| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics |
University of Iowa, Department of Biological Sciences & Center for Comparative Genomics, 239 Biology Building, Iowa City, Iowa 52242-1324 USA
Received for publication July 27, 2001. Accepted for publication January 24, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Carnegiea • chloroplast DNA • Lophocereus • mutualism • Neobuxbaumia • Pachycereus • pollination • Stenocereus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Herre, 1996
; Pellmyr et al., 1996
; Herre et al., 1999
). Phylogenies are currently being developed for the well-known yucca–yucca moth (Brown et al., 1994
; Pellmyr et al., 1996
; Pellmyr and Leebens-Mack, 1999
) and fig–fig wasp (Herre et al., 1996
; Machado et al., 1996
; Rasplus et al., 1998
; Weiblen, 2000
) systems to determine the extent of coevolution in these interactions. The phylogenetic origin of the recently discovered (Fleming and Holland, 1998
) senita–senita moth system, the third known obligate pollinating seed-eater mutualism involving active pollination, has yet to be examined.
Senita is a columnar cactus (genus Lophocereus, family Cactaceae) that inhabits the Sonoran Desert of northwestern Mexico and the southwestern United States (Lindsay, 1963
). Recent studies show that senita, unlike related columnar cacti that are visited by bats or birds, is night-pollinated by the moth Upiga virescens (family Pyralidae) (Fleming and Holland, 1998
; Holland and Fleming, 1999a
, b
). Each stage in the life cycle of U. virescens is intimately associated with the host plant. Adult senita moths (
1 cm long) rest during the day in the elongated, bristle-like spines characteristic of reproductively mature senita stems. Female moths actively pollinate and oviposit on the nocturnal white to pink flowers. The larvae feed within the developing fruit and tunnel into the stem to pupate before the damaged fruit abscise from the cactus. They overwinter within the stem, eclose, and emerge the following flowering season. The senita moth is thus both a pollinator and a seed consumer, much like yucca moths and fig wasps are on their particular hosts.
The genus Lophocereus presently includes two described species, the Baja California localized endemic L. gatesii and the more common and widespread L. schottii. The latter consists of three described varieties, L. schottii var. australis, L. schottii var. schottii, and L. schottii var. tenuis (Lindsay, 1963
). The molecular phylogenetic relationships among these varieties and between L. schottii and L. gatesii are not known. Moreover, there are currently two conflicting hypotheses concerning the evolutionary origin of Lophocereus within the North American columnar cacti (tribe Pachycereeae). In one scheme, Buxbaum (1961)
allied Lophocereus with the bat-pollinated Carnegiea gigantea in the Stenocereinae, a subtribe exhibiting floral adaptations to a variety of pollinators, including bats, birds, and insects. In the other scheme, Gibson and Horak (1978)
placed Lophocereus in the subtribe Pachycereinae, which is almost entirely bat pollinated. Furthermore, Gibson and Horak identified the hummingbird-pollinated Pachycereus marginatus as sister to Lophocereus, which would make the otherwise bat-pollinated genus Pachycereus a paraphyletic group. Although the inclusion of Lophocereus into Pachycereus' synonym has been adopted in more recent taxonomic treatments of the Cactaceae (e.g., Hunt and Taylor, 1990
; Barthlott and Hunt, 1993
; Anderson, 2001
), the exact phylogenetic position of Lophocereus within the tribe Pachycereeae has yet to be determined. The resolution of this issue has important implications not only for the classification of Lophocereus and Pachycereus, but also for understanding the transition to moth pollination in Lophocereus.
To establish the evolutionary origins of Lophocereus and its unusual pollination system, a phylogeny of the tribe and subtribe in which senita occurs is needed. Understanding the systematics of columnar cacti is complicated by the adaptive convergence of vegetative characters in response to xeric desert environments. Many morphological characters (such as leaves) are reduced to minimize transpirational water loss and show morphological homoplasy (Cota and Wallace, 1997
). Floral characters, too, may exhibit homoplasy as a result of adaptation to a common guild of floral visitors, such as bats, which are the predominant pollinators of many columnar cacti. Phylogenetic inference is further hindered by the absence of a good fossil record for the Cactaceae. As a consequence, the generic nomenclature of columnar cacti as well as the grouping of genera into tribes and subtribes has been revised extensively in the past and is still being resolved (Gibson et al., 1986
; Wallace, 1995
; Cornejo and Simpson, 1997
; Arias, Terrazas, and Cameron, 2001
).
