CSIRO
PUBLISHING www.publish.csiro.au/journals/asb Australian Systematic Botany 17, 145- 1 70
L.
C.
F
Bus
ADepartment of Botany, The University of Tennessee, Knoxville, TN 37996, USA.
Forest Service, Pacific Northwest Research Station, 3200 SW Jefferson Way, Corvallis, OR
°Corresponding author; email: rsmall@utk.edu
9733 1 , USA. cDepartment of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA 500 1 1 , USA.
Abstract.
Molecular data have had a profound impact on the field of plant systematics, and the application of
DNA-sequence data to phylogenetic problems is now routine. The majority of data used in plant molecular phylogenetic studies derives from chloroplast DNA and nuclear rDNA, while the use of low-copy nuclear genes has not been widely adopted. This is due, at least in part, to the greater difficulty of isolating and characterising low-copy nuclear genes relative to chloroplast and rDNA sequences that are readily amplified with universal primers. The higher level of sequence variation characteristic of low-copy nuclear genes, however, often compensates for the experimental effort required to obtain them. In this review, we briefly discuss the strengths and limitations of chloroplast and rDNA sequences, and then focus our attention on the use of low-copy nuclear sequences. Advantages of low-copy nuclear sequences include a higher rate of evolution than for organellar sequences, the potential to accumulate datasets from multiple unlinked loci, and bi-parental inheritance. Challenges intrinsic to the use of low-copy nuclear sequences include distinguishing orthologous loci from divergent paralogous loci in the same gene family, being mindful of the complications arising from concerted evolution or recombination among paralogous sequences, and the presence of intraspecific, intrapopulational and intraindividual polymorphism. Finally, we provide a detailed protocol for the isolation, characterisation and use of! ow-copy nuclear sequences for phylogenetic studies.
Introduction
The impact of molecular data on the field of plant systematics can hardly be overstated. In combination with explicit methods for phylogenetic analysis, molecular data have reshaped concepts of relationships and circumscriptions at all levels of the taxonomic hierarchy
(Qiu et a!. 1 999; Soltis et a!. 1 999; Crawford 2000). As molecular phylogenetic studies have accumulated, it has become apparent that different molecular tools are required for different questions because of varying rates of sequence evolution among genomes, genes and gene regions. The choice of molecular tool is of paramount importance to ensure that an appropriate level of variation is recovered to answer the phylogenetic question at hand. Nonetheless, the plant systematics community is using only a small fraction of the available molecular tools. The preponderance of molecular data applied to plant systematics problems come from two sources: chloroplast DNA (cpDNA) or nuclear ribosomal DNA (rDNA). While the contributions of cpDNA and rDNA to plant systematics are undeniable, reliance on
© CSIRO 29 April 2004 these tools to the exclusion of other, perhaps more appropriate, tools is pervasive (Alvarez and Wendel 2003).
Alternatives to cpDNA and rDNA include both mitochondrial (mtDNA) and nuclear (nDNA) sequences other than rDNA. Because of its generally slow rate of sequence evolution and fast rate of structural evolution
(Palmer 1 992; Palmer et al. 2000), mtDNA generally has been ignored by plant systematists as a potential source of data (but see e.g. Qiu et al. 1 998, 1 999; Freudenstein and
Chase
200 1 ;
Anderberg et al.
2002;
Sanjur et al.
2002).
For this reason, mtDNA will not be considered further in this review. Nuclear sequences other than rDNA represent most of the DNA contained in any given cell, comprising both high-copy repetitive DNA (e.g. transposons, centromeric and telomeric repeats), and low- to moderate-copy DNA elements, including the majority of genes. The evolutionary dynamics of these two classes of DNA can be dramatically different. For example, repetitive DNA may experience non-Mendelian transmission, be subject to concerted evolution, and/or be mobile within a genome (e.g.
I 0. 1 07 1/SB030 1 5 I 030- 1 887/04/020 145

146 Australian Systematic Botany R. L. Small et a/. transposable elements). Accordingly, distinguishing sequences related by descent (orthologs) from a sea of related (but non-orthologous) sequences may be a challenge.
Low-copy nuclear sequences on the other hand, typically evolve independently of paralogous sequences and tend to be stable in position and copy number (but see e.g. Fu and
Dooner 2002), thereby facilitating identification and isolation of orthologous sequences. It is this low-copy fraction of nONA that we wish to focus on here, particularly with reference to applications at the level that most systematists work, namely, among closely related species within one to several genera. We briefly review the pros and cons of using the now-traditional cpDNA and rONA sequences, We then focus our attention on low-copy nuclear sequences, . describing methodological, biological and conceptual issues surrounding their use in phylogenetic analysis. Finally, we describe a generally applicable strategy for the isolation, characterisation and phylogenetic use of low-copy nuclear genes.
Chloroplast DNA
The most widely used source of data in plant molecular phylogenetic analyses has been cpDNA, either in the form of restriction site or DNA sequence analysis. The use of cpDNA has been reviewed extensively (Olmstead and Palmer 1994;
Soltis and Soltis 1998a) and will not be belabored here, but several points merit restating, with respect to the nature of cpDNA evolution and the general utility of cpDNA relative to nONA.
Evolut ionary dynamics of cpDNA
The primary advantages of cpDNA as a molecular tool lie in its relatively simple genetics. The chloroplast genome is a circular molecule found in multiple copies per chloroplast, and contains both coding (gene) and non-coding (intron and intergenic spacer) sequences. The presence of multiple copies of the chloroplast genome per chloroplast, coupled with the presence of multiple chloroplasts per leaf cell, means that cpDNA is relatively high copy within a typical genomic DNA prep. This property is a significant advantage as the high copy number facilitates restriction site analysis as well as PCR amplification of specific cpDNA regions because higher copy-number sequences are more readily accessible.
Chloroplast DNA is generally characterised as structurally stable, haploid, non-recombinant and generally uniparentally inherited (primarily maternally in angiosperms, although examples of paternal and biparental inheritance are also known), all features that facilitate its use in systematic studies. Structural stability across large evolutionary scales has been demonstrated by comparative cpDNA mapping and sequencing, and numerous studies have shown that cpDNA molecules are highly conserved with respect to gene content and arrangement, especially among closely related species (Olmstead and Palmer 1994).
As evolutionary divergence increases, structural mutations
(inversions, indels, and expansion/shrinkage of the inverted repeat) become more pronounced, but overall gene content and order remain remarkably consistent. This structural stability has facilitated the design of 'universal' PCR primers and heterologous cpDNA probes. In addition, and most importantly, it allows the. expectation that specific DNA sequences isolated from different species are orthologous.
Haploidy, uniparental inheritance and the absence of recombination among cpDNA molecules are also important features. A central assumption of phylogenetic analysis is that terminal taxa are the product of bifurcating lineage-splitting events, rather than products of reticulation.
For haploid (and thus non-recombinant), uniparentally inherited cpDNA molecules, relationships are by definition bifurcating rather than reticulate, and hence these are appropriate terminals for phylogenetic analysis (Doyle
1992). Additionally, to the extent that cpDNA is haploid, intra-individual (allelic) variation is absent, thus simplifying analyses. The haploid nature of cpDNA also serves to reduce the amount of intraspecific and intrapopulation variation.
Since the effective population size of a haploid genome is smaller than that of a diploid genome (114 in dioecious plants; 1/2 in monoecious plants), coalescence times and time to fixation of cpDNA haplotypes within a population are short relative to diploid genomes.
These generalisations are, however, not without exceptions. For example, while organellar genomes are assumed to be non-recombinant, evidence from Pinus contorta (Marshall et at. 2001) suggests cpDNA recombination may occur in plants, similar to reports of mtDNA recombination in hominids (Awadalla et at. 1999;
Eyre-Walker 2000). Uniparental inheritance is usually assumed in cpDNA studies, and while most studies have confirmed uniparental inheritance of organellar genomes
(maternal in angiosperms, paternal in gymnosperms), exceptions, including both paternal and biparental inheritance, have been reported in angiosperms (Birky 1995;
Corriveau and Coleman 1988; Reboud and Zeyl 1994). The primary effect of these processes on molecular systematics is an increase in homoplasy over what might be expected from a uniparentally inherited, and thus non-recombining molecule.
There is some irony in the realisation that the properties of cpDNA that make it an attractive tool for molecular systematics also hinder its overall utility in phylogenetic analyses. These stem from the propensity of plants to hybridise and undergo polyploidisation. Because cpDNA is uniparentally inherited and haploid, it reveals only half of the parentage in plants of hybrid or polyploid origin (generally the seed parent in angiosperms because of maternal inheritance). Thus, cpDNA analysis of hybrid or polyploid plants may incorrectly identify them as belonging to a clade

Use of nuclear genes in plant phylogeny Australian Systematic Botany 147 of one of the two parents, without revealing the hybrid history. If hybridisation is followed by introgression and subsequent fixation of the alien cpDNA, then the phylogeny may be accurately resolved with regard to the maternal history; however, cpDNA fails to identify the phylogenetic conflict arising from a hybrid ancestry.
Evolut ionary rates of cpDNA
Generally, in plants the mitochondrial genome evolves at the slowest rate, the chloroplast genome at a slightly faster rate and the nuclear genome at the fastest rate (Wolfe, Li et a!.
1987;Gaut 1998). There is, of course, substantial rate variation within genomes, and coding regions evolve more slowly than non-coding regions (introns and intergenic spacers), presumably because of selective constraints. For this reason, cpDNA gene sequences (e.g. rbcL, atpB, matK and ndhF) have been used extensively at the family level and above (Chase et a!.
1993; Qiu et al.
1999; Soltis et a!.
1999), while non-coding sequences such as introns. (e.g. r pL16, rpoC I, rpS
16, t rn
L, trnK) and intergenic spacers (e.g. trnT -trnL, trnlr-trnF, atpB-rbcL, psbA-trnH) are used more frequently at lower taxonomic levels (Taberlet et a!.
1991; Sang et al.
1997a; Small et al.
1998). Given the relatively slow rate of cpDNA evolution, however, even non-coding cpDNA regions often fail to provide significant phylogenetic information at low taxonomic levels (Small et al. 1998). This latter problem represents a significant limitation of cpDNA sequences for phylogenetic reconstruction, as applicability is restricted to relatively deep divergence histories.
Nuclear ribosomal DNA
To compensate for the limitations of cpDNA, as well as to obtain additional and independent estimates of phylogeny, nuclear rDNA has been widely adopted as a tool in plant systematics, and is now as commonly used as cpDNA. At higher taxonomic levels the slowly evolving rRNA genes are used (Hamby and Zimmer 1988, 1992; Steele et a!. 1991;
Soltis et a!. 1997; Kuzoff et a!. 1998; Soltis and Soltis
1998b ), while at lower taxonomic levels internal and external intergenic spacers are more commonly employed (Baldwin et a!. 1995; Alvarez and Wendel 2003; Bailey et a!. 2003).
Evolut ionary dynam ics ofrDNA
In land plants rRNA genes are organised into two distinct sets of tandem arrays. The first set is composed of 5S rRNA genes and intergenic spacers in tandem arrays at one or more chromosomal loci. The second set includes the
18S-5.8S-26S rDNA cistron in tandem arrays at one or in ore chromosomal loci. Both sets of rDNA arrays have been used in systematic studies, although the 18S-5.8S-26S arrays have been used far more frequently than 5S. As for cpDNA, the potential phylogenetic utility of rDNA is facilitated by its structure and molecular evolution. Ribosomal genes exist in tandem arrays of genes composed of hundreds to thousands of copies per array (Baldwin et al. 1995; Cronn et a!. 1996;
Hanson et a!. 1996b; Kuzoff et a!. 1998; Soltis et a!. 1997;
Soltis and Soltis 1998b; Wendel et al. 1995a). The higher copy number facilitates evaluation of rDNA by both restriction site- and PCR-based strategies. In addition, the repetitive structure of these arrays promotes a process of homogenisation ('concerted evolution': Zimmer et a!. 1980;
Arnheim 1983) that may result in a single predominant sequence across all copies and arrays. This homogenisation allows PCR products to be directly sequenced, generally yielding a single dominant sequence that is assumed to be representative of the underlying genomic sequences.
W hile rDNA has contributed substantially to our present understanding of plant systematics, the highly repetitive nature of rDNA gives it properties that effect the potential utility and reliability of these regions in phylogenetic studies
(Alvarez and Wendel 2003; Bailey et al. 2003). These potential difficulties stem from the structure and evolutionary dynamics of rONA-specifically, the presence of both multiple copies per array and multiple arrays per genome, and the presence, absence, or variable strength of concerted evolution. As noted above, concerted evolution tends to homogenise sequences within and sometimes between rDNA arrays. The important phrase here is 'tends to.' The strength of concerted evolution is far from uniform across repeat units, arrays and taxa. In the absence of complete concerted evolution, sequence variants can arise and be maintained within and between arrays, yielding multiple distantly related rDNA types within individuals
(Suh et al. 1993; Cronn et a!. 1996; Buckler et al. 1997;
Hartmann et a!. 2001; Mayo! and Rossello 200 I; Muir et a!.
2001; Bailey et al. 2003). Indeed, rDNA pseudogenes may be regular constituents of many genomes, representing phylogenetically distant sequences that may preferentially be amplified over functional rDNA loci (Buckler and Holtsford
1996; Buckler et al. 1997; Hartmann et a!. 2001; Mayo! and
Rossello 2001; Chase et a!.
2003).
More problematic, however, is the possibility that such erratic variation is the norm for rDNA rather than the exception, even though such findings may be generally ignored, underreported or undetected (Alvarez and Wendel
2003). For example, polymorphic positions are found frequently in published rDNA datasets. Such polymorphisms generally are simply coded as polymorphic characters for phylogenetic analyses, ignoring the fact that these positions are evidence of unhomogenised sequence variants, perhaps of distantly related sequences. In addition, sequence polymorphism may go undetected in automated or manual sequencing, or else the strongest peak (or band) at a position may be scored as the 'correct' base. Finally, PCR bias may selectively amplify one of the genomic sequences because of differences in genomic copy number or primer affinity
(Wagner et a!. 1994). The sum of these observations is that

