| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics and Phytogeography |
2 Ohio State University Herbarium, Department of Evolution, Ecology and Organismal Biology, 1315 Kinnear Road, Columbus, Ohio 43212 USA 3 Department of Biological Sciences, Kent State University, Kent, Ohio 44242 USA
Received for publication 8 September 2007. Accepted for publication 11 January 2008.
ABSTRACT
Corallorhizinae are a small group of Old and New World temperate orchids of which a core monophyletic group comprises Govenia, Cremastra, Aplectrum, Oreorchis and the leafless Corallorhiza, and which according to phylogenetic analysis of nuclear ITS and plastid matK sequences, are related in this way: (Govenia (Cremastra (Aplectrum (Oreorchis (Corallorhiza))))). This hypothesis is consistent with the progressive deletion of the trnK intron and matK ORF. Frameshift-resulting indels yield a predicted loss of translation for the critical "domain X" region of matK and are evidence that matK is a probable pseudogene in Aplectrum, Oreorchis, and Corallorhiza. Within Corallorhiza, a previous hypothesis based on plastid DNA restriction site analysis is confirmed, with the thickened-labellum C. striata group being sister to the thin-labellum remainder of the genus, within which the circumboreal C. trifida is sister to the remainder, which then comprise two further sister groups: C. maculata + C. bulbosa + C. mertensiana and C. odontorhiza + C. wisteriana. A close relationship between C. striata and the recently described Appalachian C. bentleyi is shown; in particular, C. bentleyi is more closely allied to a southern Mexican population of C. striata than it is to northern North American C. striata populations, suggesting that two lineages, each with Mexican and northern North American populations, exist within the C. striata group.
Key Words: Corallorhiza • matK • molecular evolution • Orchidaceae • plastid DNA
Orchidaceae are the largest family of flowering plants and have the greatest number of independent derivations of the leafless, heterotrophic habit of any angiosperm family, numbering at least 10 (Dressler, 1993
; Molvray et al., 2000). The fungal associations believed to occur with all orchids make them preadapted for this shift. The largest temperate genus of leafless putative heterotrophs in the family is Corallorhiza Gagnebin, comprising 11 species that are confined to the northern hemisphere of the New World except for one, C. trifida, which is circumboreal. Corallorhiza is placed in subtribe Corallorhizinae (Calypsoeae; Chase et al., 2003) with several other genera of temperate terrestrials; the composition of the subtribe or its equivalent has varied throughout the history of orchid classification (cf. Freudenstein, 1994b). Dressler (1981) included Aplectrum, Corallorhiza, Cremastra, Dactylostalix, Didiciea, Ephippianthus, Govenia, Oreorchis, and Tipularia in this subtribe and distinguished it from Calypsoeae, which comprised Calypso and Yoania. Subsequently (Dressler, 1993), he merged the two groups but excluded Govenia in part on the basis of its different root velamen type (Cymbidium rather than Calanthe type). Govenia occurs from Mexico to South America, which is a somewhat anomalous distribution for this subtribe, the other members of which are predominantly North American-Eurasian. Freudenstein (1994b) analyzed morphological features for the group and resolved two principal clades—one comprising Aplectrum, Corallorhiza, Cremastra, and Oreorchis (the "Corallorhiza clade"), and another comprising Calypso, Changnienia, Tipularia, and Yoania (the "Calypso clade"). There was no resolution among the genera of the Corallorhiza clade, whose members share the unusual feature of a "winter leaf" (except for the leafless Corallorhiza). In these genera, the leaf emerges in the autumn after leaves have fallen from trees and persists through the winter, withering in spring at or before flowering. Govenia was resolved as part of the sister group to these two clades. More recently, a subfamily-wide analysis of molecular data confirmed the monophyly of the Corallorhiza clade and indicated that Govenia is sister to it, but left the relationship of the Calypso clade to this group less certain due to weak branch support (J. Freudenstein, unpublished data).
Freudenstein (1997) monographed Corallorhiza, and Freudenstein and Doyle (1994) and Freudenstein (1994a) analyzed morphological characters and restriction site variation in plastid DNA to produce a cladogram of species relationships for the genus. Although the pattern was well resolved, they were unable to find a molecular synapomorphy for Corallorhiza, and the only morphological synapomorphies were loss characters (leaves and roots), leaving open the possibility that the genus is not monophyletic.
Much morphological variation in Corallorhiza has been described in North America over the last 200 years, resulting in a proliferation of named taxa, most of which are not now recognized. Nonetheless, significant variation is still occasionally being uncovered. Freudenstein (1999a) named C. bentleyi from a single population in West Virginia; additional populations in West Virginia and Virginia have since been found. Corallorhiza bentleyi is similar to C. striata in its floral structure, but the flowers of the former are much smaller than those of eastern North American C. striata var. striata and are most similar to those from small-flowered populations of the species from southern Mexico, which are known as C. striata var. involuta. Corallorhiza bentleyi also flowers much later than C. striata (late July as opposed to early June), also resembling the southern Mexican populations.
