on
this adaptive trait, may be explained with selection
pressures able to maintain diversity over long evolutionary times (balancing, disruptive, or frequencydependent selection) as discussed in Casasoli et al.
(2006).
5.6
Conclusion
Mapping experiments in Fagaceae were hampered by
various biological constraints that have limited research activities in this field. First, for most species,
it was not possible to find adequate F2 pedigrees
that would allow us to screen the genome for QTLs
of interest. This is somehow compensated by the
high level of within-population diversity, which would
allow segregation for QTLs of interest in F1 pedigrees as well. Second, controlled crosses to obtain
mapping F1 pedigrees has been challenging in these
species, and alternatives based on open-pollinated
progeny screening using parentage analysis were implemented. Third, the development of mapping activities was restrained by the limited genomic resources available (genetic markers, ESTs) within this
group of species. Despite these limitations, important progress has been made in the recent years as
a result of international cooperation. Maps have been
developed for each economically important genus
(Quercus, Castanea, and Fagus), and the ongoing activities in comparative mapping suggest that there
184
A. Kremer et al.
is a strong macrosynteny between phylogenetically
close genera (Quercus and Castanea). For some traits,
e.g., bud burst, there is even a strong conservation
of the QTL position between the two genera. Extension of comparative mapping to Fagus might be
more problematic as illustrated by difficulties described in this review. However, comparative mapping should be much easier with Lithocarpus and
Castanopsis, as these genera are close to Quercus
and Castanea. Furthermore, the genome of the Fagaceae is of small enough size (e.g., only 3.5 to 6
times larger than Arabidopsis) to make comparative
genomics easily applicable to this family. These expectations should enhance research activities in genetics
within a large group of ecologically and economically
important species growing throughout the northern
hemisphere.
Acknowledgement. The construction of the genetic linkage
maps in European species Quercus robur, Castanea sativa and
Fagus sylvatica was carried out with the financial support of the
European Commission, DG Research (OAKFLOW, QLK5-200000960 for oaks; CASCADE, EVK2-1999-00065P for chestnut,
and DYNABEECH, QLRT-1999-01210 for beech). The study on
the Dutch Q. robur map was carried out with financial support
from Programm 381 Functions of Nature, Forest and Landscape
of the Dutch Ministry of Agriculture, Nature and Food quality.
Jeremy Derory received a PhD grant from INRA to develop the
EST used for the comparative mapping between Quercus and
Castanea. The authors are grateful to Scott Schlarbaum (University of Tennessee-Knoxville) for providing the material for
the genome size determination of different Fagaceae species,
to Preston Aldrich, Kevin McAbee, David Chagne, Paolo Piovani, Weilin Sun, Michela Troggio, for their helpful contribution. Jeanne Romero thanks Antoine Kremer for suggesting
to screen open-pollinated progenies by exclusion methods in
order to identify full-sib progeny for the mapping.
References
Aldrich PR, Michler C, Sun W, Romero-Severson J (2002) Microsatellite markers for northern red oak (Fagaceae: Quercus rubra). Mol Ecol Notes 2:472–475
Aldrich PR, Jagtap M, Michler CH, Romero-Severson J (2003a)
Amplification of north American red oak microsatellite
markers in European white oaks and Chinese chestnut.
Silvae Genet 52(3-4):176-179
Aldrich PR, Parker GR, Michler CH, Romero-Severson J (2003b)
Whole-tree silvic identifications and the microsatellite genetic structure of a red oak species complex in an Indiana
old-growth forest. Can J For Res 33:2228–2237
Arcade A, Anselin F, Faivre Rampant P, Lesage MC, Pâques LE,
Prat D (2000) Application of AFLP, RAPD and ISSR markers
to genetic mapping of European and Japanese larch. Theor
Appl Genet 100:299–307
Axelrod DI (1983) Biogeography of oaks in the Arco-Tertiary
province. Ann Mo Bot Gard 70:629–657
Bakker EG (2001) Towards molecular tools for management of
oak forests, PhD Thesis. Alterra, Wageningen, The Netherlands
Barreneche T, Bodenes C, Lexer C, Trontin JF, Fluch S, Streiff
R, Plomion C, Roussel G, Steinkellner H, Burg K, Favre JM,
Glössl J, Kremer A (1998) A genetic linkage map of Quercus robur L. (pedunculate oak) based on RAPD, SCAR, microsatellite, minisatellite, isozyme and 5S rDNA markers.