148 Australian Systematic Botany R. L. Small et a/. rONA sequences obtained by direct sequencing of PCR products may fail to reveal the complexity of nuclear rONA content, and may in fact preferentially reveal paralogous rONA sequences in different taxa or accessions.
While the prevalence of this phenomenon can be difficult to ascertain in diploid species, the problem becomes exacerbated in allopolyploid
· and hybrid taxa in which multiple divergent rONA loci are expected to exist on different chromosomes donated by the different parents. For example, allotetraploid (AD-genome) species of Gossyp ium contain genomes donated from an African A-genome diploid and a New World 0-genome diploid. Direct sequencing of
PCR-amplified ITS fragments (Wendel et a!. 1 995a) from the five allotetraploid species revealed only a single rONA sequence type characteristic of either the A- or 0-genome
(but not both), despite the fact that Southem hybridisation
(Wendel et al. 1 995a) and FISH (Hanson et al. 1 996b) evidence indicated that both genome types were maintained in the allotetraploid. Similar results have been reported in allotetraploid Glyc ine (Rauscher et a!. 2002) and elsewhere
(reviewed in Alvarez and Wendel 2003; Bailey et a!. 2003) where biased amplification of ITS sequences from specific parental diploid genomes was observed, perhaps because of differential underlying copy number.
Evolut ion m y rates ofrDNA
One of the reasons for adding rONA sequences to the arsenal of tools available to plant molecular systematists is to obtain data from a source independent of cpDNA. The other primary reason, especially at lower taxonomic levels, is to obtain sequences that evolve at a faster rate, so that more phylogenetically informative characters are obtained. In phylogenetic studies at low taxonomic levels in which both non-coding cpDNA and ITS data are collected for a set of taxa, ITS sequences generally provide greater levels of divergence and thus greater resolution and stronger support than an equivalent sample of cpDNA sequence (e.g. Hodges and Arnold 1 994; Gielly et a!. 1 996; Sang et a!. 1 997a;
Whitten et a!. 2000). The more rapid rate of evolution ofiTS, however, is tempered by its relatively short length (generally
500-600 bp in angiosperms) and the relatively high levels of homoplasy (Alvarez and Wendel 2003). Recently, attempts to add the flanking external transcribed spacer (ETS) of nuclear rONA to supplement ITS sequences have met with some success (Baldwin and Markos 1 998; Linder et al.
2000). The internally repetitive structure of the ETS region, however, can make both PCR amplification and sequence alignment difficult, thus presenting an additional obstacle to the widespread .adoption of ETS in systematic studies.
Finally, it is important to note that because of the linked nature of the 45S rONA cistron, the ETS region will fall prey to the same pitfalls and problematic evolutionary dynamics as ITS. For this reason, congruence between ITS- and
ETS-derived topologies provides not so much an independent confirmation of phylogenetic signal as confirmation of genetic and physical linkage.
Low-copy nuclear genes
While the use of low-copy nuclear genes for phylogeny reconstruction is still in its relative infancy, several conclusions may be drawn regarding both utility and limitations. Among the advantages of nuclear genes are an overall faster rate of sequence evolution than in organellar genomes, the presence of multiple independent (unlinked) loci and biparental inheritance. Disadvantages of nuclear loci stem primarily from the more complex genetic architecture and evolutionary dynamics of the nuclear genome and possible difficulties in isolating and identifying orthologous genes. Other relevant issues include the possibilities of concerted evolution and/or recombination among paralogous sequences and the presence of intraspecific, intrapopulational and intraindividual variation
(heterozygosity).
Advantages of nuclear genes
Rate var iat ion in nuclear genes
One of the primary advantages of nuclear genes for phylogenetic analysis is the elevated rate of sequence evolution relative to organellar genes. Broad surveys have revealed that synonymous substitution rates of nuclear genes are up to five times greater than those of chloroplast genes and 20 times greater than those of mitochondrial genes
(Wolfe et a!. 1 987; Gaut 1 998). This elevated evolutionary rate yields a greater efficiency of sequencing effort, since more variation is detected per unit of sequence than in organellar genes. This advantage becomes particularly acute at low taxonomic levels (Small et al. 1 998) or among lineages created by rapid divergence events (Cronn et al.
2002b; Malcomber 2002). For example, in an analysis of the relationships among the allotetraploid species of Gossyp ium
L. (Small et al. 1 998), direct comparison of non-coding cpDNA and nuclear-encoded alcohol dehydrogenase (AdhC) sequences showed that relationships were incompletely resolved and poorly supported by >7 kb of non-coding cpDNA, while data from a 1.65-kb section of AdhC provided complete and robustly supported resolution of relationships.
Extrapolation of these results suggested that to obtain equivalent phylogenetic resolution from the cpDNA as the
AdhC sequences, >40 kb of non-coding cpDNA would have to be sampled.
While this example highlights a potential advantage of nuclear genes in terms of a faster evolutionary rate, it ignores the huge range of evolutionary rates observed among nuclear genes, regions within genes, and even sites within genes. For example, the selection of
AdhC for the study of Small et a!.
( 1 998) was fortuitous, as this gene is the most rapidly evolving of the
Gossyp ium Adh gene family (Small and

Use of nuclear genes in plant phylogeny Australian Systematic Botany 149
Wendel 2000a) and the fifth most rapidly evolving gene among a sample of 49 nuclear genes in Gossypium (Senchina et a!. 2003). Variation in overall evolutionary rate is observed whenever multiple genes are sampled in a given phylogenetic context. For example, in comparisons of A- and D-genome diploid cottons for the Adh gene family of Gossypium, a greater than 3-fold range of variation was observed for synonymous sites (Small and Wendel 2000a). Broader sampling of 16 mostly anonymous nuclear loci from
Gossypium showed a 5-fold range of variation in overall evolutionary divergence (Cronn et a!. 1999), and a recent study of 36 'fibre-expressed' genes from the same species
(Senchina et a!.
2003) revealed a 7-fold range of variation in overall divergence and a 2.9-fold range in synonymous substitution rates. These ranges are in accordance with those derived from other well studied groups (Gaut 1998; Mathews et a!. 2002). Such wide ranging variation will clearly influence the lineage-specific utility of a given nuclear gene, and points to the importance of screening multiple loci and choosing markers on the basis of preliminary evidence from the taxa of interest.
In addition to rate variation among genes, there is rate variation among regions within genes. Nuclear-encoded genes can generally be divided into different functional regions, such as the 5' untranslated region (UTR), exons, introns and 3' UTR. Given the variable functions of these regions, and thus variable evolutionary constraints on them, considerable variation in evolutionary rates exists within a single locus. Important promoter elements responsible for gene regulation are generally found in the
5' UTR which may include relatively conserved domains
(Wray et a!. 2003). In some cases, however, 5' UTRs may contain introns that are highly variable (Liu et al. 2001).
Exons are likely to be more conserved at non-synonymous sites (primarily 1st and 2nd codon positions) while synonymous 3rd codon positions typically diverge at rates similar to those of non-coding regions. Nuclear spliceosomal introns generally are under fewer functional constraints at the sequence level, although there often are length limits on introns that are important for proper mRNA processing, and occasionally regulatory elements lie within introns (e.g. Ahlandsberg et a!. 2002; Hong et al. 2003) that also are likely to be highly conserved.
Elements in the 3'-UTR sequences control mRNA processing and poly-A addition signals, but otherwise are often highly variable. Perhaps counterintuitively, accumulating evidence indicates that synonymous sites within exons (i.e. 3rd codon positions) may exhibit evolutionary rates that are equal to or greater than those found at 'silent' non-coding sites (introns, UTRs)
(Charlesworth and Charlesworth 1998; Small and Wendel
2002; Senchina et a!. 2003), presumably because of conserved structural features and/or existence of regulatory domains in the latter.
Multiple independent loci
In addition to faster evolutionary rates, nuclear genes offer another vitally important feature-multiple unlinked loci may be used for independent phylogenetic inference.
Corroboration of phylogenetic hypotheses by independent datasets increases confidence in a given phylogenetic tree.
Likewise, incongruence between datasets has been of use to infer evolutionary phenomena such as hybridisation, introgression or lineage sorting (Wendel and Doyle 1998). In plant systematics, such comparisons are typically between cpDNA and nrDNA datasets. Given the complete linkage among cpDNA sites owing to its structure (a single circular chromosome) and its non-recombining nature, multiple cpDNA datasets are expected to converge on a single topology (irrespective of whether or not this reflects the organismal phylogeny). Ribosomal DNA offers a single alternative from the nuclear genome, but if incongruence between rDNA and cpDNA is identified, independent data must be obtained to sort out the source of the incongruence.
Moreover, and as alluded to above and discussed at length elsewhere (Alvarez and Wendel 2003), rDNA sequences may prove phylogenetically unreliable under certain conditions.
In Gossypium, unexpected and anomalous phylogenetic results obtained with rDNA (Wendel et a!.
1995a, 1995b) have been reconciled only through the use of multiple, unlinked low-copy loci (Small et al. 1998; Liu et a!. 200 I;
Cronn et a!. 2002b, 2003), each of which provides an independent evaluation of the unexpected results from rDNA.
Low-copy nuclear genes are capable of providing a virtually limitless source of additional, independent phylogenetic information. Because eukaryotic nuclear genomes are composed of multiple chromosomes that undergo recombination, genes found on different chromosomes, or even on the same chromosome if sufficiently far apart, are effectively evolutionarily independent of each other. Factors that may give rise to phylogenetic incongruence between datasets (e.g. hybridisation/introgression; non-homologous gene conversion) are likely to affect a limited number of genes in a localised region. Acquisition of data from multiple independent regions allows an opportunity to determine phylogenetic relationships supported by a majority of the data and can highlight particular genomic regions that may be problematic (Cronn et al. 2002b, 2003; Rokas et al.
2003).
Biparental inheritance
A final desirable property of low-copy nuclear genes is their explicitly biparental Mendelian inheritance. Because the chloroplast genome is usually uniparentally inherited, cpDNA datasets can tell only half the story in cases of hybridisation or allopolyploidisation. Likewise, while

150 Australian Systematic Botany diploid species genomes AA
2A 1A diploid species genomes BB
18 28 diploid species genomes CC
2C 1C
R. L. Small et a/. locus 1 locus 2 ancestral locus
Flg. 1. Hypothetical scenario of gene duplication, followed by speciation events to depict relationships among orthologous, paralogous and homoeologous gene copies. Orthologous genes are related solely by speciation (e.g.
!A and 18). Paralogous genes are related by gene duplication and are found both within species (e.g. ! A and 2A) or between species (e.g. lA and 28). Homeologous genes are orthologous genes that are related by polyploidisation (e.g. ! A' and IC'). A gene duplication occurs near the base of the tree, resulting in two genetic loci (Locus 1 and Locus 2). Two speciation events give rise to three extant species (Species A-C), each of which is diploid and thus contains two haploid genomes (AA, 88 and CC,respectively). Hybridisation between species
A and C, followed by polyploidisation results in the tetraploid AC species (with four haploid genomes, AACC).
Gene copies in the tetrl!ploid are noted with a' to distinguish them from copies found in the diploid progenitors.
Orthologous genes: l A, 18, lC; 2A, 28, 2C. Paralogous gene pairs: ! A, 2A; 18, 28; l C, 2C; ! A', 2A'; lC', 2C'; lA, 28; lA, 2C; lA, 2A'; lA, 2C'; 18, 2A; 18, 2C; 18, 2A'; 18, 2C'; IC, 2A; IC, 28; IC, 2A'; IC, 2C'; lA',
2A; lA', 28; lA', 2C; lA', 2C'; IC', 2A; IC', 28; IC', 2C; IC', 2A'. Homeologous gene pairs: lA', IC'; 2A',2C'. nrDNA is biparentally inherited, the process of concerted allotetraploid
Gossyp ium based on cpDNA identified the evolution, array expansion/contraction and the presence of maternal lineage (Wendel 1989; Wendel and Albert 1992), paralogous sequences can make isolation of both parental and ITS analyses were complicated by bi-directional copies difficult, if not impossible (Rauscher et a!. 2002; interlocus concerted evolution (Wendel et a!. 1995a),
Alvarez and Wendel 2003). analyses of low-copy nuclear genes from tetraploid
In contrast, low-copy nuclear genes are less frequently
Gossyp ium have unambiguously identified the ·Jineages subject to concerted evolution (although exceptions have representing the parental donor species (Small et a!. 1998; been reported-see below), thus making them ideal Cronn et al. 1999, 2002b; Small and Wendel 2000a; Liu candidates for identifying parental donors of suspected eta!. 2001). Low-copy nuclear genes have been used to hybrids or polyploids. W hile phylogenetic studies of identify the origins of hybrid or allopolyploid taxa in a

Use of nuclear genes in plant phylogeny Australian Systematic Botany 1 5 1 number of other groups as well (Sang et a!. 1 997b; Doyle et al. 1 999a, 1999b; Ford and Gottlieb 1999; Ge et a!. 1999;
Sang and Zhang 1999; Mason-Gamer 2001).
Conceptual, methodological and biological issues in using nuclear genes
Gene families and the isolation and identification of orthologous genes
The primary disadvantage of using nuclear genes stems from the complicated genetic architecture of eukaryotic nuclear genomes and the tendency for nuclear genes to exist in gene families-multiple copies of homologous genes related by gene (or genome) duplication (Fig. 1) (Henikoff et a!. 1 997;
Thornton and DeSalle 2000). In plants, nuclear-gene families vary widely in size, ranging from genes that appear to be truly single copy in some species (e.g. GBSSI in diploid
Poaceae; Mason-Gamer et a!. 1998) to gene families with dozens to hundreds of copies (e.g. actins: Moniz de Sa and
Drouin 1996; or small heat-shock proteins: Waters 1995).
Further complicating the situation is the fact that gene and genome duplication (and subsequent gene loss) appear to be ongoing and dynamic processes (Gottlieb and Ford 1996;
Clegg et a!. 1 997; Wagner 1998, 2001; Grant et at. 2000;
Lynch and Conery 2000; Small and Wendel 2000a; Bancroft
200 I; Gaut 200 I; Ford and Gottlieb 2002). Because of this, characterisation of a gene family necessarily becomes a lineage-specific problem, and inferences of gene family structure from one lineage may not be applicable to other lineages. For example, plant Adh gene families are often considered to be small, with one to three loci present in most diploids (e.g. Chang and Meyerowitz 1986; Sang et a!.
1 997 b; Gaut et a!. 1 999). In contrast, the Adh gene families in Gossypium (Small and Wendel 2000a) and Pinus (Perry and Fumier 1996) have been shown to have a minimum of seven loci in diploid species. Phylogenetic analyses of the
Adh gene family in angiosperms suggest multiple rounds of both ancient and recent gene duplication and deletion (Clegg et a!. 1997; Small and Wendel 2000a).
Given the complexity of nuclear genomes, the need to characterise gene family composition prior to phylogenetic analysis is of paramount importance. Phylogenetic analyses implemented to estimate organismal phylogeny assume that comparisons are between strictly orthologous gene copies, and the inclusion of a mixture of orthologous and para1ogous sequences can produce robust, yet erroneous, hypotheses of relationships (Wendel and Doyle 1998). Rigorous elucidation of the size of a gene family, as well as criteria for establishing orthology should be considered a first step in preliminary studies of the potential utility of nuclear genes
(see below).
As gene family structure varies widely both among gene families and among plant lineages, initial studies of nuclear genes should focus on characterising the target gene family in the taxa of interest. The usual route taken by many investigators is to design or obtain 'universal' primers for a particular gene family, usually by comparison of gene sequences from a wide phylogenetic array of taxa, with primers located in regions of strong sequence conservation
(Sang et a!. 1 997b; Strand et a!. 1 997; Mason-Gamer et a!.
1998; Small et at. 1998; Evans et a!. 2000; Small and Wendel
2000a).
PCR using 'universal' primers frequently results in amplification of more than one product, as evidenced by either multiple amplified bands or obvious sequence heterogeneity in the PCR pool. Resolution of this sequence complexity and subsequent development of locus-specific primers generally requires isolation of individual PCR products from the heterogeneous initial amplification reaction, often accomplished by cloning PCR products and screening multiple clones. The most efficient approach is to select a few representative taxa for this initial screening.
Once all apparent sequence types from the representative taxa have been obtained, phylogenetic analysis of those sequences is performed. Ideally, these preliminary analyses should take place in the context of related sequences from other taxa in GenBank (http://www.nlm.nih.gov).
Appropriate sequences for comparison can often be obtained from GenBank by reference to the literature, searching by gene name or by BLAST searches [Basic Local Alignment
Search Tool; http://www.ncbi.nlm.nih.gov/BLAST/;
(Altschul et a!. 1 990)] for similar sequences. This exercise is an important component of a preliminary study (see below), as it can differentiate between relatively recent and ancient gene-duplication events. Confidence in assumptions of orthology generally are stronger when the putatively orthologous sequences in question are monophyletic and divergent from paralogs. Preliminary phylogenetic analyses also provide important information of relative rates of sequence evolution within and among genes (see below).
Once an initial characterisation of a gene family has been conducted and appropriate loci have been selected for further study, the next logical step is to develop locus-specific PCR primers. This step can improve efficiency by reducing the necessity of cloning heterogeneous PCR products (except for those cases where heterozygosity is encountered).
Locus-specific primers have the added benefit of eliminating the possibility of recovering cloned PCR-generated chimeras
(PCR-mediated recombinants) that can form whenever two highly similar templates are co-amplified in a single PCR reaction (Bradley and Hillis 1 997; Cronn et a!. 2002a). If homogeneous sequences are obtained from direct sequences of PCR products from all taxa of interest there is also greater confidence that a single orthologous locus is being amplified and sequenced from all taxa.
Once a group of candidate loci have been identified, orthology must be evaluated empirically. A number of criteria that vary in their methodology and assumptions have