Leafless, putatively achlorophyllous angiosperms such as Corallorhiza often experience deletions in their plastid genomes, and in some cases this can be severe (e.g., Epifagus; Wolfe et al., 1992). Freudenstein and Doyle (1994) found much less deletion in the plastid genome of Corallorhiza than in other leafless angiosperms that have been examined. Their restriction site mapping showed that the species that were most affected were in the C. striata and C. maculata groups. DNA sequencing allows a much more precise investigation of molecular changes. The chloroplast genome locus matK has been used extensively in angiosperm phylogenetic reconstruction because of its relatively high substitution rate (Wolfe, 1991) in combination with a conserved overall structure. Beyond its phylogenetic utility, the locus is interesting because it resides in a group II intron in the gene for tRNA-lysine. Its substitution rates and evolutionary properties are unusual for a coding locus (Wolfe, 1991) and bear further investigation. Leafless taxa provide a unique system within which to examine the properties and changes in this gene.
The objectives of this study were to analyze DNA sequence data from the nuclear and plastid genomes to (1) resolve relationships among the genera of the Corallorhiza clade, (2) test the relationships among the species of Corallorhiza that were resolved based on plastid DNA restriction site data, (3) investigate the relationships of the newly described C. bentleyi, and (4) further investigate changes in the plastid genome in Corallorhiza by focusing on matK.
MATERIALS AND METHODS
The taxa examined are listed in Appendix 1. DNA was isolated from fresh or frozen leaf/inflorescence material by the CTAB method of Doyle and Doyle (1987). The matK and ITS sequences were generated by standard PCR methods, including amplification with Taq polymerase under conditions given in Goldman et al. (2001) and Freudenstein (1999b). To obtain sequences that included portions of the trnK gene and the entire intron, primers 2R and 3914F that are located in the trnK exons were used for amplification and also as sequencing primers (Goldman et al., 2001); additional sequencing primers for matK are given in Goldman et al. (2001). Sequencing was performed using the ABI Prism cycle sequencing (Applied Biosystems, Foster City, California, USA) or Pharmacia Autocycle sequencing kits (Pharmacia Biotech, Piscataway, New Jersey, USA) according to the manufacturer's directions, except that 1/4 reactions were used for the ABI. Reactions were run on a Pharmacia ALFexpress automated sequencer using Cy-5 labeled sequencing primers or on an ABI 3100 sequencer using the BigDye terminator reaction mix following the manufacturer's protocols. Sequences were checked and contigs assembled in Sequencher (Gene Codes, Ann Arbor, Michigan, USA) and aligned by eye for matK and with ClustalX (Thompson et al., 1997) followed by manual adjustment for ITS; we followed the recommendations of Simmons and Ochoterena (2000) and Freudenstein and Chase (2001) for coding simple gap characters. Cases with nested indels were coded as multistate unordered characters. Sequences were deposited in GenBank (Appendix 1). Outgroups (Coelia and Calypso) were chosen based on the results of previous higher level analyses (Freudenstein, 1994b; J. Freudenstein, unpublished data).
Full sequences comprising the 3' end of the trnK 5' exon, the trnK 5' intron, matK, and the trnK 3' intron were generated for all accessions of Corallorhiza, Oreorchis, Aplectrum, Govenia, and for Cremastra appendiculata because these were important for examining the length changes that have occurred in the trnK locus or for providing closely related undeleted sequences for comparison. For Cremastra unguiculata, Calypso, and Coelia, which did not experience any length changes and were relevant only to the phylogenetic objectives of the study, only the matK-coding sequence was generated. Predicted protein translations were generated with the program ExPASy (http://us.expasy.org/tools/dna.html).
Equally weighted parsimony analysis was performed on the data sets individually and combined using the program TNT (Goloboff et al., 2003). Tree search was performed using the XMULT command, which employs sectorial search, tree fusing, and tree drifting (Goloboff, 1999). Data sets were run with and without indel characters in order to make comparisons with the model-based results. The ragged ends of sequences were excluded from the analysis. Branch lengths generated by TNT are minimum possible lengths for each branch. Jackknife support (Farris et al., 1996) was determined using 5000 replications of a thorough search strategy employing XMULT for tree search and BBREAK for generating the maximal pool of most parsimonious trees. Individual probability of deletion of 37% was assigned to each character. Uncorrected genetic distances were calculated with the program PAUP* (Swofford, 2002) for matK using only the portion of the ORF that was present in all taxa.
Maximum likelihood (ML) analysis was employed using the program RAxML (Stamatakis, 2006). The ITS, matK, and combined analyses were run with indel characters removed. Each data set was analyzed 1000 times using randomized parsimony trees generated by the program and subsequently rearranged using the GTRMIX option, which employs a GTR + 
model. For the combined data set, two partitions were specified to allow matK and ITS to be modeled individually. The best likelihood tree was kept and compared with the parsimony results.
