Molecular evolution of Adh and LEAFY and the phylogenetic utility of their introns in Pyrus (Rosaceae)
© Zheng et al; licensee BioMed Central Ltd. 2011
Received: 11 February 2011
Accepted: 14 September 2011
Published: 14 September 2011
The genus Pyrus belongs to the tribe Pyreae (the former subfamily Maloideae) of the family Rosaceae, and includes one of the most important commercial fruit crops, pear. The phylogeny of Pyrus has not been definitively reconstructed. In our previous efforts, the internal transcribed spacer region (ITS) revealed a poorly resolved phylogeny due to non-concerted evolution of nrDNA arrays. Therefore, introns of low copy nuclear genes (LCNG) are explored here for improved resolution. However, paralogs and lineage sorting are still two challenges for applying LCNGs in phylogenetic studies, and at least two independent nuclear loci should be compared. In this work the second intron of LEAFY and the alcohol dehydrogenase gene (Adh) were selected to investigate their molecular evolution and phylogenetic utility.
DNA sequence analyses revealed a complex ortholog and paralog structure of Adh genes in Pyrus and Malus, the pears and apples. Comparisons between sequences from RT-PCR and genomic PCR indicate that some Adh homologs are putatively nonfunctional. A partial region of Adh1 was sequenced for 18 Pyrus species and three subparalogs representing Adh1-1 were identified. These led to poorly resolved phylogenies due to low sequence divergence and the inclusion of putative recombinants. For the second intron of LEAFY, multiple inparalogs were discovered for both LFY1int2 and LFY2int2. LFY1int2 is inadequate for phylogenetic analysis due to lineage sorting of two inparalogs. LFY2int2-N, however, showed a relatively high sequence divergence and led to the best-resolved phylogeny. This study documents the coexistence of outparalogs and inparalogs, and lineage sorting of these paralogs and orthologous copies. It reveals putative recombinants that can lead to incorrect phylogenetic inferences, and presents an improved phylogenetic resolution of Pyrus using LFY2int2-N.
Our study represents the first phylogenetic analyses based on LCNGs in Pyrus. Ancient and recent duplications lead to a complex structure of Adh outparalogs and inparalogs in Pyrus and Malus, resulting in neofunctionalization, nonfunctionalization and possible subfunctionalization. Among all investigated orthologs, LFY2int2-N is the best nuclear marker for phylogenetic reconstruction of Pyrus due to suitable sequence divergence and the absence of lineage sorting.
The genus Pyrus L. belongs to the tribe Pyreae Baill. (the former subfamily Maloideae C. Weber) of the family Rosaceae  and is geographically divided into two groups: occidental pears and oriental pears . The majority of oriental pears are native to China; a few are native to Japan and the Korean Peninsula. Chinese taxonomists agreed on 13 Pyrus species native to China, among which P. betulaefolia Bge. and P. calleryana Dcne. have retained characteristics believed to be ancestral for Pyrus  including the smallest fruit size and lowest carpel number. Based on morphological traits or crossing experiments, P. × bretschneideri Rehd., P. × serrulata Rehd., P. × sinkiangensis Yu, and P. × hopeiensis Yu are putative hybrids among other Pyrus species [4–6]. The circumscription of species, subspecies, and forms for occidental species remains controversial. It is believed that some cultivated pears frequently escaped from cultivation and became feral. These plants hybridize easily, both with cultivated and wild species, resulting in a number of intermediate forms and segregants . Therefore, morphological characters are poor indicators of Pyrus phylogeny. Other data sources, like chemical characters, were used to distinguish some pear species , but these were plagued by low number of characters, polymorphisms, and environmental plasticity. During the last decade, molecular markers including RFLPs , RAPDs [10–12], genomic-SSRs [13, 14], EST-SSRs [15, 16] and AFLPs [10, 17] have been applied in Pyrus. These data provided useful information on the origin of some cultivated pear groups, e.g. Chinese white pears (CWP), which are assigned to P. × bretschneideri. However, CWP are morphologically different from the so-called wild P. × bretschneideri in northern Hebei province . They also show a close relationship to P. pyrifolia based on multiple molecular marker data and thus were treated as P. pyrifolia White Pear Group [11, 14, 17]. However, most of the studies have focused on the relationships of several oriental or occidental species or cultivar pear groups, and the phylogeny of the genus remains unresolved.
Plastid DNA sequence data and the internal transcribed spacer (ITS) of nuclear ribosomal DNA (nrDNA) have been used for plant phylogenetic reconstruction due to ease of amplification. In the Rosaceae, these data have been applied in phylogenetic studies at different taxonomic levels [18, 19], but their utility is limited due to varied evolutionary rates [20–25]. Intra-individual ITS polymorphisms caused by incomplete concerted evolution of nrDNA arrays have been found in many Rosaceae [20, 21, 26, 27]. Such polymorphisms provided evidence for hybrid origins of some species in Rosa , but they led to a poorly resolved phylogeny in Malus . Similarly, our previous study in Pyrus  revealed a history of non-concerted evolution of ITS and a poorly resolved phylogenetic tree. Six non-coding regions of plastid DNA were found to be highly conserved in Pyrus , but they reflect only the maternal genealogies.
An alternative source of molecular sequence data, low-copy nuclear genes (LCNGs), has proven to be more phylogenetically informative than either ITS or plastid DNA . These genes reflect biparental lineages and are less prone to homogenization [31–33]. Due to the accumulation of large number of gene sequences in GenBank, it is now possible to design taxa-specific primers. However, paralog and lineage sorting problems are still challenges to applying LCNGs to phylogenetic studies, since they may lead to topological incongruence similar to those caused by hybridization [31, 34]. Gene duplication is a prominent feature of plant genome evolution, and duplicate segments account for 60% of the Arabidopsis thaliana genome . In molecular phylogenetic studies, nuclear genes undergoing gene duplications or the birth-and-death process lead to problems in the identification of orthologs and paralogs and discordance between gene and species trees. Additionally, frequent gene duplications made the terms 'paralogy' and 'orthology' ambiguous. Thus the new terms 'inparalog' for paralogs that evolved after the ingroup speciation and 'outparalog' for those that evolved before ingroup speciation occurred . Lineage sorting (or deep coalescence) results from random fixation of ancestral polymorphic alleles, which may induce similar topological incongruence to that of hybridization, and poses the most challenging problems for inter- and intra-specific phylogenetic inference [31, 37]. However, lineage sorting is a random process, and fixation of ancestral alleles among species is rarely identical for two unlinked nuclear loci. Therefore, incongruence caused by hybridization and lineage sorting could be differentiated when comparing phylogenies based on multiple unlinked nuclear loci.
LCNGs that succeeded in other Rosaceae are potentially ideal nuclear markers for phylogenetic studies of Pyrus. The coding region of GBSSI has been successfully applied at intergeneric and higher levels [26, 38], but the introns are too short and dispersed to be ideal gene regions for interspecific levels studies. Another gene region is the second intron of LEAFY, which is long enough and has been proven to be informative for studies at the interspecific level in Pyreae [39, 40]. Complete coding sequences of two LEAFY loci have been isolated in Pyrus , and the corresponding genomic sequences in Malus species are available (DQ535885-AFL1, DQ535886-AFL2). Thus obtaining introns of LEAFY in the Pyrus taxa is possible.
Alcohol dehydrogenase gene (Adh) is one of the best-studied nuclear-encoded genes in plants. Two major ADH classes, class P with alcohol activity and class III with glutathione-dependent formaldehyde activity, have been identified in flowering dicot or monocot plants. The former is common for plants and the latter has been isolated in a few taxa including Pisum sativum (P80572) , Oryza sativa (U77637) and Araobidopsis thaliana (X82647) . The Adh gene occurs in small gene families, and has proven to be a useful phylogenetic marker in the Poaceae and Paeoniaceae [44–46], but it is too complex a gene family or provided little phylogenetic resolution in other taxa such as the Gossypium and Carex [47, 48]. Two distinct Adh loci (AF031900, AF031899) have been isolated in P. communis 'Packham's Triumph' , thus it is possible to isolate Adh genes among Pyrus species. However, the only genomic Adh sequence from a species of Rosaceae is that from Fragaria ananassa (X15588) in GenBank , and the exon/intron structure is unknown in Pyrus. The phylogenetic utility of Adh coding regions and the intron region have not been determined in any Rosaceae taxa.