152 Australian Systematic Botany R. L. Small et a/. been suggested as evidence of orthology. These criteria include overall sequence similarity, tissue specificity and expression patterns (Doyle 1991; Doyle and Doyle 1999),
Southern hybridisation analysis (Cronn and Wendel 1998;
Evans et a/. 2000; Small and Wendel 2000a) and comparative genetic mapping (Cronn and Wendel 1998;
Small and Wendel 2000a; Cronn et al. 2002b).
The simplest approach to identifying orthologs is through sequence similarity and phylogenetic analysis. Clearly, orthologs are expected to be more similar and phylogenetically more closely related to each other than to any paralogous sequence, assuming complete sampling of genes within all taxa. This expectation can be violated, however, for at least two reasons. First, complete sampling of all genes of a gene family in all taxa may not be accomplished, owing to the challenges of generating and isolating all relevant gene copies from each taxon in a study.
PCR amplification with 'universal' primers may result in differential amplification of loci in different taxa because of imperfect pairing between PCR primers and templates or relative qualities of template DNA (Wagner et a!. 1994).
Subsequent cloning of heterogeneous PCR products samples only a subset of the PCR products present, and screening of numerous clones (by sequencing or restriction digestion) is required to differentiate orthologs from paralogous sequences. Subsequent phylogenetic analysis of the sequences may reveal two sequences from different taxa that are more closely related to each other than either is to suspected paralogs; yet these sequences may indeed be paralogous rather than orthologous if they are related by a recent gene-duplication event and both paralogs are not sampled in both taxa. The second issue that bears on this approach is the problem of either in v ivo or in v itro
(PCR-mediated) recombination. In this scenario, closely related paralogous loci may undergo non-homologous recombination, and phylogenetic analysis of the resulting chimeric sequences will confound rather than illuminate relationships and inferences of orthology/paralogy. It should be noted that this sequence-based method of determining orthology represents minimal evidence for orthology and is typically included as a precursor to the following three methods of homology assessment.
A second approach to identifying orthologs is to use shared expression patterns as evidence of orthology (Doyle
1991; Doyle and Doyle 1999). One of the features of nuclear-gene families may be differential expression patterns
(either timing or tissue specificity) of paralogous genes. In fact, gene families may exist primarily because paralogous copies are adapted to function differentially, and if orthologous genes serve the same function in different species they can be expected to maintain similar expression patterns. However, expression data may be difficult to obtain, often requiring either detailed RNA (e.g. Northern blot or RT -PCR) or protein (Western blot or isozyme) analysis. In some cases, expression patterns can be inferred from sequence data, as is the case with gene families that have both cytosolic- and organellar-expressed genes. Such cases usually represent ancient gene duplications that are readily identifiable from the sequence data alone and the evidence of orthology is clear. For example, both the phosphoglucose isomerase (PGI) (Gottlieb and Ford 1996;
Ford and Gottlieb 1999; Ford and Gottlieb 2002) and glutamine synthase (GS) (Doyle 1991; Emshwiller and
Doyle 1999; Emshwiller and Doyle 2002) gene families contain both cytosolicand chloroplast-expressed paralogous gene copies. Sequence conservation among these classes allows ready identification of the expression pattem of each paralog. Difficulties may arise, however, if a relatively recent gene duplication creates paralogous gene copies with similar or identical expression patterns, highlighting a second difficulty of using expression patterns as a criterion of orthology. Specifically, recent gene duplications within a particular expression class may result in paralogous gene copies that share both strong sequence similarity and expression patterns. A recent study of gene pairs duplicated by polyploidy in allotetraploid cotton
(Adams et al. 2003) shows how subsequent expression patterns of duplicated genes may be not only tissue specific, but locus- or genome-specific. This could create a situation whereby one copy (e.g. the A' genome homoeolog in Fig. 1) would be expressed in one species, while the other copy (e.g. the C' genome homoeolog in Fig. 1) could be expressed in a closely related species. For this reason estimating orthologous relationships among sequences on the basis of tissue-specific mRNA pools could prove particularly challenging, especially if polyploids were included in the analysis. An additional example is provided by the chloroplast-expressed PGI genes of Clark ia (Onagraceae)
(Gottlieb and Ford 1996; Ford and Gottlieb 1999; Ford and
Gottlieb 2002).
A third and more practical criterion that can strengthen inferences of orthology are Southern blot hybridisation experiments. Once a suite of sequences have been obtained from a given set of taxa, comparisons of sequence identity between classes of loci can be used to identify regions that are 'locus-specific,' i.e. sequence motifs found in only one of the sequence types or are sufficiently divergent between loci to minimise cross-hybridisation under conditions of high stringency. Generally, introns and 3' UTRs make the best candidates for such motifs. Locus-specific hybridisation probes are then designed from these regions, amplified by
PCR, and used to probe restriction-digested genomic DNAs of representative taxa. Results from high-stringency hybridisations using these ideal probes provide an opportunity to 'count genes,' since each band on the resulting autoradiograph represents a single locus. If a single band is detected using this approach in the taxa examined, this constitutes strong evidence that the locus is unique, and

Use of nuclear genes in plant phylogeny Australian Systematic Botany ! 53 by inference, orthologous among taxa. If multiple bands are detected under stringent hybridisation conditions, then either
( 1 ) a restriction site was present within the probe region
(which can be avoided by choosing appropriate enzymes and/or evaluating multiple enzymes), or (2) there are multiple, closely related (but paralogous) loci that share high sequence identity with the probe.
An example of the power of this approach is provided by the Adh gene family of Gossypium that contains a minimum of seven loci. Southern hybridisation experiments were informative in identifying both single-copy orthologous loci shared among species as well as documenting a small gene
'subfamily'-a group of closely related genes resulting from recent gene duplications (Small and Wendel 2000a). The
AdhA gene in Gossypium is a truly single-copy gene, and this is reflected in the Southern hybridisation results-using a small intron probe, a single band was detected in diploid species screened, and two bands were detected in the allotetraploid species, reflecting the additivity of the loci contributed from their diploid progenitors.
'AdhB' in Gossypium, on the other hand, was found to include several closely related sequence types. Again, by using a small intron probe, Southern hybridisation showed a complex banding pattern with several hybridising fragments, despite the fact that PCR-based experiments had isolated only a single AdhB-like sequence type. This conundrum was resolved when independent sequence data from genomic and eDNA clones of Adh sequences from Gossypium were described (Millar and Dennis 1 996). The Adh sequences described by (Millar and Dennis 1 996) were subsequently included in a phylogenetic analysis, with Adh sequences isolated via PCR in our laboratory (Small and Wendel
2000a) and found to be closely related but not orthologous to
AdhB, and thus, to constitute a small gene subfamily that presumably resulted from recent gene-duplication events.
Given these data and evidence of potential interlocus recombination (Millar and Dennis 1 996), it was determined that
AdhB would be a poor choice for use in phylogenetic studies.
Screening exemplar taxa by Southern hybridisation, while clearly the most efficient approach for counting genes, may not always prove completely effective. While preliminary screening of AdhA in Gossypium indicated that AdhA was a single-copy gene in the three diploid species surveyed
(spanning the diversity of Gossypium), a subsequent phylogenetic study using
AdhA of the 13 D-genome diploid species showed a previously undetected AdhA gene-duplication event (Small and Wendel 2000b). During the course of data collection for this study, all species surveyed showed little to no sequence heterogeneity, with the exception of the group of four species that constitute
Gossypium section Erioxytum subsection Erioxytum. Direct sequencing of PCR products of AdhA (amplified using AdhA locus-specific primers) detected extensive sequence heterogeneity in these four species. Given this information, we conducted Southern hybridisation experiments on these species and found that while all other D-genome species displayed a single AdhA-hybridising fragment, all species of section Erioxytum displayed two hybridising fragments.
Apparently, a gene-duplication event confined to these species had occurred, and both loci were being amplified with our PCR reaction conditions.
Finally, the most rigorous approach for demonstrating orthology can be obtained by com
arative gen tic-mapping studies. In practice, such analyses are restricted to taxa for which genetic-mapping experiments are already underway for other reasons, simply because of the extensive time and effort required for comparative genetic mapping.
Nevertheless, it is important to note that comparative mapping studies have been completed for a wide variety of plant lineages (especially crop plants), including, for example, Asteraceae, Brassicaceae, Chenopodiaceae,
Fabaceae, Malvaceae, Pinaceae, Poaceae, Rosaceae and
Solanaceae. Information from these 'data rich' species can be used to bolster inferences of orthology to more distant relatives. W hile evidence from sequence identity, shared expression patterns and Southern hybridisation analysis can all assist the process of inferring orthology, the retention of shared genomic position among species is arguably the most rigorous evidence of orthology.
Gene families and rate variation among genes
In addition to the requirement of orthology, rate variation is an important consideration when choosing an appropriate gene. Preliminary data from representative species for all . potential loci provide the raw data required to inform this decision. Given the observation of extensive rate variation among nuclear genes at all levels (among gene families, among genes within gene families, among regions within genes, among plant lineages: see e.g. Gaut 1 998; Gaut et at.
I 996; Sen china et at. 2003; Small and Wendel 2000a), the gene selected should provide an appropriate level of sequence variation to answer the question being asked. The appropriate level of variation depends on the level of resolution being sought: inter- or intrafamilial studies may utilise more slowly evolving sequences than studies at the inter- or intraspecific level. Further, the regions of a particular gene that are to be used can vary from question to question-exons are generally easily alignable across wide phylogenetic distances (in many cases even among extant land plants), while introns are often unalignable outside of individual genera. Accordingly, questions focused at higher taxonomic levels may choose to exploit regions with high exon content, while lower-level studies may emphasise intron sequence.
W hile examples of particular plant lineages that have been broadly sampled for nuclear genes are relatively few, the available examples highlight the range in variation (and

! 54 Australian Systematir; Botany R. L. Small et a/. thus phylogenetic utility) found thus far. Our . work in
Gossyp ium has resulted in characterisation of a large number of nuclear genes across the well established phylogeny of the primary genome groups. More than 50 nuclear loci have been sequenced· from multiple species, and the range of phylogenetic utility spans from an almost complete lack of variation to highly variable loci that provide robustly resolved and supported depictions of relationships (Small et a!. 1 998; Cronn et a!. 1 999, 2002b; Small and Wendel 2000a,
2000b; Senchina et al.
2003).
Other well studied examples include Zea (Gaut and Clegg
1 993; Hanson et al. 1 996a; Gaut and Doebley 1 997;
Eyre-Walker et a!. 1 998; Gaut 1 998, 2001; Hilton and Gaut
1 998; Zhang et al. 2001), Poaceae (Mathews and Sharrock
1 996; Mason-Gamer et a!. 1 998; Mathews et a!. 2000, 2002;
Mason-Gamer 2001), Brassicaceae (Galloway et al. 1 998;
Bailey and Doyle 1 999; Bailey et al. 2002), Paeon ia (Sang et a!. 1 997 b; Sang and Zhang 1 999; Ferguson and Sang
2001; Tank and Sang 2001; Sang 2002), and Glyc ine (Doyle et al. 1 996, 1 999a; 1 999b, 2000, 2002; Doyle and Doyle
1 999). In all of these examples, variation in the relative phylogenetic utility of nuclear genes is evident, again highlighting the need for preliminary studies to determine the most appropriate locus (or loci) for a given question.
Intraspecific var iat ion in nuclear genes
Two features of nuclear genes that require attention in phylogenetic studies are the high probability of allelic variation within and among populations of a species, and alleles that are shared between species. Because of the smaller effective population size of organellar genes, and concerted evolution in nrDNA sequences, allelic variation is often low for these markers. Accordingly, a single or just a few individuals of a species are often sampled as placeholders, with the implicit assumption that all alleles from individuals of that species will be more closely related to each other than they are to any other species. When sampling of individuals is increased, this assumption is often borne out, although exceptions do exist, especially with nrDNA (Suh et al. 1 993; Mason-Gamer et al. 1 995; Levy et al. 1 996; Mayer and Soltis 1 999; Hartmann et al. 2001;
Mayo! and Rossello 2001; Muir et a/. 2001). Allelic variation within individuals, populations and species in low-copy nuclear genes, however, can be extensive. This is due to the larger effective population size of nuclear genes than that of organellar genes, the process of allelic recombination and the faster rate of molecular evolution of nuclear genes than that of organellar genes. Evidence of such variation was first recognised in isozyme analyses (Crawford 1 985; Gottlieb
1 977), where extensive allelic variation was detected in many species, but also shared allelic variation among species was evident. As results from sequencing studies of plant nuclear genes have accumulated, the presence of substantial allelic variation has been confirmed, although the majority of this work has been on model systems such as maize,
Arab idops is and cotton (Miyashita et al. 1 996; Gaut and
Doebley 1 997; Innan and Tajima 1 997; Kawabe et al. 1 997;
Hilton and Gaut 1 998; Purugganan and Suddith 1 998;
Kawabe and Miyashita 1 999; Small et al. 1 999; Kuittinen and Aguade 2000; Small and Wendel 2002).
Allelic variation within species may take two forms, with different implications for phylogenetic studies. First, irrespective of the extent of allelic variation, if all alleles of a species coalesce within that species (i.e. all alleles of a species are more closely related to each other than they are to any allele of a different species), then allelic variation is irrelevant to the ultimate goal of recovering a species phylogeny. This type of allelic variation may be useful, however, for intraspecific studies, e.g. ascertaining population-level relationships, phytogeography, and studies of rates and patterns of sequence evolution.
A second possibility is that allelic variation spans species boundaries (deep coalescence); i.e. some alleles of a species are more closely related to alleles of other species than they are to those of the same species. A number of population genetic phenomena can give rise to this commonly observed pattern. One primary cause lies in the population genetics of nuclear genes. Because of the greater effective population size and faster mutation rate of nuclear genes relative to organellar genes, coupled with the process of recombination, extensive intraspecific allelic variation is both expected and observed in species with sufficient population sizes. When speciation occurs, it is likely that both descendants of an ancestral species will contain some, if not all, of the allelic variation present in the ancestral species. If the alleles contained in each of the descendant species are reciprocally monophyletic, then alleles will coalesce within species and phylogenetic analysis using any alleles should reflect this history. If, on the other hand, allelic variation in one or both of the descendant species is not monophyletic, then trans-species polymorphism will be observed.
In exceptional cases, natural selection acts to promote this trans-species polymorphism. This is the case for genes that undergo balancing selection where natural selection acts to promote and maintain allelic variation. Striking examples of this phenomenon are found for genes involved in self-recognition (reviewed in Klein et al. 1 998), for example major histocompatibility
(Mhc) loci in mammals (Hughes and Yeager 1 998) and self-incompatibility (S-genes) in plants (Charlesworth and Awadalla 1 998). In the case of
S-allele variation, analyses in Solanaceae have revealed allelic variation that exceeds not only species boundaries, but even generic boundaries (Richman et al. 1 996;
Richman and
Kohn 1 999). While clearly of interest for molecular evolutionary studies, loci that tend to undergo balancing selection would be poor choices for phylogenetic analyses attempting to infer species histories.

Use of nuclear genes in plant phylogeny Australian Systematic Botany 155
Less dramatic examples of trans-specific polymorphism are also evident for genes in which segregation of ancestral polymorphism may be the sole cause. Several examples from maize show such a pattern. For example, alleles found in Zea mays ssp. mays are found phylogenetically intermingled with alleles from other subspecies of Zea mays or even other species of Zea (Hanson et al. 1996a; Eyre-Walker et al. I 998;
Hilton and Gaut 1998; Wang et a!. 1999; White and Doebley
1999; Gaut 200 I; Zhang et al. 2001 ). Studies of several nuclear genes in Leavenworth/a (Brassicaceae) have shown similar sharing of alleles across species boundaries
(Charlesworth et a 't. 1998; Liu et a!. 1998; Filatov and
Charlesworth 1999). A study of nucleotide diversity in two pairs of homeologous Adh loci in allotetraploid Gossypium hirsutum and G. barbadense revealed an unusual pattern of non-coalescence (Small and Wendel 2002). In this particular case, four loci were sampled (two AdhA and two AdhC loci, one from each of the diploid progenitors). For three of four loci, alleles coalesced within species; however, for the fourth locus (AdhC in the D-genome of the tetraploids) alleles found in G. hirsutum were placed phylogenetically into two separate clades. One of these clades included only
G. hirsutum sequences, but the second clade included all G. barbadense alleles as well as several G. hirsutum alleles. This observation is significant because, regardless of the underlying cause
(non-coalescence or introgression), it was found for only one of four loci, indicating that these population-genetic phenomena can act differentially across loci.
A second cause of trans-species polymorphism is hybridisation and introgression. A prominent feature of plant population biology, hybridisation, and the subsequent possibility of introgression, may be responsible for instances of alleles shared among species. Many well known examples have been described of introgression of chloroplast DNA
('chloroplast capture') and many isozyme studies have documented instances of putative introgression of nuclear genes (Rieseberg and Soltis 1991; Riese berg and Wendel
1993; Rieseberg 1997). Few studies have addressed the prevalence of nuclear-gene introgression, however, despite the power of phylogenetic analysis of variab1e nuclear genes to illuminate the phenomenon (Sang and Zhang 1999). In the case of trans-specific allelic variation in maize, the variation may reflect a combination of non-coalescence and introgression between subspecies of Zea mays (Hanson et a!.
!996a; White and Doebley 1999). The paucity of empirical data may stem from the theoretical difficulty of distinguishing non-coalescence or lineage sorting from introgression, as both processes are expected to result in similar patterns of allele sharing. Arguing for one or the other generally requires independent evidence, e.g. biosystematic evidence that hybridisation is occurring, or geographic evidence that allele sharing occurs only in regions of sympatry between putatively introgressing species
(e.g. Doyle et a!. 1999a).
Recombination and concerted evolution
An additional feature of nuclear genes that has implications for phylogenetic studies is the potential for recombination both at individual loci (thus generating additional allelic diversity), and more importantly, between paralogous genes.
Such recombination can take place either in vivo or in vitro
(i.e. PCR-mediated recombination).
Allelic recombination (including both crossing-over and gene conversion) results in alleles that are chimeric between parental alleles. This phenomenon violates the assumption of phylogenetic analysis that relationships among terminals are strictly bifurcating, rather than reticulate. If, however, alleles within species are monophyletic, this phenomenon will not affect reconstruction of species relationships from a gene tree, although it may introduce homoplasy into the phylogenetic analysis (Doyle 1995, 1997). Analytical tools are available for identifying putatively recombinant sequences and their potential parental alleles (Stephens
1985; Sawyer 1989; Jakobsen and Easteal 1996; Grassly and
Holmes 1997; Drouin et al. 1999). Thus, while allelic recombination has implications for phylogenetic analysis, it does not necessarily prevent accurate reconstruction of supra-specific phylogenies from gene sequence data.
Recombination among paralogous sequences
(non-homologous recombination), however, is of greater consequence. This phenomenon may take place only sporadically, or be a general feature ofa particular gene or gene family. One noted example is concerted evolution, which tends to homogenise rDNA sequences both within and among rDNA repeats (Zimmer et a/. 1980; Arnheim I 983; Baldwin et a/.
1995; Elder and Turner 1995). In the particular case of rDNA, which generally consists of thousands of repeats per locus, concerted evolution tends to homogenise repeats such that a single predominant allelic form exists. Concerted evolution appears to be a common feature of highly repetitive nuclear sequences. Low-copy nuclear genes are not free from concerted evolution, however, and examples exist of concerted evolution even among fairly small gene families
(e.g. rbcS: Clegg et a/. 1997).
The effect of concerted evolution (or any recombination among paralogous sequences) on phylogenetic analysis depends on its extent. As noted by Sanderson and Doyle
(1992), if concerted evolution among members of a gene family is absent, then complete sampling of all genes from all species will result in an orthology-paralogy tree (OP tree) in which sequences of orthologous loci are all more closely related to each other than to any paralogous sequence, and phylogenetic relationships among each ortholog may be expected to reflect organismal relatioships. If, at the other extreme, concerted evolution results in complete homogenisation of members of a gene family (as is often assumed in studies of rDNA), then sampling of any gene of a gene family will result in its correct phylogenetic