RESULTS
matK
Alignment of the trnK region, including matK, was straightforward; the region that was sequenced is shown in Fig. 1. Six indels were coded for the matK ORF. Analysis of the full trnK/matK matrix gave a single most parsimonious tree (CI = 0.87, RI = 0.93, length = 213; Fig. 2). The first clade within the ingroup comprises Cremastra and Govenia, supported at 93%. Among these, the two species of Cremastra are very strongly united (100%). The next branch along the main spine of the tree is Aplectrum, which is then sister to Oreorchis + Corallorhiza.
|
|
The matK sequences had significant deletions among the members of this group. Most striking are the 5' deletions in Aplectrum, Oreorchis, and Corallorhiza (Fig. 1). Cremastra appendiculata has an ORF of 1554 bp, and Govenia has 1557 bp. Aplectrum has a deletion of 299 bp at the 5' end of the ORF (yielding a length of 1235 bp), while Oreorchis and all Corallorhiza species have a further 84-bp deletion (giving ca. 1152 bp total). The deletion extends well beyond the ORF into the 5' intron. In all of these taxa, the trnK 5'-coding sequence is present, as is at least a portion of the 5' end of the trnK intron—387 bp in Aplectrum and 76 bp in Oreorchis and all of the Corallorhiza accessions. This deletion pattern is in contrast to the full intron of 964 bp for Cremastra appendiculata and 1037 bp in Govenia (Fig. 1). The 3' trn K intron is much smaller, approximately 300 bp and was not sequenced completely here, but shows little length variation in the major portion sequenced among Govenia, Cremastra, Aplectrum, Oreorchis, and Corallorhiza.
There are also internal deletions and insertions in matK among these genera, which were coded for the phylogenetic analysis. Govenia has a single 3-bp insertion, the only indel detected for a species outside the Aplectrum-Corallorhiza-Oreorchis clade. Aplectrum shares a 7-bp insertion with Oreorchis and Corallorhiza, as well as a 13-bp deletion that has undergone further loss to become a 19-bp deletion in Oreorchis and Corallorhiza. Oreorchis and all of the accessions of Corallorhiza share another 7-bp deletion. The C. trifida-maculata-wisteriana clade is united by a 9-bp deletion. All the C. striata group accessions (including C. bentleyi) share a 1-bp deletion. Finally, the Oaxaca accession of C. striata var. involuta has a 9-bp deletion near the 3' end of the gene.
Examination of the translated matK sequences reveals additional insights. Translation of the outgroup sequences for Coelia and Calypso and the ingroup Govenia and Cremastra sequences gives predicted amino acid sequences that align easily with other angiosperm matK translations. In these taxa, matK represents a single ORF ending in a termination codon. Translation of matK in Aplectrum, Oreorchis, and Corallorhiza yields short reading frames because of frameshift-causing indels and major deletions. The largest of these reading frames corresponds to 241 amino acids (in Aplectrum), and it is not in the same frame as Govenia and Cremastra. A 56 amino acid frame is present in all taxa included except for C. bentleyi, C. striata (Oaxaca), and C. bulbosa, which have a stop codon 10 amino acids into the region and C. maculata var. maculata, which has a stop codon 15 amino acids into the region. No reading frame yields more than 
10 amino acids of "domain X" as defined by Mohr et al. (1993). All these stop codons are the result of G 
T mutations that result in TAA codons.
The genetic distances for matK show some trends (Table 1). Uncorrected pairwise distances between Govenia and the other ingroup taxa range between 0.028 and 0.047. The smallest distances are between Govenia and the Cremastra species (0.028, 0.033). The distance increases to 0.035 with Aplectrum and to 0.037 and beyond with Oreorchis and Corallorhiza. The C. striata-bentleyi accessions had increased distances; among those, C. striata (Oaxaca) and C. bentleyi had the greatest distances (both 0.047).
|
|
The combined ITS + matK data set yielded a single most parsimonious tree (CI = 0.77, RI = 0.85, length = 510; Fig. 4) that provides a well-supported pattern for most resolved nodes. The ML tree was identical to the parsimony tree and to that resulting from the analysis without indels. Because the patterns from the matK and ITS data sets largely agreed, the combined tree simply reflects their patterns for the most part. However, the disagreement concerning the placement of Cremastra is here resolved in a novel way, with Cremastra as sister to Aplectrum + Oreorchis + Corallorhiza. The only clade in which relationships remain unresolved is the C. maculata group.