The main goals of this study were to use molecular sequence data to investigate the relationships among taxa within the genus Lophocereus and to establish the sister group and phylogenetic origins of Lophocereus within the North American columnar cacti (Pachycereeae). Because molecular data are less likely than morphological characters to exhibit homoplasy in response to convergent environmental selection, sequence analysis is a particularly effective method for resolving higher taxonomic relationships among cacti (Wallace, 1995
; Cota and Wallace, 1997
). In this paper, we first analyze the varietal and sister species relationships within the genus Lophocereus. We then conduct a broader molecular phylogenetic examination of the origin of Lophocereus within the North American columnar cacti. The sister group relationship of Lophocereus to P. marginatus was first determined, leading to the analysis of an expanded set of taxa including all six described Pachycereus species. To test the paraphyly of Pachycereus and the monophyly of Pachycereus and Lophocereus, we also included Carnegiea gigantea and Neobuxbaumia tetetzo, two closely related species within the subtribe Pachycereinae (sensu Gibson and Horak 1978
). Stenocereus thurberi, a member of the subtribe Stenocereinae, was used as an outgroup based on the chloroplast restriction fragment length polymorphism (RFLP) analysis of Cota and Wallace (1997)
. The inclusion of these species allowed us to resolve conflicting hypotheses regarding the phylogenetic origin of Lophocereus and to speculate on the historical origin of its current pollination mutualism with Upiga virescens.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
delimited two species and three varieties of Lophocereus. Within L. schottii, the widespread variety L. schottii var. schottii is restricted to, but widely distributed within, the Sonoran Desert of northwestern Mexico and the southwestern United States. The two other varieties, L. schottii var. tenuis and L. schottii var. australis, are restricted to the coastal plain of southern Sonora/northern Sinaloa and the subtropical thornscrub of the Cape Region of southern Baja California, respectively. The second species, L. gatesii, is endemic to a small area of fog desert on the Pacific coast of southern Baja west of La Paz. More recent taxonomic treatments (e.g., Anderson, 2001
) recombine these two species within Pachycereus (but do not recognize the varieties of L. schottii). The phylogenetic distinctiveness and relationship among species and varieties within Lophocereus have yet to be examined using DNA sequence data.
Taxonomy of the genus Pachycereus
The classification of species in the genus Pachycereus within the subtribe Pachycereinae is contentious and has been modified extensively since the suggestion by Britton and Rose (1909)
to include arborescent cacti with trichome-covered flowers and bur-like fruits. Several species have been added or removed from the genus by various taxonomists who have used different sets of morphological characters for their studies, with only Pachycereus grandis, P. pecten-aboriginum, and P. pringlei consistently assigned to Pachycereus (see Gibson et al., 1986
, for review). Included in this study are the six Pachycereus species recognized by Gibson: P. pringlei, P. pecten-aboriginum, P. marginatus, P. weberi, P. hollianus, and P. grandis.
Taxonomy of the tribe Pachycereeae
Classification of cacti in the tribe Pachycereeae has changed extensively in the past (see Gibson et al., 1986
, for review). We summarize here the two predominant hypotheses regarding the phylogenetic relationships of North American columnar cacti. Buxbaum (1961)
conducted morphological and geographical analyses of cacti using as characters stem triterpenes, funicular pigments, and seed morphology. In his final phylogeny of genera within the Pachycereeae, the four subtribes Myrtillocactinae, Pachycereinae, Pterocereinae, and Stenocereinae were recognized. In contrast, based on analyses of stem chemistry, funiculus pigmentation, seed morphology, growth form, rib morphology and anatomy, flower morphology, and presence and levels of mucilage, Gibson and coworkers recognized only the two subtribes Stenocereinae and Pachycereinae in the Pachycereeae (Gibson and Horak, 1978
). In their treatment (as well as in more recent classifications), the Stenocereinae largely subsume Buxbaum's Myrtillocactinae, Pterocereinae, and Stenocereinae. Buxbaum (1961)
suggested that Lophocereus is closely related to the bat-pollinated Carnegiea gigantea in the subtribe Stenocereinae, a subtribe exhibiting floral adaptations to a variety of pollinators. Gibson and Horak (1978)
, in contrast, placed Lophocereus in the Pachycereinae, a subtribe that is almost entirely bat pollinated. The placement of Lophocereus within these subtribes has important implications for understanding the origins of its highly derived floral biology associated with the senita moth.
Tissue source and DNA extraction
To reconstruct phylogenies, we extracted DNA from seeds, flower buds, or stem sections of L. gatesii, L. schottii var. australis, L. schottii var. schottii, L. schottii var. tenuis, and Pachycereus pecten-aboriginum and from flower buds of P. pringlei, S. thurberi, and C. gigantea in the Sonoran Desert (Baja California and Sonora, Mexico). Seeds of P. grandis originated from La Noria (Puebla, Mexico). Voucher specimens for Lophocereus species and varieties, P. pecten-aboriginum, and P. grandis have been deposited in the Herbarium of the University of Iowa. Material from P. pringlei, S. thurberi, and C. gigantea is available upon request from the authors. Pachycereus marginatus and P. weberi seedlings were purchased from Mesa Garden (Belen, New Mexico, USA).