! 56 Australian Systematic Botany R. L. Small et a/. placement (assuming .concerted evolution occurs at a rate faster than speciation). If, however, concerted evolution occurs but is incomplete, then sampled genes may represent a mixture of orthologous and non-homogenised paralogous sequences. Accurate reconstruction of organismal phylogenies from such data are practically impossible
(Sanderson and Doyle 1992).
The probability of non-homologous recombination appears to be influenced by several factors. Principle among these may be the degree of sequence similarity among paralogs, and the genomic proximity of paralogs. As recombination is a similarity-driven process, paralogous sequences that are more closely related (recently diverged) are more likely to experience non-homologous recombination. This can create a circular pathway of recombinational events-after duplication, paralogous sequences diverge in sequence, but if they retain sufficient similarity, non-homologous recombination may result in homogenisation of the paralogs, which leads to greater sequence similarity, which can then promote inter-locus recombination. Genomic proximity may also play an important role in the . probability of non-homologous recombination. Because recombination occurs primarily between homologous chromosomes, paralogous loci that are closely linked on a chromosome may be more likely to undergo non-homologous recombination than paralogous loci that are located on separate chromosomes.
Finally, the methodological concern of PCR-mediated recombination must be addressed, as most gene sequences used · in molecular phylogenetic studies are generated via
PCR. PCR-mediated recombination is a well characterised phenomenon (Myerhans et a!.
1990; Bradley and Hillis
1997; Cronn et a!. 2002a) that results from either template-switching during PCR or from incompletely extended copies from one locus serving as a primer for subsequent extension from a paralogous locus. This is a notable problem for analyses of low-copy nuclear genes because studies often utilise general or universal PCR primers that amplify multiple loci, especially during preliminary stages of a study, prior to the development of locus-specific primers (Sang et a/. 1997b; Evans, Alice et a/.
2000; Small and Wendel 2000a; Cronn et a!. 2002a), Similar to in vivo non-homologous recombination, the propensity for
PCR-mediated recombination may depend on the degree of sequence similarity among para logs as well as PCR reaction conditions. Specifically, as noted by Cronn et al. (2002a), annealing temperature, amplicon length and extension time are factors that may be optimised to avoid PCR-mediated recombination.
Procedure to determine appropriate nuclear genes for phylogenetic analysis
When planning a phylogenetic study that includes nuclear genes, a generalised protocol would entail several sequential steps (Fig. 2). These include selecting candidate genes and representative taxa for a preliminary study, isolating candidate genes from representative taxa, assessing orthology among sequences isolated from representative taxa, assessing relative rates of sequence evolution in order to choose among potential loci and finally, generating sequences from the taxa of interest from the chosen loci. In an effort to facilitate a general application of these sequential experimental necessities, we will use as an example our previous investigations of the alcohol dehydrogenase (Adh) gene family in Gossypium (Small and Wendel 2000a).
Selection of candidate genes
One of the great advantages of using nuclear genes for phylogenetic analysis is the practically unlimited number of genes from which to choose (e.g. the Arabidopsis thaliana nuclear genome contains about 26000 genes: The
Arabidopsis Genome Initiative 2000). While there clearly are differences among genes and gene families in their potential phylogenetic utility, the huge number of possible genes from which to choose indicates that a multitude of genes exist that will be useful in any given plant lineage at any given phylogenetic level. It is important to note that there is no a priori reason to expect that any particular gene or gene family will be universally useful at any given phylogenetic depth because of the vagaries of the evolutionary dynamics of gene families. However, it is even more important to note that with a reasonable investment in characterisation, almost any gene will prove to be useful at some level. Thus, there is no reason to limit explorations to those genes that have previously been shown to be useful in other plant groups. To paraphrase and emphasise this point, there is nothing 'special' in the attributes of commonly used genes such as Adh, particularly in-as-much as relatively unexplored genes (Liu et a!. 2001; Mal comber 2002; Wendel et al. 2002) and even anonymous nuclear loci (Blake et a!.
1999; Cronn et a!. 2002b) have proven phylogenetically useful.
Given the diversity of genes from which to choose, where should one start? One logical place is with an assessment of genes that have been used by previous workers in taxa related to the group of interest. While the list continues to grow, at present a relatively small number of gene families have been widely investigated for their phylogenetic utility in plants
(Sang 2002). These include Adh (alcohol dehydrogenase: e.g. Gaut and Clegg 199 1 , 1993; Morton et a!. 1996; Sang et a/. 1 997b; Small et a!. 1 998; Ge et a/. 1 999; Small and
Wendel 2000a, 2000b), G3PDH (glyceraldehyde
3-phosphate dehydrogenase: e.g. Olsen and Schaal 1999;
Olsen 2002), GBSSI (granule-bound starch synthase: e.g.
Mason-Gamer et a/. 1998; Evans et a!. 2000; Mason-Gamer
200 1; Evans and Campbell 2002; Small 2003), MADS-box genes (e.g. pistil/ala, apetalal, apetala3, leafY: e.g. Bailey and Doyle 1999; Barrier et al. 1999; Bailey, Price et a!.

Use of nuclear genes in plant phylogeny
Australian Systematic Botany
I L
Select Exemplar Taxa " ''
"
' C'"dld"• " '
""
I
Generate Preliminary Sequence Data
+
•
1 PCR band?
'
Sequence PCR product
PCR amplification
(universal primers)
No polymorphism detected
+
Polymorphism detected
+
2+ PCR bands?
'
Clone PCR products
(ligation, transformation, pick colonies)
'
Screen multiple clones
(colony PCR, restriction digestion)
'
Sequence unique clones
'
Expression Data t
Assess Orthology
'
Sequence similarity t
I phylogeny
+
Southern Hybridization Comparative Mapping
'
Synonymous Rate (Ks) t
Assess Relative Rates t lntron Rate (KI)
+
Nonsynonymous Rate (Ka) t
Choose Appropriate Loci
'
Evidence of orthology
+
Appropriate evolutionary rate
+
Locate unique sites for primers t
Design and Optimize Locus-Specific Primers t
Optimize MgCI2 cone. &
+
Assess heterozygosity in direct sequences annealing temp. t
I
Generate Sequence Data From All Taxa
I
Fig. 2. Generalised protocol describing the steps necessary in foundational studies of the potential
· phylogenetic utility of nuclear genes.
! 57
2002), PHY (phytochrome: e.g. Mathews and Sharrock
1 996; Lavin et a!. 1 998; Simmons et a!. 200 I; Mathews et a!.
2002) and PGI (phospho-glucose isomerase: e.g. Gottlieb and Ford 1 996; Filatov and Charlesworth 1 999; Ford and
Gottlieb 1 999, 2002). Many of these genes have been investigated in a wide array of taxa, including most major angiosperm groups. In addition, however, many model plants, including most major crops, have associated with them extensive libraries of gene sequences for hundreds if not thousands of genes. This wealth of systematically useful information has been generated as a consequence of the many 'genome projects' being undertaken worldwide.
A first test of potential utility of a given gene or gene family, then, is whether or not it proved useful in other systems at a phylogenetic depth similar to that being attempted (e.g. interspecific, intergeneric, interfamilial). A second advantage of a literature survey is that papers will generally describe the methodology used to isolate specific

158 Australian Systematic Botany R. L. Small et a/. genes ahd include PCR-amplification primers. These may include general primers that were used for preliminary investigations, as well as locus-specific primers that were used to amplify single members of a gene family. At the time we began our investigations of Gossypium nuclear genes, the
Adh gene family was the most widely investigated nuclear-gene family in plants, making it a good candidate for further study. Additionally, previous isozyme studies in
Gossypium had shown that ADH protein products
(isozymes) were variable among Gossypium species.
Primary sources of candidate genes are the publicly available DNA-sequence databases (GenBank-United
States National Center for Biotechnology Information: http://www.ncbi.nlm.nih.gov/; EMBL-European
Molecular Biology Laboratory-European Bioinformatics
Institute: http://www.ebi.ac.uk/embllindex.html; DDJB­
DNA Data Bank of Japan: http://www.ddbj.nig.ac.jp/). DNA sequences deposited in any of these databases are ultimately shared among all three, thus searching through one database generally is sufficient, and because of our familiarity with
GenBank, it will be used as an example in the following discussion. Searching for candidate genes by using the DNA sequence databases can be conducted in several ways. First, taxonomic queries can be made at any level of the taxonomic hierarchy through the Taxonomy section of GenBank
(http://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome. htmll). Note here that GenBank uses a specific taxonomic hierarchical nomenclature for plants that follows recent phylogenetic studies (APGII 2003), and which must be followed for successful searches. Within angiosperms, groups are classified not only by traditional ranks (e.g. subclasses, orders, families), but also by rankless but well accepted groups (e.g. eudicotyledons, eurosid I). If searches for a particular genus or species do not result in any sequences of interest, the next higher level of the taxonomic hierarchy can be investigated.
Queries can be made by gene name in the search window of the GenBank homepage, if a particular gene family is of interest. Again, nomenclature is an issue that must be dealt with, given the variable names used for some genes. For example, searches for 'alcohol dehydrogenase AND
Magnoliophyta' result in matches not only for medium-chain alcohol dehydrogenase sequences that have been the primary focus of molecular systematic and molecular evolutionary studies, but also matches for short-chain alcohol dehydrogenases, cinnamyl alcohol dehydrogenases, unknown mRNA and genomic clones with some sequence similarity to alcohol dehydrogenase genes. Further, some genes may be deposited in GenBank as 'alcohol dehydrogenase', while others may be deposited as 'Adh.'
Multiple searches using different combinations of gene and taxon names may be necessary to identify candidate genes.
A final method of searching GenBank that can be especially useful is BLAST searching (Basic Local
Alignment Search Tool: http://www.ncbi.nlm.nih.gov/
BLAST/). This search tool compares an input sequence to all of the sequences in GenBank and identifies those sequences that have high sequence similarity to the input sequence. One of the advantages of BLAST searching is that results are retrieved without prior knowledge of gene name or taxonomic rank. If preliminary sequence data are available for some taxon of interest (e.g. from a genome project), then
BLAST searches can identify sequences in GenBank that are closely related to the sequence of interest, which can then provide useful information for primer design, identification of gene structure (exons and introns) and preliminary phylogenetic analyses. In addition to searches in the general
GenBank database, searches can be made through specific sequence sets such as ESTs (expressed sequence tags) and
STSs (sequence-tagged sites) that are not included in the general nucleotide database searches, as well as with specific whole genomes (e.g. Arabidopsis thaliana and 01yza sativa-the only plants with complete genome sequences).
Given the large number of genome sequencing, EST and
STS projects currently underway for various organisms (see
NCBI's Plant Genomes Central: http://www.ncbi.nlm.nih. gov/genomes/PLANTS/PlantList.html), these searches may be especially useful for identifying candidate genes.
Selection of representative taxa
When screening candidate genes for potential phylogenetic utility, one efficient approach is to select several taxa that represent the range of diversity that is ultimately to be included. This approach allows the preliminary study to be relatively contained in terms of the number of sequences that must be generated, while still providing enough data to make an informed decision with respect to phylogenetic utility.
Preliminary studies should include a minimum of two taxa, as this is required to understand relative levels of divergence.
Preferably, up to five taxa should be included to provide a more comprehensive overview of potential phylogenetic utility. The specific choice of taxa will depend on the amount of preliminary data from other sources (taxonomy, biosystematic studies, other phylogenetic hypotheses) that is available. For example, if a study is to investigate relationships among species within a genus, then species representing the major groups within that genus should be selected on the basis of, for example, taxonomic classification of species into subgenera or sections, cytogenetic data grouping species into genome groups, or previous phylogenetic hypotheses.
A second criterion that should be used when considering choice of taxa to use in pilot studies is ploidy level. Given the added layer of sequence complexity that accompanies polyploidy, it is prudent to select taxa at the lowest possible ploidy level for initial investigation. As ploidy level increases, gene family size for any given gene also increases, making it more challenging to confidently identify