|
Phylogenetic relationships within Corallorhizinae
The only point of disagreement between the two data sets presented here concerns the placement of the two species of Cremastra. A sister relationship between Cremastra and Govenia, such as indicated by the matK data, has not been proposed before, but was quite strongly supported (93%), as was the contrasting ITS pattern. This discrepancy could be a case of true signal difference among the data sets as a result of past hybridization or lineage sorting. Even more strongly supported by matK is the sister relation between Aplectrum and Oreorchis + Corallorhiza, while the ITS data strongly suggest rather that Aplectrum is sister to Cremastra + Oreorchis + Corallorhiza, with Govenia sister to this whole assemblage. The pattern of major deletions at the 5' end of the matK ORF and trnK intron may have some bearing on choice among these patterns. While the topology from matK is most consistent with progressive deletion in this region, from no deletion in Cremastra to minimal deletion in Aplectrum to maximal deletion in Oreorchis and Corallorhiza, the ITS pattern suggests either independent deletions in Aplectrum and Oreorchis + Corallorhiza, or a seemingly unlikely regain of the deleted DNA in Cremastra. The pattern from the combined analysis, which is taken to be the strongest hypothesis here, is consistent with progressive matK deletion.
The morphological analysis of Freudenstein (1994b) did not resolve the relationships among the genera of the Corallorhiza clade, but it did suggest a synapomorphy for Cremastra + Aplectrum + Oreorchis + Corallorhiza—the hamulus type of pollinium stalk (Rasmussen, 1982), which is uncommon among orchids overall. On the ITS and combined trees, the hamulus is a unique synapomorphy for these genera, while on the matK tree the hamulus would either have been derived twice or have transformed subsequently to the tegula type of stalk seen in Govenia. Such a transformation between these two types of pollinium stalk has not been proposed before in orchids, given the very different structure of the two types of stalks.
The combined molecular tree is consistent with the current taxonomy that is widely accepted for these genera. Maekawa (1971) considered Cremastra and Aplectrum to be congeneric and transferred both species of Cremastra to Aplectrum, but the present results indicate that this combined group would not be monophyletic, thus arguing for continued recognition of the two genera.
The pattern of geographical distribution of the genera of Corallorhizinae, when considered in the context of their phylogenetic relationships, is one of alternate Old and New World ranges, with Govenia occurring from Mexico to Bolivia, Cremastra in temperate east Asia, Aplectrum in eastern North America, Oreorchis in temperate east Asia, and Corallorhiza predominantly in North America (but having one circumboreal species, C. trifida, which does not fall at the base of the genus). The eastern North American–east Asian disjunct distribution pattern is well known and has been explained by fragmentation of a formerly more continuous temperate flora (see review in Wen, 1999). This scenario would be expected to lead to vicariant pairs, as have been described in many groups, including orchids (Boufford and Spongberg, 1983; Chen, 1983; Xiang et al., 1998b). None of the relationships depicted in the current study can be explained solely by vicariance, except perhaps for Oreorchis-Corallorhiza. This lack of correspondence to a vicariance paradigm could be due to extinction of members of vicariant pairs, such as possibly an Asian sister to Aplectrum, or to dispersal.
Both the ITS and matK patterns support the cladogram of relationships among Corallorhiza species based on restriction site data from the large single copy region of the plastid genome presented by Freudenstein and Doyle (1994). Two primary clades are distinguished: the C. striata clade, with flowers lacking a mentum (small nectar spur) and having a thickened, boat-shaped labellum with a fused pair of lamellae at its base; and the C. trifida-maculata-wisteriana clade, with a mentum of varying size at the summit of the ovary and a thin, flattened labellum that has two free lamellae at its base.
The least well-known species of Corallorhiza included here, C. bentleyi, has not been examined from a molecular perspective before, having only been discovered in the late 1990s. Freudenstein (1999a) described the species from West Virginia as a clear ally of C. striata, sharing with it the boat-shaped, thickened labellum with fused basal lamellae, and absence of a mentum. The flowers of this species are distinct from those of C. striata from northeastern North America because those of C. bentleyi are much smaller. They are most similar to the small-flowered plants of C. striata from southern Mexico that have been called C. involuta (recognized by Freudenstein [1997] as C. striata var. involuta). The molecular evidence from both loci supports this relationship, associating the population of C. bentleyi examined here with a particular population of C. striata var. involuta from Oaxaca on a relatively long branch (27 total changes), while the Hidalgo accession of var. involuta grouped with the large-flowered C. striata var. striata and the medium-flowered C. striata var. vreelandii from the southwestern USA. This pattern suggests that two distinct lineages may exist in the C. striata group, each with members in Mexico and northern North America. If we were to optimize the range on the C. striata-bentleyi portion of the cladogram in Fig. 4 and considered the southwestern USA and Mexico to be one region, that southern region would be plesiomorphic and the northern North American regions would be apomorphic, representing independent gains in each clade. However, no strong conclusions about biogeography can be drawn with such sparse sampling. Freudenstein and Doyle (1994) found that the Mexican accessions of C. maculata were successive sisters to the northern North American accessions. If this is the case for the C. striata group as well, it may indicate a general pattern for the genus, and perhaps for other groups that have a montane Mexican–temperate northern North American distribution. A possible explanation for such a pattern would be northward migration following glacial retreat.