Theor Appl Genet 97:1090–1103
Barreneche T, Casasoli M, Russell K, Akkak A, Meddour H,
Plomion C, Villani F, Kremer A (2004) Comparative mapping between Quercus and Castanea using simple-sequence
repeats (SSRs). Theor Appl Genet 108:558–566
Basten CJ, Weir BS, Zeng ZB (1994) Zmap-a QTL cartographer. In: Smith C, Gavora JS, Benkel B, Chesnais J, Fairfull
W, Gibson JP, Kennedy BW Burnside EB (eds) Proc 5th
World Congress On Genetics Applied to Livestock Production: Computing Strategies and Software, Guelp, Ontario,
Canada, 22:65–66
Basten CJ, Weir BS, Zeng ZB (2002) QTL cartographer, version 1.16. Department of Statistics, North Carolina State
University, Raleigh, NC
Beavis WD (1995) The power and deceit of QTL experiments:
lessons from comparative QTL studies. In: Proc 49th Annu
Corn and Sorghum Industry Res Conf, ASTA, Washington,
DC, pp 250–266
Bennett MD, Leitch IJ, Price HJ, Johnston JS (2003) Comparisons with Caenorhabditis (∼100 Mb) and Drosophila (∼
175 Mb) using flow cytometry show genome size in Arabidopsis to be ∼157 Mb and thus ∼25% larger than the
Arabidopsis Genome Initiative estimate of ∼125 MB. Ann
Bot 91:547–557
Bradshaw HD, Stettler RF (1995) Molecular genetics of growth
and development in Populus. IV. Mapping QTLs with large
effects on growth, form, and phenology traits in a forest
tree. Genetics 139:963–973
Brendel O, Le Thiec D, Scotti-Saintagne C, Bodénès C, Kremer A,
Jean-Marc Guehl JM, 2007Quantitative trait loci controlling
water use efficiency and related traits in Quercus robur L.
Tree Genetics and Genomes (in press)
Britvin E, Minkov D, Glikson L, Ronin Y, Korol A (2001) MultiQTL, an interactive package for genetic mapping of correlated quantitative trait complexes in multiple environments, version 2.0 (Demo). In: Plant & Animal Genome IX
Conf, San Diego
Brown GR, Kadel III EE, Bassoni DL, Kiehne KL, Temesgen B,
van Buijtenen JP, Sewell MM, Marshall KA, Neale DB (2001)
Anchored reference loci in loblolly pine (Pinus taeda L.)
for integrating pine genomics. Genetics 159:799–809
Brown GR, Bassoni DL, Gill GP, Fontana JR, Wheeler NC,
Megraw RA, Davis MF, Sewell MM, Tuskan GA, Neale DB
Chapter 5 Fagaceae Trees
(2003) Identification of quantitative trait loci influencing
wood property traits in loblolly pine (Pinus taeda L.). III.
QTL Verification and candidate gene mapping. Genetics
164:1537–1546
Brugnoli E, Farquhar GD (2000) Photosynthetic fractionation of carbon isotopes. In: Leegood RC, Sharkey TD,
von Caemmerer S (eds) Photosynthesis: Physiology and
Metabolism. Advances in Photosynthesis, vol 9. Kluwer,
Boston, pp 399–434
Brunner AM, Busov VB, Strauss SH (2004) Poplar genome sequence: functional genomics in an ecologically dominant
plant species. Trends Plant Sci 9:49–56
Burnham CR, Rutter PA, French DW (1986) Breeding blightresistant chestnuts. Plant Breed Rev 4:347–397
Butorina AK (1993) Cytogenetic study of diploid and spontaneous triploid oaks, Quercus robur L. Ann Sci For 50(Suppl
1):144s-150s
Camus A (1936-1954) Les chênes, Monographie du genre Quercus et Monographie du genre Lithocarpus. Encyclopédie
Economique de Sylviculture. Vol. VI, VII, VIII. Académie
des sciences, Paris, France
Cannon CH, Manos PS (2003) Phylogeography of the South
Asian stone oaks (Lithocarpus). J Biogeogr 30:211–226
Casasoli M, Mattioni C, Cherubini M, Villani F (2001) A genetic
linkage map of European chestnut (Castanea sativa Mill.)