Use of nuclear genes in plant phylogeny A ustralian Systematic Botany 1 59 orthologous sequences. Multiple homeologous loci
(orthologous loci donated by the progenitors of the allopolyploid) are expected to be found in allopolyploids, which must be distinguished from paralogous sequences.
Third, taxa for which significant background information is available are better choices for preliminary studies than relatively unknown taxa. Such background information may include ploidy level (see above), cytogenetic characterisation, previous knowledge of phylogenetic relationships, and/or previous molecular biological studies. All of this information may help place preliminary molecular phylogenetic data into a useful organismal and molecular framework, which may then guide further experimental choices.
Finally, representative taxa for which sufficient quantities of high-quality template DNA are available should be chosen. Preliminary studies may involve numerous PCR experiments to find optimal PCR conditions, and successful
PCR amplification is often dependent on DNA quality.
Additionally, if South em hybridisation experiments are to be performed (see below), large quantities (5-1 0 j.tg per digestion) of restriction enzyme-digestible DNA must be available.
For our studies of Adh in Gossypium we chose three representative taxa for our initial studies. There are so species of Gossypium distributed circumtropically with a well developed infrageneric classification system (Fryxell
1 968, 1 979, 1 992). There are three primary centers of diversity in Gossypium : Australia, Africa and the New
World. Additionally, previous cytogenetic studies in
Gossypium (reviewed in Endrizzi et a!. I 985) had identified eight different genome groups, and phylogenetic studies based on cpDNA restriction-site data (Wendel and Albert
1 992) had identified major clades. On the basis of this wealth of preliminary data we chose three species representative of the major groups: G. robinsonii (Australian C-genome),
G. herbaceum (African A-genome), and G. raimondii (New
World D-genome ). Although Gossypium includes both diploid and allotetraploid species, the representative taxa chosen were all diploids to simplifY estimates of gene copy number. It should be noted that while the amount of preliminary data available in Gossypium facilitated our selection of representative taxa, any one of the above sources of information (taxonomy, distribution, cytogenetics, previous phylogenetic hypothesis) alone would have provided sufficient information to make an informed decision on which taxa to include.
Isolation of candidate genes ji·om representative taxa
Once candidate genes and exemplar taxa have been chosen, the next step is to generate preliminary sequence data. This step generally is accomplished via PCR amplification using either general gene family primers or locus-specific primers.
The choice of approach is dictated by the amount of preliminary data available to the investigator.
If locus-specific primers are available (i.e. have been designed by other research groups), then PCR amplification and sequencing can be relatively straightforward. It is necessary, however, to remain mindful of complications that may arise from heterozygosity, particularly if alleles differ in length. If, on the other hand, little or nothing is known about a given candidate gene from the taxa of interest, an approach using general primers is necessary. This approach involves either development of general gene-family primers or use of primers developed by others. Such primers usually are designed by comparing gene sequences of available genes of the gene family (e.g. downloaded from GtmBank) and placing primers in regions of high sequence conservation
(e.g. highly conserved exon sequences). General primers can be designed to incorporate some polymorphism within the primer if necessary, and it is useful to be aware of codon structure of the exons where primers are to be placed. As the
3' end of the primer is the most important in terms of homology with the target, and especially the last 3' nucleotide, the 3' end of the primer should be placed at a second codon-position nucleotide, as these are likely to be the most highly conserved. General primers have been developed for a number of gene families, including
Adh-alcohol dehydrogenase (Sang et a!. I 997b; Gaut et a/.
I 999; Small and Wendel 2000a), GBSSI-granule-bound starch synthase I (Mason-Gamer et al. 1 998; Evans et al.
2000), GS-glutamine synthetase (Emshwiller and Doyle
I 999) and PHY -phytochrome (Mathews and Sharrock
I 996). In addition, the publication of Strand et a!. ( 1 997) gives primers for a number of different nuclear-gene families.
Once primers have been designed or obtained, initial PCR optimisation experiments should be performed. Because general primers have not been designed to be locus-specific, the number and size of PCR amplicons from a given taxon cannot be predicted a priori. The efficiency of amplification of various loci may be affected by a number of factors, including template DNA quality, efficiency of the PCR polymerase (Taq, or other available polymerases), primer annealing temperature, MgCI2 concentration and the use of other PCR additives. Relative to other loci amplified for molecular phylogenetic studies (e.g. cpDNA, rDNA), nuclear genes exist in much lower copy number and hence are more difficult to amplify. In many cases; significant experimental effort is required to optimise conditions.
Template DNA quality is important, especially when using general primers that may not match the template exactly. Optimal tissue for DNA extraction may vary across lineages. In our experience in Malvaceae, DNA extracted from fresh, young leaf material generally provides the best-quality DNA. Other researchers (personal communication from a reviewer) have found that DNA extracted from properly processed silica gel-dried material also provides high-quality DNA. In addition, in some groups

1 60 Australian Systematic Botany
3 . 0 mM
2 . 0 mM
4 6 4 8 5 0 5 2 5 4 5 6 5 8 60 62 64 4 6 4 8 5 0 52 5 4 5 6 5 8 60 62 6 4
2 . 6
2 . 0
1 . 5
1
•
2
1 . 0
0 . 5
R. L. Small
2 . 6
2 . 0
1 . 5
1 . 2
1 . 0
0 . 5 et a/.
Fig.
3. Gel photo demonstrating the effect ofMgC12 concentration and annealing temperature on the amplification of nuclear genes from genomic DNA. A single genomic DNA (Gossypium raimondii) was amplified with universal Adh primers across a range of annealing temperatures (46-64°C), with either 3.0 or 2.0 mM MgC12 final concentrations. As annealing temperature increases, the number of amplified products decreases. As the MgC12 concentration increases, the number of amplified products increases. very young leaves may contain PeR-inhibiting compounds, indicating that mature leaves may be more suitable tissue sources for DNA extraction (personal communication from a reviewer). While typical CTAB DNA isolation procedures
(e.g. Doyle and Doyle 1987, 1990) often result in amplifiable
DNA for high-copy templates, these DNAs may retain sufficient impurities to inhibit the more selective amplification of low-copy nuclear genes. In these cases, the use of DNA extraction kits (e.g. Plant DNeasy, Qiagen) that produce cleaner DNAs may be indicated. Template concentration in a PCR reaction may also be important, given the lower copy number of nuclear genes; low-concentration (<20 ng J.!L-1) DNAs that have been adequate for amplification of cpDNA or rDNA loci may be insufficient for amplification of nuclear genes.
The choice of amplification polymerase may also affect efficiency or even ability to PCR amplify some nuclear genes.
In our laboratories we have used a wide variety of polymerases from various suppliers and found significant variation in the ability of these enzymes to amplify low-copy nuclear sequences. This is, unfortunately, an example of 'you get what you pay for.' While some lower-cost polymerases may be sufficient for amplification of high-copy templates, they may be insufficient for amplification of low-copy sequences.
In addition to the specific polymerase used, other
PCR-reaction conditions will affect the ability to amplify low-copy templates. Specifically, annealing temperature,
MgC12 concentration and addition of PCR additives may need to be adjusted to optimise amplification conditions. The interplay between primer-annealing temperature and MgC12 concentration is especially important since increasing annealing temperature and decreasing MgC12 concentration results in more specific amplification. Because multiple
· paralogous copies of most genes are expected to be amplified with a general primer set, it may be desirable in some cases to amplify multiple members of a gene family for initial characterisation. At low annealing temperatures and high
MgC12 concentrations a large number of bands often are observed, many of which may turn out to be non-specific amplification products (i.e. not the target sequence).
Alternatively, at high annealing temperatures and low MgC12 concentrations, fewer to no bands may be observed because of the high stringency of the reaction conditions. A combined optimisation approach that evaluates a range of annealing temperatures (e.g. ± 5°C of the theoretical melting temperature of the primer pair) and MgC12 concentrations
(e.g. 1.5-3.0 mM final concentration) wiil provide an estimate of the combination of conditions under which a reasonable number of bands is amplified (Fig. 3). If a gradient thermal cycler is available, these optimisation experiments can be conducted in a single run by using a single DNA in one set of reactions, varying both the MgC12 concentration and the temperature gradient. Finally, a number of PCR additives have been suggested to improve amplification.
While we have not exhaustively screened these additives, our experience of amplifying low-copy loci from a variety of seed plants indicates that the addition of bovine serum albumin
(BSA; final concentration of 0.2 J.!g J.!L-1) . significantly improves amplification from problematic DNA.
If locus-specific PCR primers are used to amplify the genes of interest and a single homogenous PCR product is amplified, these products may be directly sequenced. More often, however, especially when universal primers are used, multiple PCR products that may vary in size are obtained. To isolate individual PCR products for sequencing, cloning of these PCR products into an appropriate plasmid is generally

Use of nuclear genes in plant phylogeny A ustralian Systematic Botany 1 6 1 required (e.g. Promega's pGEM-T). Alternatively, if multiple bands are widely spaced in size and can be sliced individually from a gel, these products can then be directly sequenced. An exception to these generalisations are studies of highly heterozygous outcrossing plants. For example, analysis of four low-copy loci in conifers (R. Cronn, unpubl.) has revealed that heterozygosity is nearly universal across loci and species and that length polymorphism is common within intron regions. In these more challenging cases, cloning may be necessary. One final alternative to cloning is the design of allele-specific PeR-amplification and sequencing primers (e.g. Rauscher et a!. 2002) if sufficient preliminary sequence data are available.
Once PCR products have been ligated into a plasmid and transformed into competent E. coli cells, the next decision that must be made is to choose how many transformants to screen. At this point in a preliminary study thoroughness is more important than speed, and although these screening steps are often laborious and time-consuming, they increase the probability of completely sampling the pool of PCR products. The number of colonies to pick depends on the initial starting PCR product. If only one or two unique PCR products are expected in a pool of transformants, then a smaller number of colonies can be screened, whereas if several unique PCR products are expected, a greater number of colonies must be screened to ensure identifying all products. As a starting point, we have found that picking 20 colonies is reasonable if a small number of PCR products is expected, and 40 or more if a larger number is expected.
Apparent transformants (i.e. white colonies in systems using blue-white selection · criteria with X-gal and IPTG) can quickly and easily be screened for inserts via PCR by using the following procedure.
Individual colonies are picked from plates with a pipet tip and suspended in I 0 j.!L of H20 in a numbered microcentrifuge tube. Once the colony is resuspended by pumping the pipettor several times, the pipet tip is applied to a fresh LB-agar-ampicillin plate that has a numbered grid which corresponds to the numbers on the microcentrifuge tubes. This results in (1) a suspension of bacterial colonies in the microfuge tube and (2) a grid plate with corresponding bacterial colonies that (after overnight culture) provide cells for plasmid minipreps. The suspended cells are then boiled for 1 0 min to lyse cells and release plasmid into the supernatant. After centrifugation pellet the cell debris, I
( I min, maximum speed) to j.!L of this suspension can then be used as a template in a small-volume (e.g. 10 j.!L) PCR reaction using either the gene-specific or vectorcspecific primers to screen the colonies for presence of an appropriate insert. Test gels may be run using high-percentage agarose gels (e.g. 2% or higher) to help distinguish among PCR products of slightly different lengths.
A test gel of the colony PCR results will indicate (1) how many of the apparent transformants contain the PCR products of interest and (2) the relative sizes of the inserts. If different-sized PCR products were ligated and transformed in the same reaction, this screen will provide quick identification of which colonies contain different-length
PCR products. Because a single-sized PCR product may contain amplicons from more than one gene or multiple alleles of a single locus, a secondary screening using restriction-enzyme digestion (typically frequently cutting enzymes with 4-bp recognition sequences) can be conducted to differentiate among plasmids with identically sized inserts. Ideally, this two-step screening procedure will result in the identification of unique sets of plasmids: a set for each size class, and subsets of plasmids with the same size, but different restriction digestion profiles. Once these sets have been identified, one to several plasmids from each set should be screened by sequencing. Although the foregoing pre-screening procedures may be skipped, the only way to be sure that all variants present are sampled is to sequence all of the colonies.
Template DNA for sequencing from individual colonies may be obtained either by PCR amplification from the boiled colonies or by miniprep isolation of the plasmid DNA either by using published protocols (Ausubel ef al. 1992;
Sambrook et a!. 1989) or commercially available kits. We suggest that initial sequencing efforts utilise the plasmid-specific primers that generally flank the cloning site of the plasmid (e.g. in Promega's pGEM-T plasmid primer sites for the T7 and SP6 promoters are available about 50 bp on either side of the cloning site) as this allows sequencing reads through the PCR-primer sites and into the PCR product itself. Depending on the size of the PCR product, internal sequencing primers may be necessary to fully sequence a given plasmid. Such primers usually are easily designed to work across a gene family by placing them in conserved regions of exons (see below), and if necessary, incorporating some ambiguity into the primer.
Preliminary characterisation of genes isolated ji·om representative taxa
Gene structure tends to be fairly conserved across genes of a given gene family. For example, plant Adh genes generally have a 10 exon/9 intron structure. There are examples of introns that have been lost from some lineages-for example, the Arabidopsis thaliana Adh gene (Chang and Meyerowitz
1986), Gossypium AdhA (Small and Wendel 2000a )-but most genes retain the 10 exon/9 intron structure, and the intron losses are easily discerned through sequence alignment. Identification of exon and intron regions can be accomplished through alignment of sequences obtained from the taxa of interest, with previously published sequences downloaded from GenBank. Gene sequences deposited in
GenBank usually have the exon/intron structure designated, and individual exon sequences can be isolated ·and aligned individually with the preliminary data. In addition, the

162 Australian Systematic Botany
A Coding Sequcn.ce
R. L. Small et a/.
UTR
B eDNA individual exons genomic DNA
• • • CT AAG- - - - - ----------------------------------------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -GGTCAAAA • • •
• • •
CCAAG
•
•
•
GGGCAGAC
• . •
Gl'AA@\'r'I'ACl'Gl'TCGA'l'A'l'AAl'GAA'l'A'rTACAACATTTATTTTCTGCAAGl'A1'1'Gl'CAC'l'GT'l'T1'CO"f'fAAC1'1'GGC!ii§J;G'l'TACAC . , , exon 2 exon 3
Fig. 4. Representation of the relationship among gene sequences derived from eDNA clones, individual exon sequences and genomic
DNA. (A) Diagrammatic representation of an alignment of eDNA, exon and genomic DNA sequences from a hypothetical gene with five exons (shaded and numbered boxes) and four introns (lower-case letters). (B) Alignment of a portion of Adh eDNA (Solanum tuberosum
Adhl, GenBank M25 1 54), individual exons (Zea mays Adhl, GenBank AF 123535) and genomic DNA (Gossypium raimondii AdhA,
GenBank AF 1 82 1 1 6). Note the conservation of the intron boundary dinucleotides (5' 'GT' and 3' 'AG'). presence of the highly conserved 5' 'GT' and 3' 'AG' intron-boundary dinucleotides can be used to confirm the start and end points of an intron. Thus, through sequential alignment of individual exons and confirmation of intron-boundary sequences, the exon/intron structure of preliminary sequence data can be defined (Fig. 4). Once these steps are accomplished for the exemplar taxa, the next logical step is to assess copy number and orthology of the isolated sequences.
Copy number and orthology assessment
As noted earlier in this review, several criteria can be used as evidence of orthology. In practice, the two most useful and efficient approaches are phylogenetic analysis and Southern hybridisation analysis, the former to infer orthologous relationships among sequences, and the latter to confirm that the inference of orthology is not confounded by the presence of multiple, closely related paralogs.
Once sequence data have been collected for representative taxa, sequences should be trimmed to include only exons and aligned with genes from related plants ( exons only because introns are generally unalignable outside of closely related species). Phylogenetic analyses of these data produce a hypothesis of relationships among genes, and inferred clades may represent orthologous sequence groups.
For example, in our studies of Gossypium Adh we initially isolated several Adh sequence types from three representative species. Putative exons were inferred by comparison with exons of a Solanum Adh sequence from
GenBank. A large dataset of angiosperm Adh exon sequences from GenBank was then compiled and subjected to phylogenetic analysis. The resulting phylogenetic tree
(Fig. 5) grouped the Gossypium Adh sequences into five primary clades that included sequences from the representative species. These clades were inferred to represent orthologous genes, which consequently were named AdhA-AdhE (Fig. 5). The inference of orthology was based on the fact that for each of the named genes, sequences of each type were isolated from each of the representative species, and in each case these sequences formed monophyletic groups. Note that the Gossypium Adh genes are found in two disparate parts of the angiosperm Adh tree:
AdhA, AdhB and
AdhC in one clade, and
AdhD and
AdhE in a separate clade. These data indicate an ancient gene duplication giving rise to these two primary clades, with more recent gene duplications giving rise to the genes within each clade.
Once these preliminary phylogenetic analyses had been performed, examination of the entire sequences ( exons + introns) further supported the inference of orthology of the genes. Intron sequences were easily alignable within putative orthologs, but generally unalignable between orthologs.
Furthermore, the absence of two introns in the
AdhA sequences isolated from all representative species, but present in most other angiosperm Adh sequences supported the inference of orthology among these sequences. Thus, all available phylogenetic and gene structure data corroborated the orthology of the named Adh genes.

Use of nuclear genes in plant phylogeny
A ustralian Systematic Botany 1 63
99 ..-----r--------------1
1 00
Fragan'a
Pyrus
Malus
Arabls
Arabldops/s
Leavenworth/a
Leavenworth/a
2
Leavenworth/a 3
1
75
1 00
1 00
77
99
1 00
Fig. 5. Strict consensus of seven equally parsimonious trees resulting from maximum parsimony analysis of Adh ex on sequences from a wide range of angiosperms (rooted with the gymnosperm Pinus), including representative Gossyplum sequences. Bootstrap values ( 1 000 replicates) greater than 50% are shown above each branch. For each Gossypium locus (AdhA-AdhE), sequences were obtained from each of three representative diploid taxa denoted by their genome affiliations (A = G. herbaceum or G. arboreum, C = G. robinson it, D = G. raimondii). Each inferred Gossypium Adh locus includes all three representative taxa, is monophyletic, and is supported by a 1 00% bootstrap value.
To ensure that the named genes were unique, and that closely related paralogous genes did not exist, we then conducted Southern hybridisation analysis. Because introns are unalignable between genes and exons are fairly divergent
(generally 20---30%, except for AdhD!AdhE which are only about 10% divergent: Small and Wendel 2000a) we designed hybridisation probes that included both intron and exon sequence. Probes were small (about 500 bp) and included the majority of intron 3 + exon 4 of the Adh genes. Individual probes were obtained by PCR for each of the Adh genes and used in high-stringency Southern hybridisations (see Small and Wendel 2000a for detailed hybridisation conditions).