Evolution of matK
matK is unusual among putatively coding genes in its relatively rapid overall substitution rate, the nearly equal substitution rates across positions in a codon, similar transition–transversion rates, and its low synonymous to nonsynonymous amino acid substitution rate ratio (due to higher nonsynonymous rate; Wolfe, 1991; Steele and Vilgalys, 1994; Xiang et al., 1998a). These properties are often associated with pseudogenes and some authors have suggested that matK might be a pseudogene in orchids and other angiosperms (Kores et al., 2000; Whitten et al., 2000; Cameron et al., 2001; Goldman et al., 2001). Li (1997, p.182) defined pseudogenes as "DNA sequences that were derived from functional genes but have been rendered nonfunctional by mutations that prevent their proper expression" and further added that they "are subject to no functional constraints." While it is true that pseudogenes usually have those properties detailed for matK, and while those properties are unusual for a coding locus, they are not sufficient to render a locus nonfunctional. Pseudogenes typically experience frameshift mutations that cause radical amino acid change and often premature stop codons, leading to the nonfunctionality stated by Li. For example, Wolfe and dePamphilis (1997) found frameshift indels and premature stop codons in rbc L in two species of Orobanche that also showed an increase in the nonsynonymous substitution rate. Although some studies have reported nontriplet indels for matK (e.g., Kores et al., 2000), in most cases in angiosperms matK indels occur in multiples of three bases (Soltis and Soltis, 1998), which is consistent with production of a protein product. Hilu and Alice (1999) reported frameshift mutations at the extreme 3' end of matK in grasses, but they are so close to the 3' terminus that they should not interfere with any function that the locus might have. The fact that matK sequences are alignable across angiosperms (Hilu et al., 2003) and within families such as Orchidaceae, suggests that most probably there is functional constraint. Moreover, Barthet and Hilu (2007) detected matK transcripts and probable full protein products for the maturase in rice and potato, suggesting, along with previous work (Vogel et al., 1999), probable functionality in at least some plants.
The changes described here in Corallorhiza matK sequences are the most extreme of any documented for angiosperms and would support the assertion that if matK is a pseudogene in any angiosperm, it is one in Aplectrum, Oreorchis, and Corallorhiza. Typical matK ORFs are approximately 1500–1600 bp (Soltis and Soltis, 1998; Hilu et al., 2003); the gene in Epifagus is reduced to 1320 bp due to deletions in the 5' end of the gene (Wolfe et al., 1992). Other leafless species of Orobanchaceae do not have unusually small matK ORFs, nor do they have nontriplet indels (Young et al., 1999). The Corallorhiza and Oreorchis matK sequences examined here are 1152 bp, also due to 5' deletion, and are thus the smallest matK sequences reported to date. Although Epifagus has significant 5' deletion, its changes are not so extensive that the reading frame for the "domain X" region at the 3' end is affected. This region was indicated to be most important for the DNA-binding function of matK by Mohr et al. (1993), although Young and dePamphilis (2000) did not observe a higher level of constraint on this region than on the rest of the gene. In Calypso, Govenia, and the two Cremastra s, a single large ORF is present for matK, as in most other angiosperms. In contrast, internal length changes have altered the reading frames in such a way that there is no large reading frame in Aplectrum, Oreorchis, and Corallorhiza, because of the presence of stop codons. In all of these, only smaller potential ORFs are present, and these are not in the same frame as those in Cremastra and Govenia, meaning that the amino acid sequence produced would be radically different in Aplectrum, Oreorchis, and Corallorhiza and that, unless a phenomenon such as RNA editing (cf. Gott and Emeson, 2000) is in operation here, even the majority of the "domain X" portion of the gene cannot be correctly translated. RNA editing in plants is limited, as far as currently known, to C U and U C changes, with no addition or removal of bases having been detected (Gray, 2003; Tillich et al., 2006), so it is unlikely that this phenomenon could restore the reading frame in these taxa. Hence, these may be the first carefully documented matK pseudogenes.
Because of the broader history of matK and its antecedents, the evolution of the matK ORF cannot be dissociated from changes in the trnK intron in which it resides (Chuang and Hu, 2004; Hausner et al, 2006). Chloroplast genome-encoded group II introns, such as that in trnK, have self-splicing ability (Michel et al., 1989), but may require chaperoning from a maturase such as the product of matK (Hess et al., 1994). The fact that matK does exist rarely as a freestanding ORF, without associated trnK gene, as in Epifagus and Adiantum capillis-veneris (Wolf et al., 2003), suggests that this function may be broader than just assisting splicing of the trnK intron (Ems et al., 1995). In this study we did not sequence the entire trnK exons, but the 3' portion of the 5' exon that we did sequence was present and highly conserved in all accessions. The trnK intron shows quite a different pattern, however. Although large in Govenia and Cremastra (>900 bp), the 5' portion of the trnK intron is drastically reduced in Aplectrum, Oreorchis, and Corallorhiza, paralleling the changes seen in matK.