based on RAPD, ISSR and isozyme markers. Theor Appl
Genet 102:1190–1199
Casasoli M, Pot D, Plomion C, Monteverdi MC, Barreneche T,
Lauteri M, Villani F (2004) Identification of QTLs affecting
adaptive traits in Castanea sativa Mill. Plant Cell Environ
27:1088–1101
Casasoli M, Derory J, Morera-Dutrey C, Brendel O, Porth I,
Guehl J-M, Villani F, Kremer A (2006) Comparison of QTLs
for adaptive traits between oak and chestnut based on an
EST consensus map. Genetics 172:533- 546
Ceroni M, Leonardi S, Piovani P, Menozzi P (1997) Incrocio
diallelico in faggio: obiettivi, metodologie, primi risultati.
Monti e Boschi 48(1):46-51
Cervera MT, Storme V, Ivens B, Gusmao J, Liu BH, Hostyn V, Van
Slycken J, Van Montagu M, Boerjan W (2001) Dense genetic
linkage maps of three Populus species (Populus deltoides, P.
nigra and P. trichocarpa) based on AFLP and microsatellite
markers. Genetics 158:787–809
Chang S, Puryear J, Cairney J (1993) A simple and efficient
method for isolating RNA from pine trees. Plant Mol Biol
Rep 11(2):113-116
Churchill GA, Doerge RW (1994) Empirical threshold values for
quantitative trait mapping. Genetics 138:963–971
Clark C, Kubisiak T, Lee B-C, O’Malley D, Sisco P (2001) AFLPs
– towards a saturated genetic map for Castanea. In: Plant
& Animal Genome IX Conf, San Diego
Connors BJ, Maynard CA, Powell WA (2001) Expressed sequence tags from stem tissue of the American chestnut,
Castanea dentata. Biotechnol Lett 23 (17):1407-1411
185
Connors BJ, Laun NP, Maynard CA, Powell WA (2002) Molecular
characterization of a gene encoding a cystatin expressed in
the stems of American chestnut (Castanea dentata). Planta
215:510–513
Costa P, Pot D, Dubos C, Frigerio J-M, Pionneau C, Bodénès
C, Bertocchi E, Cervera M-T, Remington D-L, Plomion
C (2000) A genetic linkage map of Maritime pine based
on AFLP, RAPD and protein markers. Theor Appl Genet
100:39–48
Crepet WL (1989) History and implications of the early North
American fossil record of Fagaceae. In: Crane PR, Blackmore S (eds) Evolution, Systematics, and Fossil History of
the Hamamelidae, vol 2: ‘Higher’ Hamamelidae’. Clarendon, Oxford, pp 45–66
D’Emerico S, Bianco P, Medagli P, Schirone B (1995) Karyotype
analysis in Quercus ssp. (Fagaceae). Silvae Genet 44:66–70
Derory J, Léger P, Garcia V, Schaeffer J, Hauser M-T, Plomion
C, Glössl J, Kremer A (2006) Transcriptome analysis of bud
burst in sessile oak (Quercus petraea (Matt.) Liebl.). New
Phytol 170:723–738
Doust AN, Devos KM, Gadberry MD, Gale MD, Kellogg EA
(2004) Genetic control of branching in foxtail millet. Proc
Natl Acad Sci USA 15:9045–9050
Farquhar GD, Ehleringer JR, Hubick KT (1989) Carbon isotope discrimination and photosynthesis. Annu Rev Plant
Physiol Mol Biol 40:503–537
Favre JM, Brown S (1996) A flow cytometric evaluation of the
nuclear content and GC percent in genomes of European
oak species. Ann Sci For 53:915–917
Gale MD, Devos KM (1998) Comparative genetics in the grasses.
Proc Natl Acad Sci USA 95:1971–1974
Gallois A, Burrus M, Brown S (1999) Evaluation of DNA content
and GC percent in four varieties of Fagus sylvatica L. Ann
For Sci 56:615–618
Gil B, Pastoriza E, Ballester A, Sanchez C (2003) Isolation and
characterization of a cDNA from Quercus robur differentially expressed in juvenile-like and mature shoots. Tree
Physiol 23:633–640
Govaerts R, Frodin DG (1998) World checklist and bibliography of Fagales (Betulaceae, Fagaceae and Ticodendraceae).