164 A ustralian Systematic Botany R. L. Small et a/.
The results of these experiments in some cases confirmed the uniqueness of putative orthologs, and in other cases refuted those inferences. For example, for most
Gossypium
Adh genes, a single hybridising band was seen for the diploid species and two hybridising bands in the tetraploid, as expected. For AdhB, however, multiple hybridising bands were observed in all species indicative of additional unsampled AdhB-like genes in the Gossypium genome.
Subsequent publication of
Adh sequences from
Gossypium genomic and eDNA libraries (Millar and Dennis 1996) corroborated this inference. Phylogenetic analysis of the sequences published by Millar and Dennis (1996) revealed that they were closely related to the AdhB sequences isolated via PCR in our study. Orthology-paralogy relationships among these AdhBand AdhB-like genes were not clear on the basis of either phylogenetic analysis or sequence alignment; thus, it was concluded that these particular genes would be poor choices for further phylogenetic studies.
In sum, the combination of phylogenetic and Southern hybridisation analyses allowed rigorous inference of orthology (or lack thereof) among specific sequence types isolated from the representative Gossypium species surveyed in our preliminary studies. This in turn allowed us to choose from among the isolated sequences those that were potentially phylogenetically useful.
Choosing among potential sequences
Once orthologous and potentially useful genes have been identified, evaluation of rates of sequence evolution can provide insight into the relative phylogenetic utility of various candidate genes. For example, in a recent study of
Gossypium phylogeny (Cronn et al. 2002b), 11 low-copy nuclear encoded sequences were evaluated, and the percentages of variable and phylogenetically informative sites varied over 3.5-fold and 7.9-fold ranges, respectively.
Percentage of variable sites ranged from 3.43 to 12.1 %, while percentage of phylogenetically informative sites ranged from 0.21 to 1.66%.
Evaluation of relative rates, and thus potential phylogenetic utility, can be conducted several ways. First and perhaps simplest, phylogenetic analysis of sequences obtained from representative taxa can be performed and relative branch lengths can be assessed. Those genes with longest branch lengths are the most variable and thus may provide the most information for a larger set of taxa.
On a finer scale, it is useful to evaluate variation on a per site basis, thus allowing inferences of the amount of information obtained per unit of sequence. Clearly, longer sequences are, in general, more likely to contain more variable sites. However, if efficiency in sequencing effort is a desirable goal, then those sequences that have the greatest variation per site may be more useful even if they are shorter than some other sequences that provide greater overall numbers of variable sites. This value can be estimated by using a variety of genetic-distance algorithms available in most phylogenetic inference packages (e.g. PAUP*:
Swofford 2002). Calculation of p-distances (proportion of variable sites) provides a straightforward estimate of relative variation. More complex models (e.g. Jukes-Cantor, Kimura
2-parameter) can be invoked if deviations from a simple model of evolution are observed (e.g. transition : transversion bias, GC-content bias).
Data collectionfi'Om all taxa of interest
Once a gene or set of genes has been identified as potentially useful in representative taxa, the next step is to generate sequence data for all taxa of interest. In our experience, the most efficient approach to accomplish this is to develop locus-specific PCR-amplification primers for each particular gene of interest. There are several advantages to this approach. First, locus-specific amplification may minimise the costly, tedious and time-consuming step of having to clone PCR products from a large number of taxa, although this may be necessary in any event if heterozygosity is high. Second, the more specific the PCR primers are, the less will be the nagging problem of chimeric, recombinant
PCR products that occasionally (to often) are produced during PCR amplification of genes (as discussed earlier in this review). Third, the ability to reliably and reproducibly amplify a single homogenous PCR product from multiple species bolsters the inference of orthology of those sequences.
Conclusions
The paths chosen by modern plant systematists increasingly lead to questions that are difficult to resolve through the application of one source, or even sometimes two sources, of molecular inference. There has been enormous growth over the past decade of publicly available gene-sequence databases including whole-genome sequences for model organisms such as Arabidopsis thaliana (The Arabidopsis
Genome Initiative 2000) and 01yza sativa (Goff et a!. 2002;
Yu et a!. 2002) and EST databases from many other species (see Plants Genome Central at NCBI: http://www.ncbi.nlm.nih.gov/gen,omes/PLANTS/PlantList.h tml). These accumulating data have set the stage for the development of new, low-copy gene markers that are up to the challenge of resolving difficult phylogenetic questions.
Low-copy nuclear markers come at a higher cost with regard to development than either cpDNA or rDNA markers, but they are certain to eclipse cpDNA and rDNA with regards to reliability and resolving power. Since single-copy nuclear genes are biparentally inherited, they have the potential to reveal the ancestry of all lineages contributing to an extant species, whether diploid (Cronn et a!. 2003) or polyploid
(Small et a!.
1998; Liu et a!.
2001; Cronn et a!.
2002b).
Low-copy loci are rarely subject to concerted evolution
(Cronn et al. 1999; Senchina et al. 2003), and they do not

Use of nuclear genes in plant phylogeny Australian Systematic Botany 165 display the high intragenic polymorphism (and corresponding increase in homoplasy) characteristic of rDNA, which requires concerted evolution to maintain intra­ and inter-array homogeneity. Finally, nuclear genes contain exonic regions that limit alignment ambiguity and facilitate homologous comparisons (Bailey and Doyle 1 999; Doyle et al. 1 999b; Bortiri et al. 2002; Sang 2002), as well as introns and other non-coding regions that diverge at a substantially higher rate than either cpDNA or rDNA.
To the extent that plant molecular systematic studies would benefit from the routine application of one nuclear gene, the development and application of multiple independent nuclear loci holds even greater promise for addressing more complex questions. For example, the application of multiple, low-copy nuclear markers may present the only viable approach for teasing apart temporally compressed divergence events, since the variance of divergence rates across sites from exonic and intronic regions almost ensures that a fraction of sites will yield sufficient phylogenetic signals to mark lineages (Small et al.
1 998; Cronn et al. 2002b; Mal comber 2002). Similarly, the topological 'stalemates' frequently presented by cpDNA and rDNA can only be confidently resolved through the use of multiple independent markers, such as low-copy nuclear genes (Small et a/. 1 998; Cronn et al. 2002b, 2003;
Mal comber
2002).
In many cases, the resolution provided by nuclear genes will help to shed light on the true topology, as well as the underlying methodological or biological basis of phylogenetic incongruence. Less frequently, attempts to resolve cpDNA-rDNA incongruence using multiple nuclear loci may tum up exciting instances where multiple nuclear markers return a consistent topology that stands in stark contrast to the topologies returned by both cpDNA and rDNA (Cronn et a/. 2003). In these challenging cases, the single accomplishment of resolving a phylogeny pales in significance to the greater challenges of understanding the evolutionary dynamics that have given rise to the genomic complexity embodied in living plant species.
Acknowledgments
We thank the many colleagues who have contributed to our thoughts on these issues, Jeff Doyle and two anonymous reviewers whose comments improved this manuscript, and acknowledge funding support from the United States
National Science Foundation.
References
Adams KL, Cronn R, Percifield R, Wendel JF (2003) Genes duplicated by polyploidy show unequal contributions to the transcriptome and organ-specific reciprocal silencing. Proceedings of the National
Academy of Sciences of the United States of America 100,
4649-4654. doi: l 0. 1 073/PNAS.06306 1 8 1 00
Ahlandsberg S, Sun C, Jansson C (2002) An intronic element directs endosperm-specific expression ofthe sbe/lb gene during barley seed development. Genetics and Genomics 20, 864-868.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1 990) Basic local alignment search tool. Journal of Molecular Biology 215, 403-4 1 0. doi: 10. 1 006/JMBI. l 990.9999
Alvarez I, Wendel JF (2003) Ribosomal ITS sequences and plant phylogenetic inference. Molecular Phylogenetics and Evolution 29,
41 7-434. doi: I 0. 1 0 1 6/S 1 055-7903(03)00208-2
Anderberg AA, Rydin C, Kallersjo M (2002) Phylogenetic relationships in the order Ericales s.l.: analyses of molecular data from five genes from the plastid and mitochondrial genomes. American Journal of
Botany 89, 677-687.
APGII (2003) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG
II. Botanical Journal of the Linnean Society 141, 399-436.
Arnheim N (1 983) Concerted evolution of multigene families. In
' Evolution of genes and proteins'. (Eds M Nei, RK Koehn) pp.
38-6 1 . (Sinauer: Sunderland, MA)
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA,
Struhl K (Eds) ( 1 992) 'Short protocols in molecular biology.' 2nd edn. (John Wiley & Sons: New York)
Awadalla P, Eyre-Walker A, Maynard Smith J (1999) Linkage disequilibrium and recombination in hominid mitochondrial DNA.
Science 286, 2524-2525. doi: I O. l 126/SCIENCE.286.5449.2524
Bailey CD, Carr TG, Harris SA, Hughes CE (2003) Characterization of angiosperm nrDNA polymorphism, paralogy, and pseudogenes.
Molecular Phylogenetics and Evolution doi: JO. l 0 1 6/J.YMPEV.2003.08.02 1
29, 435-455.
Bailey CD, Doyle JJ ( 1 999) Potential phylogenetic utility of the low-copy nuclear gene pistil/ala in dicotyledonous plants: comparison to nrDNA ITS and trnL intron in Sphaerocardamum and other
Brassicaceae. Molecular Phylogenetics and Evolution 13, 20-30. doi: I 0. 1 006/MPEV. l 999.0627
Bailey CD, Price RA, Doyle JJ (2002) Systematics of the Halimolobine
Brassicaceae: evidence from three loci and morphology. Systematic
Botany 21, 3 1 8-332.
Baldwin BG, Markos S (1 998) Phylogenetic utility of the external transcribed spacer (ETS) of l 8 S-26S nrDNA: congruence of ETS and ITS trees of Calycadenia (Compositae ). Molecular
Phylogenetics and Evolution 10, 449-463. doi: l 0. 1 006/MPEV. l 998.0545
Baldwin BG, Sanderson MJ, Porter JM, Wojciechowski MF, Campbell
CS, Donoghue MJ (1 995) The ITS region ofnuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82, 247-277.
Bancroft I (2001) Duplicate and diverge: the evolution of plant genome microstructure. Trends in Genetics 17, 89-93. doi: l 0. 1 0 1 6/SOJ 68-9525(00)02 1 79-X
Barrier M, Baldwin BG, Robichaux RH, Purugganan MD ( 1 999)
Interspecific hybrid ancestry of a plant adaptive radiation: allopolyploidy of the Hawaiian silversword alliance· inferred from duplicated floral homeotic genes. Molecular Biology and Evolution
1 6, 1 105-1 1 1 3.
Birky CW (1 995) Uniparental inheritance of mitochondrial and chloroplast genes: mechanisms and evolution. Proceedings of the
National Academy of Sciences of the United States of America 92,
1 1 33 1-1 1 338.
Blake NK, Lehfeldt BR, Lavin M, Talbert LE (1 999) Phylogenetic reconstruction based on low copy DNA sequence data in an allopolyploid: the B genome of wheat. Genome 42, 3 5 1 -360. doi: 10. 1 139/GEN-42-2-35 1
Bortiri E, Oh S-H, Gao F-Y, Potter D (2002) The phylogenetic utility of nucleotide sequences of sorbitol 6-phosphate dehydrogenase in
Prunus (Rosaceae). American Joumal of Botany 89, 1 697- 1 708.
Bradley RD, Hillis DM (1 997) Recombinant DNA sequences generated by PCR amplification. Molecular Biology and Evolution 1 4,
592-593.

166 Austmlian Systematic Botany R. L. Small et a/.
Buckler ES, Holtsford TP (1 996) Zea ribosomal repeat evolution and substitution patterns, Molecular Biology and Evolution 13,
623-632.
Buckler ES, Ippolito A, Holtsford TP ( 1 997) The evolution of ribosomal DNA: divergent paralogues and phylogenetic implications, Genetics 145, 82 1-832.
Chang C, Meyerowitz EM (1 986) Molecular cloning and DNA sequence of the Arabidopsis thaliana alcohol dehydrogenase gene.
Proceedings of the National Academy of Sciences of the United
States of Ainerica 8 3, 1408- 1 4 1 2.
Charlesworth D, Awadalla P ( 1998) Flowering plant self-incompatibility: the molecular population genetics of Bras sica
S-loci. Heredity 81 , 1-9. doi: 1 0. l 038/SJ.HDY.6884000
Charlesworth D, Charlesworth B ( 1 998) Sequence variation: looking for effects of genetic linkage. Current Biology 8, R658-R66 l ,
Charlesworth D, Liu F, Zhang L ( 1 998) The evolution of the alcohol dehydrogenase gene family in plants of the genus Leavenworthia
(Brassicaceae): loss of introns, and an intronless gene, Molecular
Biology and Evolution 15, 552-559.
Chase MW, Knapp S, Cox AV, Clarkson JJ, Butsko Y, Joseph J,
Savolainen V, Parokonny AS (2003) Molecular systematics, GISH and the orfgin of hybrid taxa in Nicotiana (Solanaceae). Annals of
Botany 92, 1 07-1 27. doi : l 0. 1 093/AOB/MCG087
Chase MW, Soltis DE, Olmstead, RG, Morgan D, Les DH ( 1 993)
Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL, Annals of the Missouri Botanical
Garden 80, 528-580,
Clegg MT, Cummings MP, Durbin ML ( 1 997) The evolution of plant nuclear genes. Proceedings of the National Academy of Sciences of the United States of America 94, 779 1-7798. doi: 10. l073/PNAS.94. 1 5.779 l
Corriveau JL, Coleman AW ( 1 988) Rapid screening method to detect potential biparental inheritance of plastid DNA and results for over
200 angiosperm species. American Jouma/ of Botany 75,
1443-1458.
Crawford DJ ( 1 985) Electrophoretic data and plant speciation.
Systematic Botany 10, 405-4 1 6.
Crawford DJ (2000) Plant macromolecular systematics in the past 50 years: one view. Taxon 49, 479-50 l .
Cronn R , Wendel JF ( 1 998) Simple methods for isolating homoeologous loci from allopolyploid genomes. Genome 41,
756-762. doi: 10. l 1 39/GEN-4 1 -6-756
Cronn RC, Zhao X, Paterson AH, Wendel JF ( 1 996) Polymorphism and concerted evolution in a tandemly repeated gene family: 5S ribosomal DNA in diploid and allopolyploid cottons. Journal of
Molecular 'Evolution 42, 685-705.
Cronn RC, Small RL, Wendel JF ( 1999) Duplicated genes evolve independently after polyploid formation in cotton. Proceedings of
· the National Academy of Sciences of the United States of America
96, 14406-1441 [ ,
Cronn R , Cedroni M, Haselkorn T, Grover C , Wendel JF (2002a)
PCR-mediated recombination in amplification products derived from polyploid cotton. Theoretical and Applied Genetics 104,
482-489. doi: 10. 1 007/S00 1 22010074l
Cronn RC, Small RL, Haselkorn T, Wendel JF (2002b) Rapid diversification of the cotton genus (Gossypium: Malvaceae) revealed by analysis of sixteen nuclear and chloroplast genes.
American Journal of Botany 89, 707-725. doi: 10. l 093/ AOB/MCF l 25
Cronn RC, Small RL, Haselkorn T, Wendel JF (2003) Cryptic repeated genomic recombination during speciation in Gossypium gossypioides. Evolution 57, 2475-2489.
Doyle JJ ( 1 99 1 ) Evolution of higher-plant glutamine synthetase genes: tissue specificity as a criterion for predicting orthology. Molecular
Biology and Evolution 8, 366-377.
Doyle JJ (1992) Gene trees and species trees; molecular systematics as one-character taxonomy. Systematic Botany 17, 1 44--1 63.
Doyle JJ (1995) The irrelevance of allele tree topologies for species delimitation, and a nontopological alternative. Systematic Botany
20, 574--588.
Doyle JJ ( 1997) Trees within trees: genes and species, molecules and morphology. Systematic Biology 46, 537-553.
Doyle JJ, Doyle JL (1 987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19, l l - 1 5.
Doyle JJ, Doyle JL ( 1 990) Isolation of plant DNA from fresh tissue.
Focus 12, 1 3-l 5.
Doyle JJ, Doyle JL (1999) Nuclear protein-coding genes in phylogeny reconstruction and homology assessment: some examples from
Leguminosae, In ' Molecular systematics and plant evolution'. (Ed.
RJ Gornall) pp. 229-254. (Taylor and Francis: London)
Doyle JJ, Kanazin V, Shoemaker RC ( 1 996) Phylogenetic utility of histone H3 intron sequences in the perennial relatives of soybean
(Glycine: Leguminosae). Molecular Phylogenetics and Evolution 6,
438-447. doi: l 0. 1006/MPEV. l 996.0092
Doyle JJ, Doyle JL, Brown AHD ( l999a) Incongruence in the diploid
B-genome species complex of Glycine (Leguminosae) revisited: histone H3-D alleles versus chloroplast haplotypes. Molecular
Biology and Evolution 16, 354--362,
Doyle JJ, Doyle JL, Brown AHD ( 1999b) Origins, colonization, and lineage recombination in a widespread perennial soybean polyploid complex. Proceedings of the National Academy of Sciences of the
United States of America 96, 1 074 1-1 0745. doi: 10. 1 073/PNAS.96. 19.10741
Doyle JJ, Doyle JL, Brown AHD, Pfeil BE (2000) Confirmation of shared and divergent genomes in the Glycine tabacina polyploid complex (Leguminosae) using histone H3-D sequences. Systematic
Botany 25, 437-448,
Doyle JJ, Doyle JL, Brown AHD, Palmer RG (2002) Genomes, multiple origins, and lineage recombination in the Glycine tomentel/a
(Leguminosae) polyploid complex: histone H3-D gene sequences.
Evolution 56, 1 388-1402.
Drouin G, Prat F, Ell M, Clarke GOP ( 1 999) Detecting and characterizing gene conversions between multigene family members. Molecular Biology and Evolution 16, 1 3 69-1 390.
Elder JF, Turner BJ ( 1 995) Concerted evolution of repetitive DNA sequences in eukaryotes. The Quarterly Review of Biology 70,
297-320. doi: 1 0. 1086/4 19073
Emshwiller E, Doyle JJ (1999) Chloroplast-expressed glutamine synthetase (nepOS): potential utility for phylogenetic studies with an example from Oxalis (Oxalidaceae). Molecular Phy/ogenetics and Evolution 12, 3 10-3 19. doi; 10.1 006/MPEV. 1999.06 1 3
Emshwiller E , Doyle JJ (2002) Origins of domestication and polyploidy in Oca ( Oxalis tuberosa: Oxalidaceae ). 2. Chloroplast-expressed glutamine synthetase data. American Jouma/ of Botany 89,
1 042-1056,
Endrizzi JD, Turcotte EL, Kohel RJ ( 1 985) Genetics, cytology, and evolution of Gossypium. Advances in Genetics 2 3, 27 1-375,
Evans RC, Campbell CS (2002) The origin of the apple subfamily
(Maloideae; Rosaceae) is clarified by DNA sequence data from duplicated GBSSI genes. American Joumal of Botany 89,
1 478-1484.
Evans RC, Alice LA, Campbell CS, Kellogg EA, Dickinson TA (2000)
The granule-bound starch synthase (GBSSI) gene in the Rosaceae:
Multiple loci and phylogenetic utility. Molecular Phy/ogenetics and
Evolution 17, 3 88-400, doi: 1 0. l006/MPEV.2000.0828
Eyre-Walker A (2000) Do mitochondria recombine in humans?
Philosophical Transactions of the Royal Society of London. Series
B, Biological Sciences 355, 1 573-1 580. doi: l 0.1 098/RSTB.2000,07 1 8