Since no LCNG analysis had been applied to phylogenetic studies of Pyrus, we explored the utility of LEAFY and Adh. In this study, a comparison of genomic and RT-PCR-based approaches yielded an initial description of the composition and functionality of the Adh gene family in Pyrus. The phylogenetic utility of Adh gene regions and the second intron of LEAFY were determined after examining the sequence divergence, gene duplications, lineage sorting and recombination.
Plant taxa used in this study and subparalogs of LFY1int2, LFY2int2 and Adh1-1 recovered in each accession
Leaf source b
Subparalogs or copy typesc
P. × sinkiangensis Yu
a, b, c
Liaoning Province, China
a, b, c
P. pyrifolia Nakai
Kanagawa Pref. Japan
Chiba Pref. Japan
a, b, c
P. pyrifolia White pear group
Hebei Province, China
P. pyrifolia White pear group
Anhui Province, China
a, b, c
P. ussuriensis Maxim.
Liaoning Province, China
Liaoning Province, China
a, b, c
P. pashia 1
P. pashia D.Don
Yunnan Province, China
P. pashia 2
Yunnan Province, China
a, b, c
P. dimorphophylla Makino
Mie Pref. Japan
P. calleryana Dcne.
P. fauriei Schneid.
P. betulaefolia Bge.
Gansu Province, China
P. × serrulata
P. × serrulata Rehd.
Hubei Province, China
P. xerophila Yu
Gansu Province, China
P. × hopeiensis
P. × hopeiensis Yu
Hebei Province, China
Hebei Province, China
P. × phaeocarpa
P. × phaeocarpa Rehd.
a, b, c
Nakai & Kikuchi
a, b, c
P. communis L.
a, b, c
P. elaeagrifolia Pall.
Turkey, Crimea, South East Europe
P. amygdaliformis Vill.
Mediterranean area, South Europe
P. cossonii Rehder.
M. sieboldii (Regel.) Rehd
Yunnan Province, China
a, b, c
M. rockii Schneid.
Yunnan Province, China
M. domestica subsp. chinensis
M. domestica subsp. chinensis Li Y. N.-(Nai)
a, b, c
M. neidzwetzkyana (Dieck) Langenf.
Xinjiang (Uygur Autonomous Region)
M. domestica Borkh
Gene structure and paralog identification based on long Adh sequences
Primers used for PCR amplification and sequencing in this study
Primer sequence (5'-3')
Long partial region
Sequence divergence (mean value) of the coding regions between and within Adh homologs (excluding putative pseudogenes)
Phylogenetic analyses based on reduced Adh1 sequences
Sequence variations of Adh1-1 and LEAFY subparalogs in Pyrus (excluding the Malus accessions)
The second intron of LEAFY
Transcription of Adh homologs
As indicated in Additional file 2 the transcription of an Adh1-2 copy was not observed. We speculate that Adh1-2 is a nonfunctional outparalog, since putative pseudogenes have been identified and its 3'UTR region is divergent from Adh1-1 as described above. Among the Adh1-1 outparalogs the most frequently cloned Adh1-1c was not recovered by RT-PCR, indicating that it was a degenerate outparalog. This also explains its preferential amplification. Transcription of the other Adh1-1 outparalogs was detected among different tissues or cultivars. Transcription of Adh2-1 could be detected in all investigated tissues and cultivars. An Adh2-1 sequence was not obtained in the three Malus accessions by G-PCR, but its transcription was found in 'Fuji' (M. domestica) (Additional file 3). Adh2-2 was only observed in Malus taxa, and transcription of this paralog was not detected. We could not deduce distinct tissue-specific or cultivar-specific expression for either Adh1-1 or Adh2-1. Several anomalous Adh2-1 copies with only intron 6 were recovered by cloning of RT-PCR products in 'Flemish Beauty' (P. communis) and 'Nanguoli' (P. ussuriensis) (HQ912054, HQ912055 and HQ912056). Since intron-containing cDNA has not been reported, those copies may be amplified due to genomic DNA contamination and indicate existence of an additional Adh2 paralog with more intron loss in Pyrus.
Sequence variation and paralogs of LEAFY
Twenty-six LFY1 and 27 LFY2 sequences including a partial exon 2 were obtained. Four groups with different length variations in the exon region were observed: LFY1-Malus (407 bp), LFY1-Pyrus (395 bp), LFY2-Malus (410 bp) and LFY2-Pyrus (401 bp), which were congruent with those from RT-PCR in P. pyrifolia 'Housui' (LFY1-AB162029, LFY2-AB162035) and M. domestica 'Fuji' (LFY1-AB162028 LFY2-AB162034), respectively. The length of these indels only had an effect on the length of the deduced amino acid sequences. Among Pyrus accessions, NSD of this partial exon 2 between LFY1 and LFY2 was relatively high (0.076), while that within LFY1 and LFY2 was low at 0.013 and 0.015, respectively (data not shown).
The entire intron 2 of LFY1 among Pyrus species ranged from 774-783 bp, while that of LFY2 ranged from 670 to 700 bp. LFY1int2 and LFY2int2 were amplified in all accessions using specific primers, but LFY2int2 of M. rockii and M. domestica subsp. chinensis were amplified by another forward primer MLFY2-F (Table 2) due to a large deletion (approximately 220 bp, GU991522 vs DQ535886) in these two accessions. LFY2int2 of all Malus accessions contained a 211-bp insertion, which made alignment difficult and was removed from the analyses. NSD within LFY1int2 and LFY2int2 among Pyrus accessions was 0.019 and 0.029, respectively (Table 4), while that between Malus and Pyrus was much higher at 0.057 for LFY1int2 and 0.066 for LFY2int2 (data not shown).
Sequence variation and phylogenetic analyses of LFY1int2 suggest two subparalogs, LFY1int2-a and LFY1int2-b, among Pyrus species. Compared with LFY1int2-a, LFY1int2-b contains a 6-bp insertion. Recovery of LFY1int2-a and LFY1int2-b in each accession is shown in Table 1. Among oriental species, NSD within LFY1int2-a was lower than that within LFY1int2-b for both ORF regions (0.002, 0.009) and intron regions (0.005, 0.013). Between LFY1int2-a and LFY1int2-b, NSD was 0.007 for the ORF region and 0.017 in for the intron region. For LFY2int2, subparalogs with an 8-bp insertion (LFY2int2-Ins8) and 2-bp deletion (LFY2int2-Del2) were only recovered in a few accessions. The common LFY2int2-N, with no indel, was recovered in all accessions but 'Yali' (P. pyrifolia, CWP) (Table 1). Coexistence of these subparalogs in one individual could be detected by direct sequencing due to the fixed position of indels. A minimum of three clones were sequenced. It was found that PCR or direct sequencing sometimes did not reflect the subparalogs existing in one genome, probably due to amplification preference of different nuclear alleles. For example, direct sequencing of LFY2int2 in P. calleryana identified LFY2int2-N, but a LFY2int2-Del2 sequence was obtained by cloning.
Several anomalous LFY2int2 copies were exclusively found in particular accessions. LFY2int2 from P. communis had a 21-bp deletion. LFY2int2 from P. fauriei had a 525-bp insertion that was partially homologous (reverse and complement) to the noncoding region of the S-RNase gene (AB308360), and was named LFY2int2-S. To eliminate the possibility of genetic recombination during PCR, all accessions were tested with an insert-specific upstream forward primer 'LFY2S-F' (Table 1) and the reverse primer LFY2-R. As a result, the LFY2int2-S was detected in 'Yali' (P. pyrifolia, CWP), 'Jianbali' (P. pyrifolia CWP), and 'Yaguang' (P. ussuriensis), but the sequences were not included in the phylogenetic analyses. LFY2int2-S in P. fauriei was still included in the dataset after the exclusion of its large insertion.