Although it is tempting to try to associate the sequence changes observed in matK in Corallorhizinae with the loss of leaves, the situation is not straightforward. First, because the large 5' deletions and frameshift indels occur in the leaf-bearing Aplectrum and Oreorchis as well as in Corallorhiza, they precede the loss of leaves (phylogenetically). This pattern could indicate that changes in the plastid genome foreshadow physical leaf loss in the shift to increased heterotrophy. However, the role of matK and its known variation in other plants need to be considered as well because it is not at all clear that the gene should be expected to be nonfunctional in nonphotosynthetic plants, given that its presumed maturase role is not necessarily related to photosynthesis. Accordingly, Wolfe et al. (1992) found that matK is one of the few genes remaining in the plastome of the leafless direct (haustorial) parasite Epifagus, even though the trnK gene in which it normally resides has been lost. Epifagus retains this ORF in spite of its plastid genome being much more deleted overall than the plastid genome of Corallorhiza (dePamphilis and Palmer, 1990; Freudenstein and Doyle, 1994). It is clear that, if indeed related to increased heterotrophy, the changes seen in Corallorhiza and allies in matK indicate a very different situation with respect to its role than in Epifagus.
Although a normal, functional matK product could not be produced in Aplectrum, Oreorchis, or Corallorhiza, the rate of base substitution, as assessed by genetic distance, is not very different in these taxa with respect to Govenia than it is in Cremastra. This raises the question of why, in the most convincing case where matK could be postulated to be a pseudogene, the rate has not increased substantially and sequence conservation seems to remain the rule. Perhaps it is just that the group is still in the early stages of the shift to heterotrophy and that the various taxa provide windows into positions on a continuum of change. While the rate of substitution for matK is elevated overall somewhat in the C. striata clade, it is only in C. striata var.involuta (JVF 2155) and C. bentleyi that we begin to see a notable increase in substitution rate. This pattern of substitution rate increase in the C. striata clade also holds for ITS, which is consistent with the observation of Freudenstein and Doyle (1994) that the greatest degree of plastid DNA deletion in Corallorhiza was among the C. striata accessions, particularly in C. striata var. involuta. This range of substitution rates within Corallorhiza, in combination with the molecular changes seen even in related leafy taxa, argues for the importance of this group as a natural system in the early stages of plastid genome change, from which we can continue to learn about the transformation from autotrophy to heterotrophy in plants.
Appendix 1. List of taxa examined, voucher specimens, origin, and GenBank numbers. All vouchers are at BH unless otherwise specified.
|
FOOTNOTES
1 The authors thank K. Inoue and M. W. Chase for providing plant material and DNA, respectively, E. Greenwood and G. Salazar for field assistance, and C. Barrett and A. Chaudhuri for technical assistance. This project was supported by NSF grant DEB-0415920 to J.V.F. 
4 Author for correspondence (e-mail: freudenstein.1{at}osu.edu)
LITERATURE CITED
Barthet, M. M., AND K. W. Hilu. 2007. Expression of matK: Functional and evolutionary implications. American Journal of Botany 94: 1402–1412.
Boufford, D. E., AND S. A. Spongberg. 1983. Eastern Asian–eastern North American phytogeographical relationships—a history from the time of Linnaeus to the twentieth century. Annals of the Missouri Botanical Garden 70: 423–439.[CrossRef][Web of Science]
Cameron, K. C., M. W. Chase, W. R. Anderson, AND H. G. Hills. 2001. Molecular systematics of Malpighiaceae: Evidence from plastid rbcL and matK sequences. American Journal of Botany 88: 1847–1862.
Chase, M. W., K. M. Cameron, R. L. Barrett, AND J. V. Freudenstein. 2003. DNA data and Orchidaceae systematics: A new phylogenetic classification. In K. M. Dixon, S. P. Kell, R. L. Barrett, and P. J. Cribb [eds.], Orchid conservation, 69-89. Natural History Publications, Kota Kinabalu, Sabah, Malaysia.
Chen, S.-C. 1983. A comparison of orchid floras of temperate North America and eastern Asia. Annals of the Missouri Botanical Garden 70: 713–723.[CrossRef][Web of Science]
Chuang, S.-L., AND J.-M. Hu. 2004. The evolution of chloroplast mat K genes, including identification of new homologues from Ophioglossum petiolatum and two lycophytes. Taiwania 49: 273–287.
dePamphilis, C. W., AND J. D. Palmer. 1990. Loss of photosynthetic and chlororespiratory genes from the plastid genome of a parasitic flowering plant. Nature 348: 337–339.[CrossRef][Medline]
Doyle, J. J., AND J. L. Doyle. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.
Dressler, R. L. 1993. Phylogeny and classification of the orchid family. Cambridge University Press, Cambridge, UK.