Royal Botanical Garden, Kew, UK
Grattapaglia D, Sederoff R (1994) Genetic linkage maps of Eucalyptus grandis and Eucalyptus urophylla using a pseudo
test-cross mapping strategy and RAPD markers. Genetics
137:1121–1137
Grattapaglia D, Bertolucci F, Sederoff R (1995) Genetic mapping of QTLs controlling vegetative propagation in Eucalyptus grandis and E. urophylla using a pseudo testcross strategy and RAPD markers. Theor Appl Genet
90:933–947
Griffin GJ, Hebard FV, Wendt RW, Elkins JR (1983) Survival
of American chestnut trees: Evaluation of blight resistance and virulence of Endothia parasitica. Phytopathology
73:1084–1092
186
A. Kremer et al.
Hall AE, Richards RA, Condon AG, Wright GC, Farquhar GD
(1994) Carbon isotope discrimination and plant breeding.
Plant Breed Rev 4:81–113
Herendeen PS, Crane PR, Drinman AN (1995) Fagaceous flowers, fruits and cupules from the Campanian (late Cretaceous) of Central Georgia, USA. Int J Plant Sci 156:93–116
Howe GT, Saruul P, Davis J, Chen THH (2000) Quantitative
genetics of bud phenology, frost damage, and winter survival in an F2 family of hybrid poplars. Theor Appl Genet
101:632–642
Hurme P, Sillanpää MJ, Arjas E, Repo T, Savolainen O (2000) Genetic basis of climatic adaptation in Scot pine by Bayesian
quantitative trait locus analysis. Genetics 156:1309–1322
Jansen RC, Stam P (1994) High resolution of quantitative
trait into multiple loci via interval mapping. Genetics
136:1447–1445
Jermstad KD, Bassoni DL, Jech KS, Wheeler NC, Neale DB (2001)
Mapping of quantitative trait loci controlling adaptive traits
in coastal Douglas-fir. I. Timing of vegetative bud flush.
Theor Appl Genet 102:1142–1151
Jermstad KD, Bassoni DL, Jech KS, Ritchie GA, Wheeler NC,
Neale DB (2003) Mapping of quantitative trait loci controlling adaptive traits in coastal Douglas fir. III. Quantitative trait loci-by-environment interactions. Genetics
165:1489–1506
Jones JH (1986) Evolution of the Fagaceaeae: the implication of
foliar features. Ann Mo Bot Gard 73:213–229
Kremer A, Dupouey JL, Deans JD, Cottrell J, Csaikl U, Finkeldey
R, Espinel S, Jensen J, Kleinschmit J, Van Dam B, Ducousso
A, Forrest I, de Heredia UL, Lowe A, Tutkova M, Munro
RC, Steinhoff S, Badeau V (2002) Leaf morphological differentiation between Quercus robur and Quercus petraea is
stable across western European mixed oak stands. Ann For
Sci 59:777–787
Kremer A, Xu LA, Ducousso A (2004) Genetics of oaks. In:
Burley J, Evans J, Youngquist J (eds) Encyclopedia of Forest
Sciences. Elsevier, Amsterdam, pp 1501–1507
Kubisiak TL, Hebard FV, Nelson CD, Zhang J, Bernatzky R,
Huang H, Anagnostakis SL, Doudrick RL (1997) Molecular
mapping of resistance to blight in an interspecific cross in
the genus Castanea. Phytopathology 87:751–759
Kvaèek Z, Walther H (1989) Paleobotanical studies in Fagaceae
of the European Tertiary. Plant Syst Evol 162:213–229
Lauteri M, Scartazza A, Guido MC, Brugnoli E (1997) Genetic
variation in photo synthetic capacity, carbon isotope discrimination and mesophyll conductance in provenances of
Castanea sativa adapted to different environments. Funct
Ecol 11:675–683
Lauteri M, Monteverdi MC, Sansotta A, Cherubini M, Spaccino L, Villani F (1999) Adaptation to drought in European chestnut. Evidences from a hybrid zone and from
controlled crosses between drought and wet adapted populations. Acta Hort 494:345–353
Lin YR, Schertz KF, Paterson AH (1995) Comparative analysis of QTLs affecting plant height and maturity across the
Poaceae, in reference to an interspecific sorghum population. Genetics 141:391–411
Manos PS, Steele KP (1997) Phylogenetic analyses of “higher”
Hamamelidiae based on plastid sequence data. Am J Bot
84:1407–1419
Manos PS, Doyle JJ, Nixon KC (1999) Phylogeny, biogeography,
and processes of molecular differentiation in Quercus subgenus Quercus (Fagaceae). Mol Phylogenet Evol 12:333–349
Manos PS, Zhe-Kun Zhou, Cannon CH (2001) Systematics of
Fagaceae: phylogenetic tests of reproductive trait evolution.