Phylogenetic analyses of LFY1int2 and LFY2int2
Malus sieboldii and M. domestica (DQ535885-ALF1, DQ535886-ALF2) without the 220-bp deletion in LFY2int2 were used as outgroups in phylogenetic analyses of LFY1int2 and LFY2int2. Putative recombinants were identified by RDP or by investigating abnormal substitution patterns and ambiguous alignment positions. The putative recombinants displayed unique substitutions of sequences from two distinct subclades and always formed well-separated clades in the tree, thus they were excluded from the final analyses (data not shown). A total of 57 LFY1int2 (four from Malus) and 46 LFY2int2 (three from Malus) sequences were included in two separate datasets. Excluding sequences from Malus, the LFY1int2 dataset had an aligned length of 653 sites, of which 61 (9.3%) were parsimony informative, while LFY2int2 had an aligned length of 562 sites and 46 (8.2%) were parsimony informative (Table 4). Similarly, only Bayesian trees were shown for both LFY1int2 and LFY2int2 datasets.
Frequency of Adh and LEAFY duplication
Gene duplication plays an important role in increasing the diversity of gene function and expression, which can enable plants to colonize diverse habitats. Most monocots and dicots have at least two Adh genes, indicating that an initial Adh gene duplication occurred before the divergence of these plant taxa, and separate duplications have subsequently taken place . Two major loci in Malus and Pyrus, Adh1 and Adh2, are outparalogs derived from an ancient gene split. Based on an estimated 0.66% rate of nucleotide substitution per million year of Adh in Drosophila  and 0.2-0.3% in mammalian nuclear genes , the split occurred approximately 50 million years ago. In maize Adh1 and Adh2 share 87% identity at the amino acid sequence level but are located on different chromosomes and differ in the level of tissue-specific expression . The expression of three Adh genes with 85% and 87% shared amino acid identity in Vitis vinifera varied in developmental stage of grape berries and affinity to either ethanol or acetaldehyde as a substrate . ADH from apple had optimal acetaldehyde activity at pH 5.5-6.0 and ethanol activity at pH 7.0-10.0 . Therefore, Adh1-1 and Adh2-1 in Pyrus (ASD = 0.170, Table 3) with less shared identity, likely also have diversified their expression patterns and substrate affinity. ADH plays an essential role in the biosynthetic pathway of aroma volatiles in apple and pear fruits by reducing aldehydes to alcohols [60–62]. It will be interesting to determine functional divergence of Adh genes and kinetic properties of the corresponding ADH enzymes in Pyrus and Malus.
Two paralogs representing Adh1 (Adh1-1 and Adh1-2) and Adh2 (Adh2-1 and Adh2-2) were observed by G-PCR. Among these the Adh2-1 may be the most ancestral since it has the classical nine-intron Adh structure, which is widely conserved among angiosperms and gymnosperms . Intron losses have been found in Adh genes of diverse taxa like Arabidopsis thaliana [63, 64] and Mangifera indica . In some species of Leavenwortia, an expressed intronless Adh3 locus occurs and is thought to have arisen by an mRNA intermediate . The single intron loss found in our study lends support to the 'intron exclusion hypothesis', which suggests that a single intron could be precisely removed by double strand breaks from a multiple-intron gene . Adh1-2, with the same gene structure as Adh1-1, was a putative nonfunctional outparalog derived by gene duplication (loss of intron 7, Figure 1). Gene duplication leading to Adh1-2, as inferred by the tree topology (Figure 2), occurred before Malus and Pyrus diverged, but probably after diversification of Pyreae taxa, since the orthologous Adh gene in Fragaria (Rosoideae) displayed the classical nine-intron structure. Additionally, the NSD between Adh1-1 and Adh1-2 in Pyrus was 0.06 (Table 3), which was much lower than that between Adh1-1 in Pyrus and Adh in Fragaria (0.10) (data not shown). However, the NSD in the coding region between these outparalogs was too low to confirm the non-functionality of Adh1-2. Adh2-1 was not obtained in Malus accessions by G-PCR, probably due to an amplification preference for Adh2-2, but its transcription was detected in 'Fuji' (Additional file 3). Adh2-2 was recovered neither in Pyrus by G-PCR and SG-PCR nor in Malus or Pyrus by RT-PCR. Compared with Adh1-1 and Adh1-2, the ASD between Adh2-1 and Adh2-2 was much higher (Table 3) and exon/intron structure also varied (Figure 1). Due to the lack of highly homologous Adh2 sequences from other Rosaceae taxa, is the origin of Adh2-2 is uncertain. It may be a duplicated inparalog derived from Adh2-1 and restricted to Malus taxa or a functional outparalog that appeared before Malus and Pyrus diverged and was subsequently lost during diversification of Pyrus species. The latter theory is similar to the paralog sorting of RPB1 and RPB2 in different core eudicots taxa . Specific RT-PCR for additional tissues and specific G-PCR for more Rosaceae taxa will be needed to examine the origin and transcription of Adh2-2. Three subparalogs representing Adh1-1 (Adh1-1a, Adh1-1b, Adh1-1c) were identified by accession sequencing (Figure 3), although Adh1-1c may not be transcribed. It is unknown whether similar subparalogs have evolved for Adh1-2 and Adh2-2.
Our study revealed that ancient and recent duplications led to the complex structure of Adh outparalogs in Pyrus and Malus resulting in neofunctionalization, nonfunctionalization and possible subfunctionalization, the three common fates of gene duplications. Adh homologs in Malus and Pyrus were more complex than those in other angiosperms like Paeonia , grasses  or legumes , but similar to those in Gossypium . Gossypium has at least seven Adh loci of two primary lineages in diploid species and the Adh gene family is dynamic with pseudogenization and gene elimination. Genomic data suggest that almost all angiosperms, perhaps even all plant groups, have experienced one to several rounds of polyploidy [71, 72]. Though Malus and Pyrus accessions used in this study were all diploid (x = 17), Pyreae taxa with x = 17 are derived from autopolyploidization of the formerly Spiraeoid ancestors with x = 9 . This apparently accounts for such complex paralogs, and similar Adh gene structures could be imputed for other Pyreae taxa with x = 17.
LEAFY was first found to be a homeotic gene encoding a transcription regulator for differentiation of the floral meristem and flowering time in Arabidopsis and was expected to be a single-copy gene in diploid angiosperms . In our study, two major lineages, LFY1 and LFY2, were recovered in both Pyrus and Malus, as in many other Pyreae taxa including the formerly Spiradeae taxa with x = 9 [39, 41], suggesting gene duplication of these two paralogs before diversification of the Rosaceae. LFY1 and LFY2 in apple are located on distinct chromosomes and thus are not alleles . In a study including the pear cultivars 'Housui' (P. pyrifolia) and 'Barlett' (P. communis), the transcriptional patterns of two LEAFY homologs differed in developmental stages and tissues, and each homolog varied among plant taxa . In our study, genus-specific and locus-specific indels were discovered in coding regions, which would alter the length of the corresponding amino acid sequences. These might be responsible for the diversification of LEAFY gene functions.