Ems, S. C., C. W. Morden, C. K. Dixon, K. H. Wolfe, C. W. dePamphilis, AND J. D. Palmer. 1995. Transcription, splicing and editing of plastid RNAs in the nonphotosynthetic plant Epifagus virginiana. Plant Molecular Biology 29: 721–733.[CrossRef][Web of Science][Medline]
Farris, J. S., V. A. Albert, M. Källersjö, D. L. Lipscomb, AND A. G. Kluge. 1996. Parsimony jackknifing outperforms neighbor-joining. Cladistics 12: 99–124.[CrossRef][Web of Science]
Freudenstein, J. V. 1994a. Character transformation and relationships in Corallorhiza (Orchidaceae: Epidendroideae). II. Structural features and phylogenetic analysis. American Journal of Botany 81: 1458–1467.[CrossRef][Web of Science]
Freudenstein, J. V. 1994b. Gynostemium structure and relationships of the Corallorhizinae (Orchidaceae: Epidendroideae). Plant Systematics and Evolution 193: 1–19.[CrossRef][Web of Science]
Freudenstein, J. V. 1997. A monograph of Corallorhiza. Harvard Papers in Botany 10: 5–52.[Medline]
Freudenstein, J. V. 1999a. A new species of Corallorhiza from West Virginia, USA. Novon 9: 511–513.[CrossRef][Web of Science]
Freudenstein, J. V. 1999b. Relationships and character transformation in Pyroloideae (Ericaceae) based on ITS sequences, morphology, and development. Systematic Botany 24: 398–408.[CrossRef][Web of Science]
Freudenstein, J. V., AND M. W. Chase. 2001. Analysis of mitochondrial nad 1b-c intron sequences in Orchidaceae: Utility and coding of length-change characters. Systematic Botany 26: 643–657.[Web of Science]
Freudenstein, J. V., AND J. J. Doyle. 1994. Character transformation and relationships in Corallorhiza (Orchidaceae: Epidendroideae). I. Plastid DNA. American Journal of Botany 81: 1449–1457.[CrossRef][Web of Science]
Goldman, D. H., J. V. Freudenstein, P. J. Kores, M. Molvray, D. C. Jarrell, W. M. Whitten, K. M. Cameron, AND M. W. Chase. 2001. Phylogenetics of Arethuseae (Orchidaceae) based on plastid matK and rbcL sequences. Systematic Botany 26: 670–695.[Web of Science]
Goloboff, P. 1999. Analyzing large data sets in reasonable times: Solutions for composite optima. Cladistics 15: 415–428.[CrossRef][Web of Science]
Goloboff, P., J. Farris, AND K. Nixon. 2003. TNT: Tree analysis using new technology. Program and documentation available from the authors and at website http://www.zmuc.dk/public/phylogeny/TNT/.
Gott, J. M., AND R. B. Emeson. 2000. Functions and mechanisms of RNA editing. Annual Review of Genetics 34: 499–531.[CrossRef][Web of Science][Medline]
Gray, M. W. 2003. Diversity and evolution of mitochondrial RNA editing systems. International Union of Biochemistry and Molecular Biology Life 55: 227–233.[CrossRef]
Hausner, G., R. Olson, D. Simon, I. Johnson, E. R. Sanders, K. G. Karol, R. M. McCourt, AND S. Zimmerly. 2006. Origin and evolution of the chloroplast trnK (matK) intron: A model for evolution of group II intron RNA structures. Molecular Biology and Evolution 23: 380–391.
Hess, W. R., B. Hoch, P. Zeltz, T. Hübschmann, H. Kössel, AND T. Börner. 1994. Inefficient rpl2 splicing inn barley mutants with ribosome-deficient plastids. Plant Cell 6: 1455–1465.[Abstract]
Hilu, K. W., AND L. A. Alice. 1999. Evolutionary implications of matK indels in Poaceae. American Journal of Botany 86: 1735–1741.
Hilu, K. W., T. Borsch, K. Müller, D. E. Soltis, P. S. Soltis, V. Savolainen, M. W. Chase, M. P. Powell, L. A. Alice, R. Evans, H. Sauquest, C. Neinhuis, T. A. B. Slotta, J. G. Rohwer, C. S. Campbell, AND L. W. Chatrou. 2003. Angiosperm phylogeny based on matK sequence information. American Journal of Botany 90: 1758–1776.
Kores, P. J., P. Weston, M. Molvray, AND M. W. Chase. 2000. Phylogenetic relationships within Diurideae:Inferences from plastid matK DNA sequences. In K. L. Wilson, and D. A. Morrison [eds.], Monocots: Systematics and evolution, 449–456. CSIRO Publishing, Collingwood, Victoria, Australia.
Li, W.-H. 1997. Molecular evolution. Sinauer, Sunderland, MA.
Maekawa, F. 1971. The wild orchids of Japan in colour. Seibundo-shinkousha, Tokyo, Japan.