Int J Plant Sci 162:1361–1379
Manos PS, Stanford AM (2001) The historical biogeography of
Fagaceae: tracking the Tertiary history of temperate and
subtropical forests of the northern hemisphere. Int J Plant
Sci 162(6 Suppl):s77-s93
Marques CM, Araujo JA, Ferreira JG, Whetten R, O’Malley
DM, Liu BH, Sederoff R (1998) AFLP genetic maps of Eucalyptus globulus and E. tereticornis. Theor Appl Genet
96:727–737
Marques CM, Brondani RPV, Grattapaglia D, Sederoff R (2002)
Conservation and synteny of SSR loci and QTLs for vegetative propagation in four Eucalyptus species. Theor Appl
Genet 105:474–478
Mehra PN, Hans AS, Sareen TS (1972) Cytomorphology of Himalayan Fagaceae. Silvae Genet 21:102–109
Naujoks G, Hertel H, Ewald D (1995) Characterisation and
propagation of an adult triploid pedunculate oak (Quercus robur). Silvae Genet 44:282–286
Ohri D, Ahuja MR (1990) Giemsa C-banding in Quercus L. (oak).
Silvae Genet 39:216–219
Ohri D, Ahuja MR (1991) Giemsa C-banding in Fagus sylvatica
L., Betula pendula Roth and Populus tremula L. Silvae Genet
40:72–75
Parmesan C, Yohe G (2003) A globally coherent fingerprint
of climate change impacts across natural systems. Nature
421:37–42
Pernas M, Sanchez-Monge R, Gomez L, Salcedo G (1998)
A chestnut seed cystatin differentially effective against cysteine proteinase from closely related pests. Plant Mol Biol
38:1235–1242
Pernas M, Lopez-Solanilla E, Sanchez-Monge R, Salcedo G,
Rodriguez-Palenzuela P (1999) Antifungal activity of
a plant cystatin. Mol Plant-Micr Interact 12:624–627
Porth I, Koch M, Berenyi M, Burg A, Burg K (2005a) Identification of adaptation specific differences in the mRNA expression of sessile and pedunculate oak based on osmotic
stress induced genes. Tree Physiol 25:1317–1329
Porth I, Scotti-Saintagne C, Barreneche T, Kremer A, Burg K
(2005b) Linkage mapping of osmotic induced genes of oak.
Tree Genet Genom 1:31–40
Powell W, Machray GC, Provan J (1996) Polymorphism revealed
by simple sequence repeats. Trends Plant Sci 1:215–222
Romero-Severson J, Aldrich P, Mapes E, Sun W (2003) Genetic
mapping in northern red oak (Quercus rubra L.). In: Plant
& Animal XI Genome Conf, San Diego
Chapter 5 Fagaceae Trees
Saintagne C, Bodénès C, Barreneche T, Pot D, Plomion C, Kremer A (2004) Distribution of genomic regions differentiating oak species assessed by QTL detection. Heredity
92:20–30
Scalfi M, Troggio M, Piovani P, Leonardi S, Magnaschi G, Vendramin GG, Menozzi P (2004) A RAPD, AFLP, and SSR
linkage map, and QTL analysis in European beech (Fagus
sylvatica L.) Theor Appl Genet 108:433–441
Scalfi M (2005) Studio della variabilita‘ genetica neutrale ed
adattativa in Fagus sylvatica L. Tesi di Dottorato in Ecologia
XVII ciclo. University of Parma, Italy
Schwarz O (1964) Quercus. In: Tutin TG, Heywood VH, Burges
NA, Valentine DH, Walters SM, Webb DA (eds) Flora Europea. Vol 1: Lycopodiaceae to Platanaceae, Cambridge University Press, Cambridge, UK, pp 61–64
Scotti-Saintagne C, Bodénès C, Barreneche T, Bertocchi E,
Plomion C, Kremer A (2004) Detection of quantitative trait
loci controlling bud burst and height growth in Quercus
robur L. Theor Appl Genet 109:1648–1659
Scotti-Saintagne C, Bertocchi E, Barreneche T, Kremer A,
Plomion C (2005) Quantitative trait loci mapping for
vegetative propagation in pedunculate oak. Ann For Sci
62:369–374
Sewell MM, Neale DB (2000) Mapping quantitative traits in forest trees. In: Jain SM, Minocha SC (eds) Molecular Biology
of Woody Plants. Kluwer, Dordrecht, pp 407–423
Sisco PH, Kubisiak TL, Casasoli M, Barreneche T, Kremer A,
Clark C, Sederoff RR, Hebard FV, Villani F (2005) An improved genetic map for Castanea mollissima / Castanea
dentata and its relationship to the genetic map of Castanea
sativa. Acta Hort 693:491–495
Steinkellner H, Lexer C, Turetschek E, Glössl J (1997) Conservation of (GA)n microsatellite loci between Quercus species.