Multiple inparalogs of LFY1int2 and LFY2int2 were observed, and their recovery varied among our accessions (Table 1). LFY1int2-a and LFY1int2-b were unequally observed among Pyrus species (Table 1). LFY1int2-a is monophyletic with shorter branch lengths than LFY1int2-b (Figure 5), suggesting it might be an inparalog derived from LFY1int2-b by a recent duplication. This would have occurred after the divergence of occidental and oriental pears, because only LFY1int2-a is not found in occidental pears. Consequently, LFY1int2-b was lost during diversification of some oriental species, which explains paralog sorting during diversification of Pyrus. Three LFY1int2-c sequences in two west Asian species (P. amygdaliformis and P. elaeagrifolia) formed a separate clade (Figure 5). We suggest two possible explanations for their origin: 1) they are pseudogenes derived from LFY1int2-b, but have evolved more rapidly, and thus are highly divergent from LFY2int2-b; 2) they represent another outparalog of LFY1int2, derived from gene duplication that occurred before diversification of the occidental species, that was subsequently lost in some occidental species. However, only the second intron of LFY1int2-c was sequenced. To help differentiate between these two possibilities the entire exonic region must be obtained and its presence in more Pyrus species investigated. For LFY2int2, the common LFY2int2-N was recovered in all accessions but 'Yali' (P. pyrifolia, CWP). LFY2int2-Ins8 and LFY2int2-Del2 are inparalogs that originated recently after Pyrus diversification and were only recovered in a few accessions (Figure 6). LFY2int2-S, with a long insertion homologous noncoding region of S-RNase gene, was similar to functional AFL1a copies found in some apple cultivars . Genomic Southern analysis also showed that apple had other homologues in addition to AFL1 and AFL2 . However, the relationships among these homologs have not been published. Only a few accessions contained LFY2int2-S (Table 1), and only the one from P. fauriei was included in our analyses. It is unknown how the intron of the RNase gene was inserted in the second intron of LFY2 and whether LFY2int2-S is functional. Additional research found both LFY2int2-S and LFY2int2-N in multiple species, and showed the LFY2int2-S sequences were all highly divergent from LFY2int2-N even after exclusion of the large insertion (unpublished data).
Incongruence and poor resolution
Paralogs and lineage sorting are two major challenges when conducting phylogenetic analyses based on LCNGs, because they can lead to incongruent patterns similar to those resulting from hybridization and polyploidization . Paralogs reflect a horizontal event, the gene duplication in one species, while orthologs reflect a vertical event, the speciation in a lineage [36, 76]. Thus, it is crucial to differentiate paralogs from orthologs by investigating their origins and monophyletic positions in the tree. We have clearly identified the outparalogs and inparalogs for both Adh and LEAFY genes. If inparalogs representing LFY1int2 and LFY2int2 were not identified, P. calleryana possessing both LFY1int2-a and LFY1int2-b as well as LFY2int2-N and LFY2int2-Del2 would be polyphyletic in both gene trees (Figure 5 and 6) and presumed to be hybrids involving other Pyrus species. However, P. calleryana is one of the most ancestral species in Pyrus and should not be a hybrid of other Pyrus species. Additionally, Pyrus ussuriensis, P. × hopeiensis, P. × phaeocarpa and P. hondoensis containing only LFY1int2-a fell in the same clade. This shows a close relationship (Figure 5). However, P. betuleafolia was not in the clade and only contained LFY1int2-b. These findings were inconsistent with the hypothesis that P. betuleafolia was involved in the origin of P. × hopeiensis and P. × phaeocarpa and to the phylogeny based on LFY2int2-N.
Interspecific hybridization has been considered the major mode of evolution for Pyrus , and LCNG has been useful for testing the hypothesis of hybridization, since homologs of a nuclear locus from both parents could be detected in putative hybrids through cloning . Excluding the possibility of paralogs, incongruence caused by hybridization and lineage sorting could be differentiated by comparing phylogenies of multiple unlinked nuclear loci. Only Adh1-1c and LFY2int2-N are shown to be two independent orthologs, and they were recovered in all accessions (except LFY2int2-N in 'Yali'). As described above, relationships revealed by Adh1-1c were poorly resolved, and most intra-species and intra-individual sequences were polymorphic. As shown in Figure 3, intra-individual sequences of P. calleryana, P. hondoensis and P. dimorphophylla were polyphyletic, and the occidental species were not monophyletic. However, the putative interspecific hybrids, P. × hopeinensis and P. × phaeocarpa, were monophyletic, which was incongruent with other gene trees and our previous understanding of these species. In contrast, only four accessions were polymorphic in the LFY2int2-N tree, including P. × phaeocarpa, P. × hopeiensis, and 'Korlaxiangli' (P. × sinkiangensis) (marked in red in Figure 6), all of which were putative interspecific hybrids. Therefore, lineage sorting of ancestral polymorphic Adh1-1c alleles may have occurred during diversification of Pyrus.
The phylogenetic relationships revealed by LFY2int2-N were mostly congruent to other orthologous gene trees and previous studies based on other data. Close relationships among P. calleryana, P. dimorphophylla, P. pashia, P. pyrifolia, and/or P. × serrulata were supported by all gene trees, suggesting a close relationship among these species. Two distinct LFY2int2-N sequences of 'Korlaxiangli' (P. sinkiangensis) were grouped with occidental species and P. pyrifolia, respectively (Figure 6). Similar relationships were found in Adh1-1a, Adh1-1c and Adh1-2 clades (Figure 3), which supports the hypothesis that P. sinkiangensis is an interspecific hybrid involving at least P. communis and P. pyrfolia . Intra-individual copies of P. × hopeiensis and P. × phaeocarpa were grouped with P. ussuriensis, P. hondoensis or P. betulaefolia in both the Adh1-1a clade (Figure 3) and LFY2int2-N (Figure 6) clade. Pyrus × phaeocarpa was a putative hybrid involving P. betulaefolia and P. ussuriensis, and P. × hopeiensis was a hybrid involving P. × phaeocarpa and P. ussuriensis. Pyrus hondoensis, which was once classified as a variety of P. ussuriensis by morphological data , and P. ussuriensis were found to be closely related . Phylogenetic relationships among these species were supported by multiple orthologous gene data, suggesting ancient hybridization rather than lineage sorting. More wild individuals of these species are needed to test such complex evolutionary histories.
The relationships based on all separate orthologs were mostly poorly resolved. In our study, different Adh and LEAFY paralogs showed a relative low sequence divergence (< 0.03). LFY2int2-N showed the highest proportion of informative sites (38/562, 6.8%, Table 4) which was similar to results in Neillia and Stephanandra (7.4%) . Low sequence divergence of multiple DNA regions suggests rapid radiation during divergence  and this has been hypothesized for many Pyreae taxa (the former Maloideae taxa) [26, 40]. This may also explain the poor resolution of the gene trees. Another contribution to the poor resolution in this study is the conflicting signals caused by recombinants. Recombinants are derived from two homologous chromosomes in one genome during meiosis (genic recombinants) or PCR (artifacts), leading to incorrect phylogenetic inferences [78–80]. As predicted by statistical principles, we found that putative recombinants formed separated clades. Recombinants represent substitutions of two distinct lineages, and thus receive no bootstrap support from either of the lineages in a cladistic phylogeny (data not shown). In this study, most putative recombinants represent one of the intra-individual polymorphic copies and were excluded from analyses. The four LFY2int2-N copies of P. xerophila all displayed the characteristics of recombinants, and formed separate clades in the tree (Figure 6). Polymorphic LFY2int2-N copies in P. xerophila may all be ancient genetic recombinants that arose by interspecific hybridization involving both oriental and occidental species. More individuals of P. xerophila and occidental species are necessary to confirm this hypothesis and investigate the origin of this species.
Phylogenetic utility of the introns
Among three Adh1-1 subparalogs, Adh1-1c was orthologous and recovered in all accessions, but it resulted in a poorly resolved phylogeny due to lack of informative sites and possible lineage sorting (Table 3). This makes it inadequate for the phylogenic reconstruction of Pyrus. The two introns of Adh1-2, Adh2-1 and Adh2-2, were not sequenced and analyzed in the current study, and it is unknown whether multiple subparalogs also exist for these paralogs. Adh2-2 might not exist in Pyrus species as discussed above. The phylogenetic utility of the introns of Adh2-1 and Adh1-2 needs to be estimated, which will require primers designed specifically to the paralogs.