Michel, F., K. Umesono, AND H. Ozeki. 1989. Comparative and functional anatomy of group II catalytic introns—a review. Gene 82: 5–30.[CrossRef][Web of Science][Medline]
Mohr, G., P. S. Perlman, AND A. M. Lambowitz. 1993. Evolutionary relationships among group II intron-encoded proteins and identification of a conserved domain that may be related to maturase function. Nucleic Acids Research 21: 4991–4997.
Molvray, M., P. J. Kores, AND M. W. Chase. 2000. Polyphyly of mycoheterotrophic orchids and functional influences on floral and molecular characters. In K. L. Wilson, and D. A. Morrison [eds.], Monocots: Systematics and evolution, 441-448. CSIRO Publishing, Collingwood, Victoria, Australia.
Rasmussen, F. N. 1982. The gynostemium of the neottioid orchids. Opera Botanica 65: 1–96.
Simmons, M. P., AND H. Ochoterena. 2000. Gaps as characters in sequence-based phylogenetic analyses. Systematic Biology 49: 369–381.[CrossRef][Web of Science][Medline]
Soltis, D. E., AND P. S. Soltis. 1998. Choosing an approach and an appropriate gene for phylogenetic analysis. In D. E. Soltis, P. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants II: DNA sequencing, 1–42. Kluwer, Boston, Massachusetts, USA.
Stamatakis, A. 2006. RaxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics (Oxford, England) 22: 2688–2690.[CrossRef]
Steele, K. P., AND R. Vilgalys. 1994. Phylogenetic analyses of Polemoniaceae using nucleotide sequences of the plastid gene matK. Systematic Botany 19: 126–142.[CrossRef][Web of Science]
Swofford, D. L. 2002. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4. 0b. Sinauer, Sunderland, MA, USA.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, AND D. G. Higgins. 1997. The Clustal_X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876–4882.
Tillich, M., P. Lehwark, B. R. Morton, AND U. G. Maier. 2006. The evolution of chloroplast DNA editing. Molecular Biology and Evolution 23: 1912–1921.
Vogel, J., T. Börner, AND W. R. Hess. 1999. Comparative analysis of splicing of the complete set of chloroplast group II introns in three higher plant mutants. Nucleic Acids Research 27: 3866–3874.
Wen, J. 1999. Evoluton of eastern Asian and eastern North American disjunction distributions in flowering plants. Annual Review of Ecology and Systematics 30: 421–455.[CrossRef][Web of Science]
Whitten, W. M., N. H. Williams, AND M. W. Chase. 2000. Subtribal and generic relationships of Maxillarieae (Orchidaceae) with emphasis on Stanhopeinae: Combined molecular evidence. American Journal of Botany 87: 1842–1856.
Wolf, P. G., C. A. Rowe, R. B. Sinclair, AND M. Hasebe. 2003. Complete nucleotide sequence of the chloroplast genome from a leptosporangiate fern, Adiantum capillus-veneris L. DNA Research 10: 59–65.[Abstract]
Wolfe, A. D., AND C. W. dePamphilis. 1997. Alternate paths of evolution for the photosynthetic gene rbcL in four nonphotosynthetic species of Orobanche. Plant Molecular Biology 33: 965–977.[CrossRef][Web of Science][Medline]
Wolfe, K. H. 1991. Protein-coding genes in chloroplast DNA: Compilation of nucleotide sequences, data base entries, and rates of molecular evolution. In L. Bogorad, and I. K. Vasil [eds.], The photosynthetic apparatus: Molecular biology and operation, vol. 7B, Cell culture and somatic cell genetics of plants, 467–482. Academic Press, San Diego, California, USA.
Wolfe, K. H., C. W. Morden, AND J. D. Palmer. 1992. Function and evolution of a minimal plastid genome from a nonphotosynthetic parasitic plant. Proceedings of the National Academy of Sciences, USA 89: 10648–10652.
Xiang, Q.-Y., D. E. Soltis, AND P. S. Soltis. 1998a. Phylogenetic relationships of Cornaceae and close relatives inferred from matK and rbcL sequences. American Journal of Botany 85: 285–297.[Abstract]
Xiang, Q.-Y., D. E. Soltis, AND P. S. Soltis. 1998b. The eastern Asian and eastern and wstern North American floristic disjunction: Congruent phylogenetic patterns in seven diverse genera. Molecular Phylogenetics and Evolution 10: 178–190.[CrossRef][Web of Science][Medline]
Young, N. D., AND C. W. dePamphilis. 2000. Purifying selection detected in the plastid gene matK and flanking ribozyme regions within a group II intron of nonphotosynthetic plants. Molecular Biology and Evolution 17: 1933–1941.
Young, N. D., K. E. Steiner, AND C. W. dePamphilis. 1999. The evolution of parasitism in Scrophulariaceae/Orobanchaceae: Plastid gene sequences refute an evolutionary transition series. Annals of the Missouri Botanical Garden 86: 876–893.[CrossRef][Web of Science]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