Mol Ecol 6:1189–1194
Tognetti R, Michelozzi M, Giovannelli A (1997) Geographical
variation in water relations, hydraulic architecture and terpene composition of Aleppo pine seedlings from Italian
provinces. Tree Physiol 17:241–250
Van Buijtenen H (2001) Genomics and quantitative genetics.
Can J For Res 31:617–622
Verhaegen D, Plomion C (1996) Genetic mapping in Eucalyptus
urophylla and Eucalyptus grandis using RAPD markers.
Genome 39:1051–1061
187
Villani F, Pigliucci M, Benedettelli S, Cherubini M (1991) Genetic differentiation among Turkish chestnut (Castanea
sativa Mill.) populations. Heredity 66:131–136
Villani F, Pigliucci M, Lauteri M, Cherubini M, Sun O (1992)
Congruence between genetic, morphometric, and physiological data on differentiation of Turkish chestnut (Castanea sativa). Genome 35:251–256
Visscher PM, Thompson R, Haley CS (1996) Confidence
intervals in QTL mapping by bootstrapping. Genetics
143:1013–1020
Voorrips RE (2002) MapChart: software for the graphical presentation of linkage maps and QTLs. J Hered 93:77–78
Vos P, Hogers R, Bleeker M, Reijans M, van de Lee T, Hornes
M, Fritjers A, Pot J, Peleman J, Kuiper M, Zabeau M (1995)
AFLP: a new concept for DNA fingerprinting. Nucleic Acids
Res 23:4407–4414
Wen J (1999) Evolution of eastern Asian and eastern North
American disjunct distributions in flowering plants. Annu
Rev Ecol Syst 30:421–455
Williams JGK, Kubelik AR, Livak KJ, Rafalski JA, Tingey
SV (1990) DNA polymorphisms amplified by arbitrary
primers are useful as genetic markers. Nucleic Acids Res
18:6531–653
Xiang QY, Soltis DE, Soltis PS, Manchester SR, Crawford DJ
(2000) Timing the Eastern Asia-Eastern North America
floristic disjunction: molecular clock corroborates paleontological estimates. Mol Phylogenet Evol 15:462–472
Xu LA, Kremer A, Bodénès C, Huang MR (2005) Chloroplast
DNA diversity and phylogenetic relationship of botanical
sections of the genus Quercus. In: Proc Joint Meeting of
IUFRO Working Groups on “Genetics of Quercus & Improvement and Silviculture of Oaks,” 29 Sept-3 Oct 2003,
Tsukuba, Japan
Zeng ZB (1994) Precision mapping of quantitative trait loci.
Genetics 136:1457–1468
Zhang J, Marshall JD (1995) Variation in carbon isotope discrimination and photosynthetic gas exchange among populations of Pseudotsuga menziesii and Pinus ponderosa in
different environments. Funct Ecol 9:402–412
Zoldos V, Papes D, Brown SC, Panaud O, Siljak-Yakovlev S
(1998) Genome size and base composition of seven Quercus species: inter- and intra-population variation. Genome
4:162–168