LFY1int2 was not suitable for studying interspecific relationships due to sorting of LFY1int2-a and LFY1int2-b paralogs and the unclear origin of LFY1int2-c. In contrast, LFY2int2-N showed the highest sequence divergence, resulting in the best-resolved tree. Inparalogs of LFY2int2-Ins8 and LFY2int2-Del2, as well as the LFY2int2-S of unclear origin, could be easily identified and removed from phylogenetic inferences. Most importantly, relationships based on LFY2int2-N were congruent to previous studies based on morphological and molecular marker data. Conflicting placement of species may be resolved by using LFY2int2-N. It provides reliable evidence of ancient hybridization, since incomplete lineage sorting was not imputed for LFY2int2-N. Phylogenetic studies of Pyrus based on nuclear gene regions have been rare. Only the ITS region has been applied to a wide range of East Asian Pyrus species, but it resulted in a poorly resolved tree . One study based on the 18S gene focused only on two species, P. pyrifolia and P. communis . We conclude that LFY2int2-N is currently the most useful nuclear gene region for phylogenetic inference in Pyrus. It is as yet unknown whether additional inparalogs representing LFY2int2 will be found by analyzing more occidental species and individuals of oriental species.
This is the first study that explores LCNGs for phylogenetic analyses in Pyrus. It is also the first to document the gene structures and transcription of Adh homologs in the Rosaceae taxa. We demonstrated that frequent gene duplications contributed to complex outparalogs and inparalogs of Adh genes with functional diversification or nonfunctionalization. Paralogs, lineage sorting of alleles, and recombinants are three major problems when applying LCNGs in plant phylogenetic analyses. One ortholog of LEAFY, LFY2int2-N, is currently the best nuclear marker for studying interspecific relationships of Pyrus. Complex reticulate histories likely complicate the phylogenetic reconstruction of some Pyrus species. To better resolve interspecific relationships and examine the evolutionary processes of Pyrus, we are extending our phylogenetic studies with plastid DNA and nuclear DNA, including LFYint2-N, and by sampling a wider assortment of species and individuals.
Taxon sampling, DNA extraction, primer design and amplification
Twenty-five accessions from 13 oriental species and four occidental species of genus Pyrus were included. Six accessions of four Malus species were used as outgroups (Table 1). Total genomic DNA was isolated from fresh leaf tissue using a modified sodium dodecyl sulfate (SDS) method [12, 82].
To date, a complete cDNA sequence representing Adh1 from 'Granny Smith' (M. domestica, Z48234) and two 5' partial (beginning at the 3' end of exon 2) cDNA sequences representing two distinct Adh loci (AF031900-Adh1, AF031899-Adh2) from 'Packham's Triumph' (P. communis) are available. The Adh series (Adh1 and Adh2) were named randomly and do not correspond to previously named alleles. A forward primer (Adh-F1) based on sequence of Z48243 and three downstream primers (Adh1-R1, Adh1-R2 and Adh2-R) based on AF031900-Adh1 and AF031899-Adh2 were designed to obtain the entire gene region in several accessions including P. communis, 'Flemish Beauty' (P. communis), 'Nanguoli' (P. ussuriensis), 'Cuiguan' (P. pyrifolia), 'Korlaxiangli' (P. sinkiangensis), 'Ralls' (M. domestica), M. rockii and M. domestica subsp. chinensis. However, these primer pairs only succeeded in amplifying 12 Adh1 sequences in select accessions. Therefore, two additional forward primers, Adh1-F2 and Adh2-F, were designed based on AF031900-Adh1 and AF031899-Adh2, respectively, targeting a partial Adh region lacking exon 1 and intron 1. These primers amplified Adh2 sequences and additional Adh1 sequences. All of the above PCR products were designated as long partial genomic Adh sequences (G-PCR). Considering labor costs and difficulties in amplifying and sequencing fragments greater than 2 kb, a smaller region covering only introns 2 and 3 (about 650 bp) of Adh1 (reduced Adh1) was used in all accessions to construct a phylogeny.
For LEAFY, a long partial region of LFY1 and LFY2 spanning exon 2 and intron 2 was first amplified in some accessions using the primer pair 'LFY-F+LFY-R' developed in an exonic region of M. domestica 'Pinova' (DQ535885, DQ535886). After initial sequence analyses, specific primer pairs of 'LFY1-F + LFY1-R' and 'LFY2-F + LFY2-R' were developed to amplify partial intron 2 of LFY1 and LFY2 (LFY1int2, LFY2int2), respectively. This was done independently in all accessions except the three commercial cultivars ('Cuiguan', 'Ralls', 'Flemish Beauty' and 'Fuji'). Sequence information for primers used in this study is listed in Table 2 and their locations illustrated in Figure 1.
PCR was carried out in a final reaction volume of 50 μL, containing 10-20 ng total DNA, 2.5 mM MgCl2, 0.4 μM of each primer, 5% DMSO (v/v), 0.2 mM dNTP, 2 U Taq DNA polymerase (Takara Biotechnology Company Co., Ltd, Kyoto, Japan) and 1 × PCR buffer supplied by the manufacturer. Amplification of the long partial Adh region was performed for 4 min at 94°C, followed by 35 cycles of 40 s at 94°C, 40 s at 58°C, 2 min of 20 s at 72°C, and a final extension for 7 min at 72°C. For other shorter regions like locus-specific RT-PCR as described below, the PCR procedure was identical, but only 1 min was needed for the extension step.
Cloning and sequencing
PCR products were verified by 2% agarose gel electrophoresis, and the target bands were separated and purified using 3S spin DNA Agarose Gel Purification (Shenergy Biocolor, Shanghai, China). For long partial genomic Adh amplified by G-PCR, the purified PCR products were cloned using TA cloning kit Pmd19 (Takara) and more than three clones per sample were sequenced using M13+, M13- primers and internal primers located at exonic regions (Adh1-F5 for Adh1, and Adh2-F5 for Adh2, Table 2). For the reduced Adh1 region, five to ten clones per sample were sequenced using the M13+ primer. For the long partial LEAFY region, more than three clones were sequenced to obtain reads representing putative LFY1 and LFY2. For LFY1int2 and LFY2int2, the purified PCR products were directly sequenced by amplification primers. Additionally, more than three clones were sequenced to capture all the copies indicated by direct sequencing results.
Intronic and exonic boundaries were determined by comparison with available cDNA sequences and preservations of the 'GT' and 'AG' at two ends of introns. Sequences were aligned with Clustal X . Sequence divergence within and between different homologs was calculated using MEGA4  with gaps treated as pairwise deletions. Putative recombinants were detected using RDP3 software package , and some putative recombinants were identified manually.
Transcription of Adh homologs
The expression patterns of LFY1 and LFY2 have been well documented in Malus and Pyrus , but little was known about expression patterns of Adh1 and Adh2 in Pyrus. Therefore, 'Cuiguan' (P. pyrifolia), 'Nanguoli' (P. ussuriensis), 'Korlaxiangli' (P. sinkiangensis) and 'Flemish Beauty' (P. communis), representing four major pear cultivar groups together with 'Fuji' (M. domestica) were selected to examine transcription of Adh1 and Adh2. For 'Cuiguan' (P. pyrifolia), fresh young leaves, ripe fruits and seeds were collected in our campus yard for Adh expression analyses, while for the other accessions, only ripe fruits were used. The plant tissues were frozen in liquid nitrogen and stored at -80°C. Total RNA was isolated using a modified CTAB method. First strand cDNA was synthesized from 1.0 μg of total RNA using a poly (T)18 as primer and AMV reverse transcriptase (Bio Basic Inc, New York, USA) according to the manufacturer's instructions. To efficiently detect transcription of Adh1 and Adh2 independently, two specific primer pairs 'spAdh1-F+spAdh1-R' and 'spAdh2-F+spAdh2-R' (Table 2) targeting a shorter region were used for RT-PCR and specific genomic PCR (SG-PCR). PCR products were directly sequenced followed by cloning to identify copies involved in transcription. Actin was analyzed as a reference gene. The primers Pact-F and Pact-R were designed based on the Actin gene sequences from P. communis and 'Yali' (P. pyrifolia, CWP) (AB190176, GU830958) (Table 2).