Total genomic DNA was extracted from seedlings, stem sections, or flower buds using a protocol based on the hexadecyltrimethylammonium bromide (CTAB) method (Doyle and Doyle, 1987
). Modifications were implemented to optimize DNA extraction from the polysaccharide-rich cactus tissues as described previously (Hartmann, Nason, and Bhattacharya, 2001
). For initial amplification of the trnC-trnD intergenic regions, we used purified chloroplast-DNA as template, which was isolated by CsCl gradient ultracentrifugation based on a protocol for extraction of chloroplast genomes of kelp (Fain, Druehl, and Baillie, 1988
). DNA preparations from P. hollianus and N. tetetzo were a gift from R. Wallace (Iowa State University) and were used directly as template for the polymerase chain reactions (PCRs).
Polymerase chain reaction and sequence analysis
The three chloroplast intergenic spacers, trnL-trnF, trnC-trnD, and trnS-trnfM, were PCR-amplified, sequenced, and used as phylogenetic markers for L. schottii var. schottii, L. schottii var. australis, and L. gatesii as well as for S. thurberi, C. gigantea, N. tetetzo, and the six Pachycereus species. The PCRs were carried out with either Taq polymerase (Fisher Scientific, Pittsburgh, Pennsylvania, USA) or Biolase DNA polymerase (Bioline, Randolph, Massachusetts, USA) in 50-µL reaction volumes. The template DNA was first denatured at 94°C for 10 min and then subjected to 35 cycles of the following PCR program: 94°C for 1 min, annealing for 1 min, extension at 72°C. Annealing temperatures and extension times were specific for each intergenic fragment as detailed below. The final cycles included an extension of 10 min at 72°C. The PCR products were gel-purified using glassmilk (Geneclean Kit, Qbiogene, Carlsbad, California, USA) according to the manufacturer's instructions and either directly sequenced or first cloned into pGEM-T (Promega, Madison, Wisconsin, USA). Sequencing reactions used a dye termination sequencing protocol and were carried out on a 373 A Fluorescent Automated Sequencer (Perkin Elmer-Applied Biosystems, Wellesley, Massachusetts, USA).
Primers "e" (5'-GGTTCAAGTCCCTCTTCCC-3') and "f" (5' ATTTGAACTGGTGACACGAG-3') (Taberlet et al., 1991
) were used to amplify the trnL-trnF intergenic region of approximately 450 base pairs (bp). The PCR cycle used an annealing temperature of 55°C and an extension time of 1 min. The PCR primer "f" was used as a sequencing primer for all taxa. Because we were unable to amplify the trnC-trnD intergenic region from total genomic DNA of senita, we initially used purified chloroplast DNA of L. s. schottii as template. Primers trnC (5'-CAAGTTCAAATCCGGGTGTC-3') and trnD (5'-GGGATTGTAGTTCAATTGGT-3') (Demesure, Sodzi, and Petit, 1995
) were used to amplify this region of approximately 3 kilobases (kb). The fragment was partially sequenced from both ends using the PCR primers as sequencing primers. Sequences were used to design internal primers CD-2F and CD-2R for subsequent PCR and sequencing reactions. Using primer trnC in combination with the internal primer CD-2R (5'-GGTTACCCTGAC-AAAATAC-3'), we were able to amplify about 2.5 kb of the trnC-trnD intergenic region from total genomic DNA of the remaining 11 taxa. Sequencing reactions were done using primers trnC and CD-2F (5'-GGATTTATTGAATTGC TTC-3'). Primers trnS (5'-GAGAGAGAGGGATTCGAACC-3') and trnfM (5'-CATAACCT-TGAGGTCACGGG-3') (Demesure, Sodzi, and Petit, 1995
) were used to amplifiy the trnS-trnfM intergenic region of approximate size 1.5 kb. The PCR cycle used an annealing temperature of 62°C and an extension time of 2 min. The trnfM PCR primer was used as a sequencing primer. Sequences determined in this study have been deposited in the EMBL/GenBank/DDBJ databases (see Table 1).