Use of nuclear genes in plant phylogeny Australian Systematic Botany 167
Eyre-Walker A, Gaul RL, Hilton H, Feldman DL, Gaul BS (1998)
Investigation of the bottleneck leading to domestication of maize,
Proceedings of the National Academy of Sciences of the United
States of America 95, 444 1---4446. doi: 10.1 073/PNAS.95.8.444 1
Ferguson D, Sang T (200 I) Speciation through homoploid hybridization between allotetraploids in peonies (Paeonia),
Proceedings of the National A cademy of Sciences of the United
States of America 98, 391 5-39 1 9. doi: I 0, I 073/PNAS.06 1 288698
Filatov DA, Charlesworth D (1 999) DNA polymorphism, haplotype structure and balancing selection in the Leavenworth/a PgiC locus.
Genetics 153, 1423-1434.
Ford VS, Gottlieb LD ( 1 999) Molecular characterization of PgiC in a tetraploid plant and its diploid relatives. Evolution 53, 1 060-1 067.
Ford VS, Gottlieb LD (2002) Single mutations silence PGIC2 genes in two very recent allotetraploid species of Clarkia. Evolution 56,
699-707.
Freudenstein JV, Chase MW (200 I) Analysis of mitochondrial nad Jb-c intron sequences in Orchidaceae: utility and coding of length-change characters. Systematic Botany 26, 643-657.
Fryxell PA ( 1 968) A redfinition of the tribe Gossypieae. Botanical
Gazette 129, 296-308. doi: 10. 1 086/336448
Fryxell PA ( 1 979) 'The natural history of the cotton tribe.' (Texas A&M
University Press: College Station)
Fryxell PA ( 1 992) A revised taxonomic interpretation of Gossypium L.
(Malvaceae). Rheedea 2, 1 08-1 65.
Fu H, Dooner HK (2002) Intraspecific violation of genetic co linearity and its implications in maize. Proceedings of the National Academy of Sciences of the United States of A merica 99, 9573-9578.
Galloway GL, lv!almberg RL, Price RA ( 1 998) Phylogenetic utility of the nuclear gene arginine decarboxylase: an example from
Brassicaceae. Molecular Biology and Evolution 15, 1 3 1 2-1 320.
Gaul BS (1 998) Molecular clocks and nucleotide substitution rates in higher plants. Evolutionary Biology 30, 93-120.
Gaut BS (200 I) Patterns of chromosomal duplication in maize and their implications for comparative maps of the grasses. Genome
Research 1 1 , 55-66. doi: I O. I I O I/GR. l 60601
Gaut B S, Clegg MT (1991) Molecular evolution of alcohol dehydrogenase I in members of the grass family. Proceedings oft he
National Academy of Sciences of the United States of America
2060-2064.
88,
Gaut BS, Clegg MT (1 993) Molecular evolution of the Adhl locus in the genus Zea. Proceedings of the National A cademy of Sciences of the United States of America 90, 5095-5099,
Gaut BS, Doebley JF ( 1 997) DNA sequence evidence for the segmental allotetraploid origin of maize. Proceedings of the National Academy of Sciences of the United States of America doi: I 0, I 073/PNAS.94, 13.6809
94, 6809-6 8 1 4.
Gaut BS, Morton BR, McCaig BC, Clegg MT ( 1996) Substitution rate comparisons between grasses and palms: synonymous rate differences at the nuclear gene Adh parallel rate differences at the plastid gene rbcL. Proceedings ofthe National Academy of Sciences ofthe United States of America 93, 1 0274--1 0279. doi: 1 0, 1 073/PNAS.93. 19. 1 0274
Gaut BS, Peek AS, Morton BR, Clegg MT ( 1999) Patterns of genetic diversification within the Adh gene family in the grasses (Poaceae).
Molecular Biology and Evolution 16, I 086-l 097,
Ge S, Sang T, Lu B-R, Hong D-Y (1 999) Phylogeny of rice genomes with emphasis on origins of allotetraploid species. Proceedings of the National A cademy of Sciences of the United States of America
96, 14400-14405. doi: I 0, I 073/PNAS.96.25. 1 4400
Gielly L, Yuan Y-M, Kupfer P, Taberlet P (1 996) Phylogenetic use of noncoding regions in the genus Gentiana L.: chloroplast tmL
(UAA) intron versus nuclear ribosomal internal transcribed spacer sequences. Molecular Phylogenetics and Evolution 5, 460---466, doi: I 0, I 006/MPEV. I 996.0042
Goff SA, Ricke D, Lan T-H, Presting G, Wang R, et a/. (2002) A draft sequence of the rice genome (01yza sativa L, ssp. japonica).
Science 296, 92-100. doi: I 0, 1 126/SCJENCE. I 068275
Gottlieb LD (1977) Electrophoretic evidence and plant systematics.
Annals of the Missouri Botanical Garden 64, 1 61-1 80.
Gottlieb LD, Ford VS ( 1996) Phylogenetic relationships among the sections of Clarkia (Onagraceae) inferred from the nucleotide sequences of PgiC. Systematic Botany 21, 45-62.
Grant D, Cregan P, Shoemaker RC (2000) Genome organization in dicots: genome duplication in Arabidopsis and synteny between soybean and Arabidopsis. Proceedings of the National A cademy of
Sciences of the United States of America doi: 1 0. 1073/PNAS.070430597
97, 4 1 68---4'! 73,
Grassly NC, Holmes EC (1 997) A likelihood method for the detection of selection and recombination using sequence data. Molecular
Biology and Evolution 14, 239-247.
Hamby RK, Zimmer EA ( 1 988) Ribosomal RNA sequences for inferring phylogeny within the grass family (Poaceae). Plant
Systematics and Evolution 160, 29-37.
Hamby RK, Zimmer EA ( 1992) Ribosomal RNA as a phylogenetic tool in plant systematics. In ' Molecular systematics of plants'. (Eds PS
Soltis, DE Soltis, JJ Doyle) pp. 50-9 1 . (Chapman & Hall: New
York)
Hanson MA, Gaut BS, Stec AO, Fuerstenberg SI, Goodman MM, Coe
EH, Doebley JF ( ! 996a) Evolution of anthocyanin biosynthesis in maize kernels: the role of regulatory and enzymatic loci. Genetics
143, 1 395-1407.
Hanson RE, Islam-Faridi M, Percival EA, Crane CF, Ji Y, McKnight
TD, Stelly DM, Price HJ ( 1996b) Distribution of 5S and 1 8S-28S rDNA loci in a tetraploid cotton (Gossypium hirsutum L.) and its putative diploid ancestors. Chromosoma 105, 55-6 1 . doi: 1 0. 1007/S004 1 20050 ! 59
Hartmann S, Nason JD, Bhattacharya D (200 I ) Extensive ribosomal
DNA genic variation in the columnar cactus Lophocereus. Journal of Molecular Evolution 53, 124--1 34.
Henikoff S, Greene EA, Pietrokovski S, Bork P, Attwood TK, Hood L
( 1 997) Gene families: the taxonomy of protein paralogs and chimeras. Science 278, 609-614. doi: I 0. 1 1 26/SCIENCE.278.5338.609
Hilton H, Gaut BS (1 998) Speciation and domestication in maize and its wild relatives: evidence from the Globulin-I gene. Genetics 150,
863-872.
Hodges SA, Arnold ML (1 994) Columbines: a geographically widespread species flock. Proceedings of the National Academy of
Sciences of the United States of America 91, 5 1 29-5 1 32.
Hong RL, Hamaguchi L, Busch MA, Weigel D (2003) Regulatory elements of the floral homeotic gene AGAMOUS identified by phylogenetic footprinting and shadowing. The Plant Cell 15,
1 296-1309. doi: I 0. I I 05/TPC.009548
Hughes AL, Yeager M ( 1 998) Natural selection at major histocompatibility complex loci of vertebrates. Annual Review of
Genetics 32, 41 5---435, doi: I 0.1 146/ ANNUREV.GENET.32. 1 .4 1 5
In nan H, Tajima F (1 997) The amounts o f nucleotide variation within and between allelic classes and the reconstruction of the common ancestral sequence in a population, Genetics 147, 143 1-1444.
Jakobsen IB, Easteal S ( 1 996) A program for calculating and displaying compatibility matrices as an aid in determining reticulate evolution in molecular sequences. CABIOS 12, 291-295.
Kawabe A, M iyashita NT (1 999) DNA variation in the basic chitinase locus (ChiB) region of the wild p!ant
Arabidopsis tha/iana. Genetics
153, 1445-1453.
Kawabe A, Innan H, Terauchi R, Miyashita NT ( 1 997) Nucleotide polymorphism in the acidic chitinase locus (ChiA) region of the wild plant Arabidopsis tha/iana. Molecular Biology and Evolution
14, 1 303- 1 3 1 5 ,

168 Australian Systematic Botany R . L . Small et a/.
Klein J, Sato A, Nag! S, O'huigin C ( 1998) Molecular trans-species polymorphism. Annual Review of Ecology and Systematics 29,
1-2 1 . doi: l0. 1 146/ANNUREV.ECOLSYS.29. l . l
Kuittinen H , Aguade M (2000) Nucleotide variation at the CHALCONE
ISOMERASE
863-872. locus in Arabidopsis tha/iana. Genetics 155,
Kuzoff RK, Sweere JA, Soltis DE, Soltis PS, Zimmer EA (1 998) The phylogenetic potential of entire 26S rONA sequences in plants.
Molecular Biology and Evolution 15, 2 5 1 -263.
Lavin M, Eshbaugh E, Hu J-M, Matthews S, Sharrock RA ( 1 998)
Monophyletic subgroups of the tribe Millettieae (Leguminosae) as revealed by phytochrome nucleotide sequence data. American
Journal of Botany 85, 4 1 2-433.
Levy F, Antonovics J, Boynton JE, Gillham NW (1 996) A population genetic analysis of chloroplast DNA in Phacelia. Heredity 76,
143-1 55.
Linder CR, Goertzen LR, Heuvel BV, Francisco-Ortega J, Jansen RK
(2000) The complete external transcribed spacer of 1 8S-26S rDNA: amplification and phylogenetic utility at low taxonomic levels ih
Asteraceae and closely allied families. Molecular Phylogenetics and Evolution 14, 285-303. doi: 1 0. 1 006/MPEV. l 999.0706
Liu F, Zhang L, Charlesworth D ( 1 998) Genetic diversity in
Leavenworthia populations with different inbreeding levels.
Proceedings of the Royal Society of London. Series B. Biological
Sciences 265, 293-30 l . doi: 1 0 . 1 098/RSPB . l 998.0295
Liu Q, Brubaker CL, Green AG, Marshall DR, Sharp PJ, Singh SP
(200 1) Evolution of the FAD2-l fatty acid desaturase 5' UTR intron and the molecular systematics of Gossypium (Malvaceae).
American Joumal of Botany 88, 92-1 02.
Lynch M, Conery JS (2000) The evolutionary fate and consequences of duplicate genes. Science 290, 1 1 5 1- 1 1 55. doi: l 0 . 1 1 26/SCIENCE.290.5494. 1 1 5 1
Malcomber ST (2002) Phylogeny of Gaertnera Lam. (Rubiaceae) based on multiple DNA markers: evidence of a rapid radiation in a widespread, morphologically diverse genus. Evolution 56, 42-57.
Marshall HD, Newton C, Ritland K (200 1) Sequence-repeat polymorphisms exhibit the signature of recombination in
Lodgepole Pine chloroplast DNA. Molecular Biology and
Evolution 18, 21 36-2 1 38.
Mason-Gamer RJ (200 I) Origin of North American Elymus (Poaceae:
Triticeae) allotetraploids based on granule-bound starch synthase gene sequences, Systematic Botany 26, 757-768.
Mason-Gamer RJ, Holsinger KE, Jansen RK ( 1995) Chloroplast DNA haplotype variation within and among populations of Coreopsis grandlj/ora (Asteraceae). Molecular Bio logy and Evolution 12,
37 1-3 8 1 .
Mason-Gamer RJ, Wei! CF, Kellogg E A ( 1998) Granule-bound starch synthase: structure, function, and phylogenetic utility. Molecular
Biology and Evolution 15, 1658-1673.
Mathews S, Sharrock RA ( 1 996) The phytochrome gene family in grasses (Poaceae): a phylogeny and evidence that grasses have a subset of the loci found in dicot angiosperms. Molecular Biology and Evolution 1 3, 1 141-1 1 50,
Mathews S, Tsai RC, Kellogg EA (2000) Phylogenetic structure in the grass family (Poaceae ): evidence from the nuclear gene
Phytochrome B. American Joumal of Botany 87, 96-107.
Mathews S, Spangler RE, Mason-Gamer RJ, Kellogg EA (2002)
Phylogeny of Andropogoneae inferred from Phytochrome B,
GBSSI, and NDHR Intemational Journal of Plant Sciences 163,
441-450. doi: 10. 1 086/339 155
Mayer MS, Soltis PS ( 1 999) Intraspecific phylogeny analysis using ITS sequences: insights from studies of the Streptanthus g/andulosus complex (Cruciferae). Systematic Botany 24, 47-6 1 .
Mayo! M, Rossello JA (200 1) Why nuclear ribosomal DNA spacers
(ITS) tell different stories in Quercus. Molecular Phylogenetics and
Evolution 19, 1 67-176. doi: 10. 1 006/MPEV.200 1 .0934
Millar AA, Dennis ES ( 1 996) The alcohol dehydrogenase genes of cotton, Plant Molecular Biology 31, 897-904.
Miyashita NT, Innan H, Terauchi R ( 1996) Intra- and interspecific variation of the alcohol dehydrogenase locus region in wild plants
Arab is gemmifera and Arabidopsis thaliana. Molecular Biology and
Evolution 13, 433-436.
Moniz de Sa M, Drouin G ( 1 996) Phylogeny and substitution rates of angiosperm actin genes. Molecular Biology and Evolution 13,
1 1 98-1 2 12.
Morton BR, Gaut BS, Clegg MT (1996) Evolution of alcohol dehydrogenase genes in the Palm and Grass families. Proceedings of the National Academy of Sciences of the United States of America
93, 1 1 735-1 1 739.
Muir G, Fleming CC, Schlotterer C (200 1) Three divergent rONA clusters predate the species divergence in Quercus petraea (Matt.)
Liebl. and Quercus robur L. Molecular Biology and Evo/11/ion 1 8,
1 12-1 1 9.
Myerhans A, Vartanian J-P, Wain-Hobon S ( 1 990) DNA recombination during PCR. Nucleic Acids Research 18, 1 687- 1 69 1 .
Olmstead RG, Palmer J D ( 1 994) Chloroplast DNA systematics: a review of methods and data analysis. American Journal of Botany
81, 1 205-1224.
Olsen KM (2002) Population history of Manihot esculenta
(Euphorbiaceae) inferred from nuclear DNA sequences. Molecular
Ecology 11, 901-9 1 1 . doi: 10. 1046/J. l365-294X.2002.0 1493.X
Olsen KM, Schaal BA ( 1999) Evidence on the origin of cassava: phytogeography of Manihot escu/enta. Proceedings oft he National
Academy of Sciences of the United States of America
5586-559 1 . doi: 1 0 . 1 073/PNAS.96: 1 0.5586
96,
Palmer JD (1992) Mitochondrial DNA in plant systematics: applications and limitations. In ' Molecular systematics of plants'.
(Eds PS Soltis, DE Soltis, JJ Doyle) pp. 36-49. (Chapman and Hall:
New York)
Palmer JD, Adams KL, Cho Y, Parkinson CL, Qiu Y-L, Song K (2000)
Dynamic evolution of plant mitochondrial genomes: mobile genes and introns and highly variable mutation rates, Proceedings of the
National Academy of Sciences of the United States of America 97,
6960-6966. doi: 1 O. l 073/PNAS.97. 1 3.6960
Perry DJ, Fumier GR ( 1996) Pinus banksiana has at least seven expressed alcohol dehydrogenase genes in two linked groups.
Proceedings of the National Academy of Sciences of the United
States of America 93, 1 3 020-13023.
Purugganan MD, Suddith JI ( 1 998) Molecular population genetics of the A rabidopsis CAULIFLOWER regulatory gene: nonneutral evolution and naturally occurring variation in floral homeotic function. Proceedings of the National Academy of Sciences of the
United States of America 95, 8 1 30-8 1 34. doi: l0. 1 073/PNAS.95. 14.8 1 30
Qiu Y-L, Cho Y, Cox JC, Palmer JD ( 1998) The gain of three mitochondrial introns identifies liverworts as the earliest land plants. Nature 394, 67 1-674. doi: I 0.1 038/29286
Qiu Y-L, Lee J, Bernasconi-Quadroni F, Soltis DE, Soltis PS, Zonis M,
Zimmer EA, Chen Z, Sauolainen V, Chase MW ( 1 999) The earliest angiosperms: evidence from mitochondrial, plastid, and nuclear genomes. Nature 402, 404-407. doi: 1 0. 1038/46536
Rauscher JT, Doyle JJ, Brown AHD (2002) Internal transcribed spacer repeat-specific primers and the analysis of hybridization in the
Glycine tomentella (Leguminosae) polyploid complex. Molecular
Ecology 11, 269 1-2702. doi: I 0.1 046/J. I 365-294X.2002.0 1 640.X