The predicted amino acid sequences of long partial Adh genes in Pyrus and Malus from G-PCR were compared with those from other well-studied plant taxa by conducting NJ and MP analyses using PAUP 4.0b10 . For reduced Adh1, nucleotide sequences including the exonic and intronic regions were both included. LFY1int2 and LFY2int2 were analyzed separately, since their sequence homology was too low to be aligned. MP analyses were conducted using PAUP 4.0b10 with gaps treated as missing data. MP analyses were performed using a heuristic search with the TBR and Multree options. To estimate support for the clades, non-parametric bootstraps were estimated with 1000 replicates. Bayesian analyses were performed with MrBayes 3.1 . The best fitting substitution models for each dataset were determined with the Akaike Information Criterion (AIC) using ModelTest 3.06 . The AIC favored the HKY+G for the reduced Adh1 and the K81uf+Gfor both LFY1int2 and LFY2int2. Markov chains were run for 10,000,000 generations with sample frequency of 100. The average standard deviation of split frequency was 0.003 for reduced Adh1, 0.005 for LFY1int2 and 0.002 for LFY2int2, indicating the runs have reached convergence for each dataset. The first 25% of the trees were discarded as burn-in. Clade posterior probabilities were calculated from the combined sets of trees. Both MP and Bayesian analyses resulted in largely congruent tree topologies. Sequences included in the final phylogenetic analyses were deposited in GenBank (Accessions GU991401-991522, HM003976-004066, HQ912028-HQ912076). Alignments of these datasets are deposited as additional files 4, 5, 6 and 7.
This work has been financed by the project (No. 30871690) from the National Natural Science Foundation of China, Project (R307605) from Zhejiang Provincial Natural Science Foundation of China, and Project (No. 20090451480) of the China Postdoctoral Science Foundation.
- Potter D, Eriksson T, Evans RC, Oh S, Smedmark JEE, Morgan DR, Kerr M, Robertson KR, Arsenault M, Dickinson TA, Campbell CS: Phylogeny and classification of Rosaceae. Pl Syst Evol. 2007, 266: 5-43. 10.1007/s00606-007-0539-9.View Article
- Bailey LH: Pyrus. Standard cyclopedia of horticulture. 1917, New York: Macmillan, 5: 2865-2878.
- Kikuchi A: Horticulture of fruit trees. 1948, Tokyo: Yokendo, (in Japanese)
- Kikuchi A: Assessment of Chinese pear species and cultivars. Collec Rec Hort Res Fac Agr Kyoto Univ. 1946, 3: 1-11. (in Japanese)
- Yu T: Taxonomy of the fruit tree in China. 1979, Beijing: China Agriculture Press, (in Chinese)
- Yu T, Kuan K: Taxa nava Rosacearum (1). Acta Phytotaxon Sinica. 1963, 8: 202-236. (in Chinese with English summary)
- Aldasoro JJ, Aedo C, Garmendia FM: The genus Pyrus L. (Rosaceae) in south-west Europe and North Africa. Bot J Linn Soc. 1996, 121: 143-158.
- Challice JS, Westwood MN: Numerical taxonomic studies of the genus Pyrus using both chemical and botanical characters. Bot J Linn Soc. 1973, 67: 121-148. 10.1111/j.1095-8339.1973.tb01734.x.View Article
- Iketani H, Manabe T, Matsuta N, Akihama T, Hayashi T: Incongruence between RFLPs of chloroplast DNA and morphological classification in east Asian pear (Pyrus spp.). Genet Resour Crop Evol. 1998, 45: 533-539. 10.1023/A:1008646016181.View Article
- Monte-Corve L, Cabrita L, Oliveira C, Leitao J: Assessment of genetic relationships among Pyrus species and cultivars using AFLP and RAPD markers. Genet Resour Crop Evol. 2000, 47: 257-265. 10.1023/A:1008794809807.View Article
- Teng Y, Tanabe K, Tamura F, Itai A: Genetic relationships of pear cultivars in Xinjiang, China, as measured by RAPD markers. J Hort Sci Biotech. 2001, 76: 771-779.
- Teng Y, Tanabe K, Tamura F, Itai A: Genetic relationships of Pyrus species and cultivars native to East Asia revealed by randomly amplified polymorphic DNA markers. J Amer Soc Hort Sci. 2002, 127: 262-270.
- Yamamoto T, Kimura T, Sawamura Y, Manabe T, Kotobuki K, Hayashi T, Ban Y, Matsuta N: Simple sequence repeats for genetic analysis in pear. Euphytica. 2002, 124: 129-137. 10.1023/A:1015677505602.View Article
- Bao L, Chen K, Zhang D, Cao Y, Yamamoto T, Teng Y: Genetic diversity and similarity of pear (Pyrus L.) cultivars native to East Asia revealed by SSR (simple sequence repeat) markers. Genet Resour Crop Evol. 2007, 54: 959-971. 10.1007/s10722-006-9152-y.View Article
- Bassil N, Postman JD: Identification of European and Asian pears using EST-SSRs from Pyrus. Genet Resour Crop Evol. 2010, 57: 357-370. 10.1007/s10722-009-9474-7.View Article
- Yao L, Zheng X, Cai D, Gao Y, Wang K, Cao Y, Teng Y: Exploitation of Malus EST-SSRs and the utility in evaluation of genetic diversity in Malus and Pyrus. Genet Resour Crop Evol. 2010, 57: 841-851. 10.1007/s10722-009-9524-1.View Article
- Bao L, Chen K, Zhang D, Li X, Teng Y: An assessment of genetic variability and relationships within Asian pears based on AFLP (amplified fragment length polymorphism) markers. Sci Hort. 2008, 116: 374-380. 10.1016/j.scienta.2008.02.008.View Article
- Morgan DR, Solits DE, Robertson KR: Systematics and evolutionary implications of rbcL sequence variation in Rosaceae. Amer J Bot. 1994, 81: 890-903. 10.2307/2445770.View Article
- Campbell CS, Donoghue MJ, Baldwin BG, Wojciechowski MF: Phylogenetic relationships in Maloideae (Rosaceae), evidence from sequences of the internal transcribed spacer of nuclear ribosomal DNA and ITS congruence with morphology. Amer J Bot. 1995, 82: 903-918. 10.2307/2445977.View Article
- Alice LA, Campbell CS: Phylogeny of Rubus based on nuclear ribosomal DNA internal transcribed spacer region sequences. Amer J Bot. 1999, 86: 81-97. 10.2307/2656957.View Article
- Robinson JP, Harris SA, Juniper BE: Taxonomy of the genus Malus Mill. (Rosaceae) with emphasis on the cultivated apple, Malus domestica Borkh. Plant Syst Evol. 2001, 226: 35-58. 10.1007/s006060170072.View Article
- Potter D, Gao F, Bortiri E, Oh S, Baggett S: Phylogenetic relationships in Rosaceae inferred from chloroplast matK and trnL-trnF nucleotide sequence data. Plant Syst Evol. 2002, 231: 77-89. 10.1007/s006060200012.View Article
- Eriksson T, Hibbs M, Yoder AD, Delwiche CF, Donoghue MJ: The phylogeny of Rosoideae (Rosaceae) based on sequences of the internal transcribed spacer (ITS) of nuclear ribosomal DNA and the trnL-F region of chloroplast DNA. Int J Plant Sci. 2003, 164: 197-211. 10.1086/346163.View Article
- Wissemann V, Ritz CM: The genus Rosa (Rosoideae, Rosaceae) revisited: molecular analysis of nrITS-1 and atpB-rbc L intergenic spacer (IGS) versus conventional taxonomy. Bot J Linn Soc. 2005, 147: 275-290. 10.1111/j.1095-8339.2005.00368.x.View Article
- Ohta S, Yamamoto T, Nishitani C, Katsuki T, Iketani H, Omura M: Phylogenetic relationships among Japanese flowering cherries (Prunus subgenus Cerasus) based on nucleotide sequences of chloroplast DNA. Plant Syst Evol. 2007, 263: 209-225. 10.1007/s00606-006-0474-1.View Article
- Campbell CS, Evans RC, Morgan DR, Dickinson TA, Arsenault MP: Phylogeny of subtribe Pyrinae (formerly the Maloideae, Rosaceae): Limited resolution of a complex evolutionary history. Plant Syst Evol. 2007, 266: 119-145. 10.1007/s00606-007-0545-y.View Article
- Iwata H, Kato T, Ohno S: Triparental origin of Damask roses. Gene. 2000, 259: 53-59. 10.1016/S0378-1119(00)00487-X.View ArticlePubMed
- Zheng X, Cai D, Yao L, Teng Y: Non-concerted ITS evolution, early origin and phylogenetic utility of ITS pseudogenes in Pyrus. Mol Phylogenet Evol. 2008, 48: 892-903. 10.1016/j.ympev.2008.05.039.View ArticlePubMed
- Kimura T, Iketani H, Kotobuki K, Matsuta N, Ban Y, Hayashi T, Yamamoto T: Genetic characterization of pear varieties revealed by chloroplast DNA sequences. J Hort Sci Biotech. 2003, 78: 241-247.