Use of nuclear genes in plant phylogeny Australian Systematic Botany 1 69
Reboud X, Zeyl C (1994) Organelle inheritance in plants. Heredity 72,
1 32-140.
Richman AD, Kohn JR (1999) Self-incompatibility alleles from
Physalis: implications for historical inference from balanced genetic polymorphisms. Proceedings of the National A cademy of
Sciences of the United States of America 96, 168-172. doi: I 0. I 073/PNAS.96. I , 168
Richman AD, Uyenoyama MK, Kohn JR ( 1 996) Allelic diversity and gene genealogy at the self-incompatibility locus in the Solanaceae,
Science 273, 1 2 1 2- 1 2 1 6.
Rieseberg LH (1 997) Hybrid origins of plant species. Annual Review of
Ecology and Systematics 28, 359-389, doi: I 0. 1 146/ ANNUREV.ECOLSYS.28. I .359
Rieseberg LH, Soltis DE (1991) Phylogenetic consequences of cytoplasmic gene flow in plants. EvolutionmJ' Ji·ends in Plants 5,
65-84.
Rieseberg LH, Wendel JF ( 1993) Introgression and its consequences in plants. In 'Hybrid zones and the evolutionary process'. (Ed. RG
Harrison) pp. 70-109, (Oxford University Press: New York)
Rokas A, Williams BL, King N, Carroll SB (2003) Genome-scale approaches to resolving incongruence in molecular phylogenies.
Nature 425, 798-804, doi: 10. I 038/NATURE02053
Sambrook J, Fritsch EF, Maniatis T ( 1 989) ' Molecular cloning: a laboratory manual.' 2nd edn. (Cold Spring Harbor Laboratory
Press)
Sanderson MJ, Doyle JJ ( 1 992) Reconstruction of organismal and gene phylogenies from data on multigene families: concerted evolution, homoplasy, and confidence. Systematic Biology 4 1 , 4-17.
Sang T (2002) Utility of low-copy nuclear gene sequences in plant phylogenetics. Critical Reviews in Biochemistry and Molecular
Biology 37, 1 2 1 - 1 47.
Sang T, Zhang D ( 1 999) Reconstructing hybrid speciation using sequences of low copy nuclear genes: hybrid origins of five Paeonia species based on Adh gene phylogenies. Systematic Botany 24,
148-163.
Sang T, Crawford DJ, Stuessy TF ( l997a) Chloroplast phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae).
American Journal of Botany 84, 1 1 20-1 1 36.
Sang T, Donoghue MJ, Zhang D ( 1997b) Evolution of alcohol dehydrogenase genes in peonies (Paeonia): phylogenetic relationships of putative nonhybrid species. Molecular Biology and
Evolution 14, 994-1 007.
Sanjur OJ, Piperno DR, Andres TC, Wessel-Beaver L (2002)
Phylogenetic relationships among domesticated and wild species of
Cucurbita (Cucurbitaceae) inferred from a mithochondrial gene: implications for crop plant evolution and areas of origin.
Proceedings of the National A cademy of Sciences of the United
States of America 99, 535-540. doi: 1 0. I 073/PNAS.O 12577299
Sawyer S (1 989) Statistical tests for detecting gene conversion.
Molecular Biology and Evolution 6, 526-536.
Senchina DS, Alvarez I, Cronn RC, Liu B, Rong J, Noyes RD, Paterson
AH, Wing RA, Wilkins TA, Wendel JF (2003) Rate variation among nuclear genes and the age of polyploidy in Gossypium. Molecular
Biology and Evolution 20, 633-643. doi: I 0. I 093/MOLBEV /MSG065
Simmons MP, Savoia in en V, Clevinger CC, Archer RH, Davis JI (200 I )
Phylogeny o f the Celastraceae inferred from 26S nuclear ribosomal
DNA, phytochrome B, rbcL, atpB, and morphology. Molecular
Phy/ogenetics and Evolution 1 9, 353-366. doi: I 0, I 006/MPEV.200 I .0937
Small RL (2004) Phylogeny of Hibiscus sect. Muenchhusia
(Malvaceae) based on chloroplast 1pL16 and ndhF, and nuclear ITS and GBSSJ sequences. Systematic Botany, in press.
Small RL, Wendel JF (2000a) Copy number lability and evolutionary dynamics of the Adh gene family in diploid and tetraploid cotton
(Gossypium). Genetics 155, 1 9 1 3- 1 926.
Small RL, Wendel JF (2000b) Phylogeny, duplication, and intraspecific variation of Adh sequences in New World diploid cottons
(Gossypium, Malvaceae). Molecular Phy/ogenetics and Evolution
1 6, 73-84, doi : I O. I 006/MPEV. 1 999.0750
Small RL, Wendel JF (2002) Differential evolutionary dynamics of duplicated paralogous Adh loci in allotetraploid cotton
(Gossypium). Molecular Biology and Evolution 19, 597-607.
Small RL, Ryburn JA, Cronn RC, Seelanan T, Wendel JF (1998) The tortoise and the hare: choosing between noncoding plastome and nuclear Adh sequences for phylogenetic reconstruction iti a recently diverged plant group. American Journal of Botany 85, 1 30 1 - 1 3 1 5 .
Small RL , Ryburn JA, Wendel J F ( 1 999) Low levels o f nucleotide diversity at homoeologous Adh loci in allotetraploid cotton
(Gossypium L.). Molecular Biology and Evolution 16, 49 1-50 I ,
Soltis D E et a/. ( 1997) Angiosperm phylogeny inferred from l 8S ribosomal DNA sequences. Annals of the Missouri Botanical
Garden 84, 1-49.
Soltis DE, Soltis PS ( 1998a) Choosing an approach and an appropriate gene for phylogenetic analysis. In 'Molecular systematics of plants
II: DNA sequencing'. (Eds DE Soltis, PS Soltis, JJ Doyle) pp. 1-42.
(Kluwer Academic Publishers: Boston)
Soltis PS, Soltis DE (1 998b) Molecular evolution of I SS rDNA in angiosperms: implications for character weighting in phylogenetic analysis. In ' Molecular systematics of plants II: DNA sequencing'.
(Eds DE Soltis, PS Soltis, JJ Doyle) (Kluwer Academic Publishers:
Boston, MA)
Soltis PS, Soltis DE, Chase MW ( 1999) Angiosperm phylogeny inferred from multiple genes as a tool for comparative biology.
Nature 402, 402-404. doi: I 0, I 038/46528
Steele KP, Holsinger KE, Jansen RK, Taylor DW (1991) Assessing the reliability of 5S ribosomal-RNA sequence data for phylogenetic analysis in green plants, Molecular Biology and Evolution 8,
240-248.
Stephens JC (1 985) Statistical methods of DNA sequence analysis: detection of intragenic recombination or gene conversion.
Molecular Biology and Evolution 2, 539-556.
Strand AE, Leebens-Mack J, Milligan BG (1 997) Nuclear DNA-based markers for p lal]t evolutionary biology. Molecular Ecology 6,
1 1 3- 1 1 8. doi: I 0. 1 046/J. 1 365-294X. 1 997.00 1 53.X
Suh Y, Thien LB, Reeve HE, Zimmer EA (1 993) Molecular evolution and phylogenetic implications of internal transcribed spacer sequences of ribosomal DNA in Winteraceae. American Journal of
Botany 80, 1 042-1055,
Swofford DL (2002) 'PAUP*. Phylogenetic analysis using parsimony
(*and other methods).' (Sinauer Associates: Sunderland, MA)
Taberlet P, Gielly L, Pautou G, Bouvet J ( 1 99 1 ) Universal primers for amplification of three non-coding regions of chloroplast DNA.
Plant Molecular Biology 17, 1 1 05-1 1 09.
Tank DC, Sang T (200 1 ) Phylogenetic utility of the glycerol-3-phosphate acyltransferase gene: evolution and implications in Paeonia (Paeoniaceae), Molecular Phylogenetics and Evolution 19, 42 1-429. doi: 1 0. I 006/MPEV.2001 .093 1
The Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabi do psis thaliana. Nature 408,
796-8 15. doi: I 0. I 038/35048692
Thornton JW, DeSalle R (2000) Gene family evolution and homology: genomics meets phylogenetics. Annual Review of Genomics and
Human Genetics 1 , 4 1-73. doi: 10, 1 1 46/ANNUREV.GENOM. I . 1 .41

170 Australian Systematic Botany R. L. Small et a/.
Wagner A (1998) The fate of duplicated genes: loss or new function?
BioEssays 20, 785-788. doi: 10. 1002/(SICI) l 52 1 - l 878(199810)20: 103.3.C0;2-D
Wagner A (200 I ) Birth and death of duplicated genes in completely sequenced eukaryotes. 'JYends in Genetics 17, 237-239. doi: 10. 10 1 6/SO 1 68-9525(0 I )02243-0
Wagner A, Blackstone N, Cartwright P, Dick M, Misof B, Snow P,
Wagner GP, Barttels J, Murtha M, Pendleton J ( 1994) Surveys of gene families using polymerase chain reaction: PCR selection and
·
PCR drift. Systematic Biology 43, 250-26 1 .
Wang R-L, Stec A, Hey J, Lukens L, Doebley J (1 999) The limits of selection during maize domestication. Nature 398, 236-239. doi: 10. 1038/ 1 8435
Waters ER (1995) The molecular evolution of the small heat-shock proteins in plants. Genetics 141, 785-795.
Wendel JF (1989) New World tetraploid cottons contain Old World cytoplasm. Proceedings of the National Academy of Sciences oft he
United States of America 86, 4 132-4 1 36.
Wendel JF, Albert VA (1992) Phylogenetics of the cotton genus
( Gossypium L.): character-state weighted parsimony analysis of chloroplast DNA restriction site data and its systematic and biogeographic implications. Systematic Botany 17, 1 1 5-143.
Wendel JF, Doyle JJ (1998) Phylogenetic incongruence: window into genome history and molecular evolution. In 'Molecular systematics of plants II. DNA sequencing'. (Eds D Soltis, P Soltis, J Doyle).
(Kiuwer Academic Publishing)
Wendel JF, Schnabel A, Seelanan T (1995a) B idirectional interlocus concerted evolution following allopolyploid speciation in cotton
( Gossypium). Proceedings of the National A cademy of Sciences of the United States of America 92, 280-284.
Wendel JF, Schnabel A, Seelanan T (J 995b) An unusual ribosomal
DNA sequence from Gossypium gossypioides reveals ancient, cryptic, intergenomic introgression. Molecular Phylogenetlcs and
Evolution 4, 298-3 13. doi: I O. l 006/MPEV. l 995. 1027
Wendel JF, Cronn RC, Johnston JS, Price HJ (2002) Feast and famine in plant genomes. Gene/lea 1 15, 37-42. doi: 10. 1023/ A: 1 0 1 6020030189
White SE, Doebley JF ( 1 999) The molecular evolution of terminal earl , a regulatory gene in the genus Zea. Genetics 153, 1455-1462.
Whitten WM, Williams NH, Chase MW (2000) Subtribal and generic relationships of Maxillarieae (Orchidaceae) with emphasis on
Stanhopeinae: combined molecular evidence. American Journal of
Botany 87, 1 842-1 856.
Wolfe KH, Li W-H, Sharp PM (1987) Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear
DNAs. Proceedings of the National Academy of Sciences of the
United States of America 84, 9054-9058.
Wray GA, Hahn MW, Abouheif E, Balhoff JP, Pizer M, Rockman MY,
Romano L (2003) The evolution of transcriptional regulation in eukaryotes. Molecular Biology and Evolution 20, 1 377-1 419. doi: I 0.1 093/MOLBEV /MSG 1 40
Yu J, Hu S, Wang J, Wong GK-S, Li S et a/. (2002) A draft sequence of the rice genome (01yza sativa L. ssp. indica). Science 296, 79-92. doi: I 0. 1 1 26/SCIENCE. I 068037
Zhang L, Pond SK, Gaut BS (2001) A survey of the molecular evolutionary dynamics of twenty-five multigene families from four grass taxa. Jouma/ of Molecular Evolution 52, 144-156.
Zimmer EA, Mattin SL, Beverly SM, Kan W, Wilson AC (I 980) Rapid duplication and loss of genes coding for the a chains ofhemoglobin.
Proceedings of the National Academy of Sciences of the United
States of America 77, 2 1 58-2 1 62.
Manuscript received 1 2 June 2003, accepted 28 January 2004 http://www.publish.csiro.au/journals/asb