- Bailey CD, Doyle JJ: Potential phylogenetic utility of the low-copy nuclear gene pistillata in Dicotyledonous Plants: Comparison to nrDNA ITS and trnL intron in Sphaerocardamum and other Brassicaceae. Mol Phylogenet Evol. 1999, 13: 20-30. 10.1006/mpev.1999.0627.View ArticlePubMed
- Sang T: Utility of low-copy nuclear gene sequences in plant phylogenetics. Crit Rev Biochem Mol Biol. 2002, 37: 121-147. 10.1080/10409230290771474.View ArticlePubMed
- Evans RC, Campbell CS: The origin of the apple subfamily (Maloideae; Rosaceae) is clarified by DNA sequence data from duplicated GBSSI genes. Amer J Bot. 2002, 89: 1478-1484. 10.3732/ajb.89.9.1478.View Article
- Mort ME, Crawford DJ: The continuing search: low-copy nuclear sequences for lower-level plant molecular phylogenetic studies. Taxon. 2004, 53: 257-261. 10.2307/4135604.View Article
- Ané C, Larget B, Baum DA, Smith SD, Rokas A: A Bayesian estimation of concordance among gene trees. Mol Biol Evol. 2007, 24: 412-426.View ArticlePubMed
- Lawton-Rauh A: Evolutionary dynamics of duplicated genes in plants. Mol Phylogenet Evol. 2003, 29: 396-409. 10.1016/j.ympev.2003.07.004.View ArticlePubMed
- Sonnhammer ELL, Koonin EV: Orthology, paralogy and proposed classification for paralog subtypes. Trends Genet. 2002, 18: 619-620. 10.1016/S0168-9525(02)02793-2.View ArticlePubMed
- Meng C, Kubatko LS: Detecting hybrid speciation in the presence of incomplete lineage sorting using gene tree incongruence: A model. Theor Popul Biol. 2009, 75: 35-45. 10.1016/j.tpb.2008.10.004.View ArticlePubMed
- Evans RC, Alice LA, Campbell CS, Kellogg EA, Dickinson TA: The granule-bound starch synthase (GBSSI) gene in the Rosaceae: multiple loci and phylogenetic utility. Mol Phylogenet Evol. 2000, 17: 388-400. 10.1006/mpev.2000.0828.View ArticlePubMed
- Oh SH, Potter D: Phylogenetic utility of the second intron of LEAFY in Neillia and Stephanandra (Rosaceae) and implications for the origin of Stephanandra. Mol Phylogenet Evol. 2003, 29: 203-215. 10.1016/S1055-7903(03)00093-9.View ArticlePubMed
- Lo EYY, Stefanović S, Christensen KI, Dickinson TA: Evidence for genetic association between East Asian and Western North American Crataegus L. (Rosaceae) and rapid divergence of the Eastern North American lineages based on multiple DNA sequences. Mol Phylogenet Evol. 2009, 51: 157-168. 10.1016/j.ympev.2009.01.018.View ArticlePubMed
- Esumi T, Tao R, Yonemori K: Isolation of LEAFY and TERMINAL FLOWER 1 homologs from six fruit tree species in the subfamily Maloideae of the Rosaceae. Sex Plant Reprod. 2005, 17: 277-287. 10.1007/s00497-004-0239-3.View Article
- Shafqat J, El-Ahmad M, Danielsson O, Martínez MC, Persson B, Parés X, Jornvall H: Pea formaldehyde-active class III alcohol dehydrogenase: common derivation of the plant and animal forms but not of the corresponding ethanol-active forms (classes I and P). Proc Natl Acad Sci USA. 1996, 93: 5595-5599. 10.1073/pnas.93.11.5595.View ArticlePubMedPubMed Central
- Martinez MC, Achkor H, Persson B, Shafqat J, Farrés J, Jörnvall H, Parés X: Arabidopsis formaldehyde dehydrogenase. Molecular properties of plant class III alcohol dehydrogenase provide further insights into the origins, structure and function of plant class p and liver class I alcohol dehydrogenases. Eur J Biochem. 1996, 241: 849-857. 10.1111/j.1432-1033.1996.00849.x.View ArticlePubMed
- Gaut BS, Peek AS, Morton BR, Clegg MT: Patterns of genetic diversification within the Adh gene family in the grasses (Poaceae). Mol Biol Evol. 1999, 16: 1086-1097.View ArticlePubMed
- Petersen G, Seberg O, Aagesen L, Frederiksen S: An empirical test of the treatment of indels during optimization alignment based on the phylogeny of the genus Secale (Poaceae). Mol Phylogenet Evol. 2004, 30: 733-742. 10.1016/S1055-7903(03)00206-9.View ArticlePubMed
- Thompson CE, Salzano FM, de-Souza ON, Freitas LB: Sequence and structural aspects of the functional diversification of plant alcohol dehydrogenases. Gene. 2007, 396: 108-115. 10.1016/j.gene.2007.02.016.View ArticlePubMed
- Small RL, Wendel JF: Copy number lability and evolutionary dynamics of the Adh gene family in diploid and tetraploid cotton. Genetics. 2000, 155: 1913-1926.PubMedPubMed Central
- Roalson EH, Friar EA: Phylogenetic analysis of the nuclear alcohol dehydrogenase (Adh) gene family in Carex section Acrocystis (Cyperaceae) and combined analyses of Adh and nuclear ribosomal ITS and ETS sequences for inferring species relationships. Mol Phylogenet Evol. 2004, 33: 671-686. 10.1016/j.ympev.2004.08.005.View ArticlePubMed
- Chervin C, Truett J, Speirs J: Alcohol dehydrogenase expression and the production of alcohols during pear fruit ripening. J Amer Soc Hort Sci. 1999, 124: 71-75.
- Wolyn DJ, Jelenkovic G: Nucleotide sequence of an alcohol dehydrogenase gene in octoploid strawberry (Fragaria × ananassa Duch.). Plant Mol Biol. 1990, 14: 855-857. 10.1007/BF00016518.View ArticlePubMed
- Korban SS, Skirvin RM: Nonmenclature of the cultivated apple. HortScience. 1984, 19: 177-180.
- Rubstov GA: Geographical distribution of the genus Pyrus and trends and factors in its evolution. Amer Nat. 1944, 78: 358-366. 10.1086/281206.View Article
- Perry DJ, Furnier GR: Pinus banksiana has at least seven expressed alcohol dehydrogenase genes in two linked groups. Proc Natl Acad Sci USA. 1996, 93: 13020-13023. 10.1073/pnas.93.23.13020.View ArticlePubMedPubMed Central
- Martin DP, Lemey P, Lott M, Moulton V, Posada D, Lefeuvre P: RDP3: a flexible and fast computer program for analyzing recombination. Bioinformatics. 2010, 26: 2462-2463. 10.1093/bioinformatics/btq467.View ArticlePubMedPubMed Central
- Cohn VH, Thompson MA, Moore GP: Nucleotide sequence comparison of the Adh gene in three drosophilids. J Mol Evol. 1984, 20: 31-37. 10.1007/BF02101983.View ArticlePubMed
- Kimura M: The neutral theory of molecular evolution. 1983, Cambridge, Cambridge University PressView Article
- Dennis ES, Sach MM, Gerlach WL, Finnegan EJ, Peacock WJ: Molecular analysis of the alcohol dehydrogenase 2 (Adh2) gene in maize. Nucleic Acids Res. 1985, 13: 727-743. 10.1093/nar/13.3.727.View ArticlePubMedPubMed Central
- Tesnié C, Verriès C: Molecular cloning and expression of cDNAs encoding alcohol dehydrogenase from Vitis vinifera L. during berry development. Plant Science. 2000, 157: 77-88. 10.1016/S0168-9452(00)00274-0.View Article
- Bartley IM, Hindley SJ: Alcohol dehydrogenases of apple. J Exper Bot. 1980, 31: 449-459. 10.1093/jxb/31.2.449.View Article
- Bartley IM, Stoker PG, Martin ADE, Hatfield SGS, Knee M: Synthesis of aroma compounds by apples supplied with alcohols and methyl esters of fatty acids. J Sci Food Agric. 1985, 36: 567-574. 10.1002/jsfa.2740360708.View Article
- Dixon J, Heweit EW: Factors affecting apple aroma/flavour volatile concentration: a review. New Zealand J Crop Hort Sci. 2000, 28: 155-173. 10.1080/01140671.2000.9514136.View Article
- Chervin C, Speirs J, Loveys B, Patterson BD: Influence of low oxygen storage on aroma compounds of whole pears and crushed pear flesh. Postharvest Biol Technol. 2000, 19: 279-285. 10.1016/S0925-5214(00)00096-X.View Article
- Chang C, Meyerowitz EM: Molecular cloning and DNA sequence of the Arabidopsis thaliana alcohol dehydrogenase gene. Proc Natl Acad Sci USA. 1986, 83: 1408-1412. 10.1073/pnas.83.5.1408.View ArticlePubMedPubMed Central
- Miyashita NT, Innan H, Terauchi R: Intra- and interspecific variation of the alcohol dehydrogenase locus region in wild plants Arabis gemmifera and Arabidopsis thaliana. Mol Biol Evol. 1996, 13: 433-436.View ArticlePubMed
- Singh RK, Sane VA, Misra A, Ali SA, Nath P: Differential expression of the mango alcohol dehydrogenase gene family during ripening. Phytochemistry. 2010, 71: 1485-1494. 10.1016/j.phytochem.2010.05.024.View ArticlePubMed
- Charlesworth D, Liu FL, Zhang L: The evolution of the alcohol dehydrogenase gene family by loss of introns in plants of the genus Leavenworthia (Brassicaceae). Mol Biol Evol. 1998, 15: 552-559.View ArticlePubMed
- Hu K: Intron exclusion and the mystery of intron loss. FEBS Letters. 2006, 580: 6361-6365. 10.1016/j.febslet.2006.10.048.View ArticlePubMed
- Luo J, Yoshikawa N, Hodoson MC, Hall BD: Duplication and paralog sorting of RPB2 and RPB1 genes in core eudicots. Mol Phylogenet Evol. 2007, 44: 850-862. 10.1016/j.ympev.2006.11.020.View ArticlePubMed
- Sang T, Donoghue MJ, Zhang D: Evolution of alcohol dehydrogenase genes in Peonies (Paeonia) phylogenetic relationships of putative non hybrid species. Mol Biol Evol. 1997, 14: 994-1007.View ArticlePubMed
- Fukuda T, Yokoyama J, Nakamura T, Song I, Ito T, Ochiai T, Kanno A, Kameya T, Maki M: Molecular phylogeny and evolution of alcohol dehydrogenase (Adh) genes in legumes. BMC Plant Biol. 2005, 5: 6-10.1186/1471-2229-5-6.View ArticlePubMedPubMed Central
- Wendel JF: Genome evolution in polyploids. Plant Mol Biol. 2000, 42: 225-249. 10.1023/A:1006392424384.View ArticlePubMed
- Soltis DE, Albert VA, Leebens-Mack J, Bell CD, Patterson AH, Zheng C, Sankoff D, DePamphilis CW, Wall PK, Soltis PS: Polyploidy and angiosperm diversification. Amer J Bot. 2009, 96: 336-348. 10.3732/ajb.0800079.View Article
- Blázquez MA, Soowal LN, Lee I, Weigel D: LEAFY expression and flower initiation in Arabidopsis. Development. 1997, 124: 3835-3844.PubMed
- Wada M, Ureshino A, Cao Q, Bessho H: Genomic varieties of apple AFL genes. Plant Sci. 2007, 193: 559-566.View Article
- Wada M, Cao Q, Kotoda N, Soejima J, Masuda T: Apple has two orthologues of FLORICAULA/LEAFY involved in flowering. Plant Mol Biol. 2002, 49: 567-577. 10.1023/A:1015544207121.View ArticlePubMed
- Fitch WM: Distinguishing homologous from analogous proteins. Syst Zool. 1970, 19: 99-113. 10.2307/2412448.View ArticlePubMed
- Calviño CI, Martínez SG, Downie SR: The evolutionary history of Eryngium (Apiaceae, Saniculoideae): Rapid radiations, long distance dispersals, and hybridizations. Mol Phylogenet Evol. 2008, 46: 1129-1150. 10.1016/j.ympev.2007.10.021.View ArticlePubMed
- Posada D, Crandall KA: Evaluation of methods for detecting recombination from DNA sequences: Computer simulations. Proc Natl Acad Sci USA. 2001, 98: 13757-13762. 10.1073/pnas.241370698.View ArticlePubMedPubMed Central
- Zhang D, Hewitt GM: Nuclear DNA analyses in genetic studies of populations: practice, problems and prospects. Mol Ecol. 2003, 12: 563-584. 10.1046/j.1365-294X.2003.01773.x.View ArticlePubMed
- Poke FS, Martin DP, Vaillancourt RE, Reid JB: The impact of intragenic recombination on phylogenetic reconstruction at the sectional level in Eucalyptus when using a single copy nuclear gene (cinnamoyl CoA reductase). Mol Phylogenet Evol. 2006, 39: 160-170. 10.1016/j.ympev.2005.11.016.View ArticlePubMed
- Kim CS, Lee CH, Park KW, Kang SJ, Shin IS, Lee GP: Phylogenetic relationships among Pyrus pyrifolia and P. communis detected randomly amplified polymorphic DNA (RAPD) and conserved rDNA sequences. Sci Hort. 2005, 106: 491-501. 10.1016/j.scienta.2005.04.009.View Article
- Dellaporta SL, Wood J, Hicks JB: A plant DNA minipreparation, Version II. Plant Mol Biol Rpt. 1983, 1: 19-21. 10.1007/BF02712670.View Article
- Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997, 25: 4876-4882. 10.1093/nar/25.24.4876.View ArticlePubMedPubMed Central
- Tamura K, Dudley J, Nei M, Kumar S: MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol. 2007, 24: 1596-1599. 10.1093/molbev/msm092.View ArticlePubMed
- Swofford DL: PAUP*: Phylogenetic analysis using parsimony (and other methods) 4.0 beta. 2002, Sunderland, Massachusetts: Sinauer Associates
- Ronquist F, Huelsenbeck JP: MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003, 19: 1572-1574. 10.1093/bioinformatics/btg180.View ArticlePubMed
- Posada D, Crandall KA: Modeltest: testing the model of DNA substitution. Bioinformatics. 1998, 14: 817-818. 10.1093/bioinformatics/14.9.817.View ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.