|
). Alignments for the individual cpDNA sequences were 435 bp for trnL-F, 1149 bp for trnC-D, and 375 bp for trnS-fM. The alignment of the concatenated sequences using all three markers for the 11 cactus taxa was 1960 bp in length. The cpDNA phylogeny was outgroup rooted with sequences from S. thurberi. We used the separate sequences as well as the concatenated sequences for phylogenetic reconstructions using the maximum parsimony, neighbor-joining, maximum likelihood, and quartet puzzling methods to infer a phylogeny. Gaps were included, but treated as missing data, and all phylogenetic methods were implemented using the computer program PAUP* (Swofford, 2001
). The unweighted maximum parsimony method was done using a heuristic search of ten replicates with random stepwise addition and tree bisection-reconnection (TBR) and the MulTrees option. One thousand bootstrap replicates were analyzed with this method. Hierarchical likelihood ratio tests were done to estimate the best-fit model for the cpDNA data set (MODELTEST v3.06; Posada and Crandall, 1998
). This procedure showed that the F81 (Felsenstein, 1981
) model incorporating the proportion of invariant sites (F81 + I) best fit our data. The F81 + I model was used to calculate the distance matrix for neighbor-joining tree-building. The bootstrap method (1000 replicates) was used to determine the support for nodes in this tree. The F81 + I model was then used to do the heuristic maximum likelihood (ML) analysis. Starting trees were obtained with stepwise addition (randomly drawn, ten rounds) and rearranged with tree bisection-reconnection. One hundred bootstrap replicates were analyzed using these settings. Quartet puzzling (10 000 steps) was done using the maximum likelihood optimality criterion of PAUP* under the F81 + I model (Swofford, 2001
).
Phylogenetic signal
To determine the structure of our data, we used the skewness of tree length distribution (g1 value) as an indication of strength of phylogenetic signal (Hillis and Huelsenbeck, 1992
). We completed an exhaustive search using the 1960 bp alignment of the concatenated sequences. The distribution of the lengths of the trees was significantly skewed with a g1 value of –1.110391 (P < 0.01), indicating that a strong nonrandom structure was present in our data matrix.
Constrained topologies
Because parts of our phylogeny are not consistent with existing taxonomic schemes (Buxbaum, 1961
; Gibson and Horak, 1978
), we wanted to test whether the topologies predicted by Buxbaum and by Gibson and Horak are significantly less likely than our best tree given the sequence data. We therefore constructed alternative tree topologies reflecting these two classifications using the computer program MacClade v3.04 (Maddison and Maddison, 1992
). The likelihoods of these two constrained phylogenies were compared to the likelihood of our best maximum likelihood tree using the Shimodaira-Hasegawa nonparametric bootstrap test (10 000 replications; Shimodaira and Hasegawa, 1999
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Phylogenetic origins of Lophocereus
The number of polymorphic, informative sites for alignments involving all 11 cactus species was 6 for trnT-L, 8 for trnC-D, 5 for trnS-fM, and 19 for the concatenated sequence. Using the combined data set for tree reconstructions, we obtained the same topology using maximum likelihood, neighbor-joining, and quartet puzzling methods (Fig. 1). In an exhaustive maximum parsimony search, two trees of equal lengths were found that differed only in the positions of N. tetetzo and P. hollianus, which were switched. Trees constructed using the three sequences separately were consistent as shown in Fig. 1, such as the monophyly of the clade containing P. pringlei, P. pecten-aboriginum, P. grandis, and C. gigantea, and the sister taxa relationship of Lophocereus and P. marginatus and of P. pecten-aboriginum and P. grandis. The concatenated sequences, however, gave higher resolution, reflected in an increase in bootstrap values for nodes in the tree, and we hereafter limit our discussion to the concatenated sequences.
|
) and Fig. 2C (according to Buxbaum, 1961
). Maximum likelihood values and results of the Shimodaira-Hasegawa test for these alternative topologies are given in Table 2. Our results showed that the likelihood of a constrained tree according to the classification scheme of Buxbaum (1961)
was lower than that of the tree with a topology suggested by Gibson and Horak (1978)
. Both constrained trees were, however, significantly "worse" than the unconstrained optimal tree shown in Figs. 1 and 2A.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). In contrast, we found 0/3000 polymorphic sites between members of the genus Lophocereus, and only 19/1960 (0.92%) informative polymorphic sites in the ten cactus species examined. The genomic regions we assayed have been used with other species to successfully resolve evolutionary relationships at the taxonomic levels investigated here (e.g., Mayer and Soltis, 1994
; Levy et al., 1996
; Dumolin-Lapegue et al., 1997
; Nicolosi et al., 2000
). For several species, these markers have even shown sufficient variation to permit studies of intraspecific diversity (e.g., Soltis et al., 1991
, 1992
; El Mousadik and Petit, 1995
; Maskas and Cruzan, 2000
; Mohanty, Martin, and Aguinagalde, 2000
). There are more recently diverged taxa, however, that exhibit levels of chloroplast sequence variation comparable to that found here. Small et al. (1998)
, for example, found that five tetraploid Gossypium species that diverged 0.5–2 million years ago contained only four informative nucleotide substitutions (0.05%) in over 7 kb of chloroplast sequences.
Comparative studies have shown that the chloroplast genome exhibits significant rate heterogeneity both among lineages and among different DNA regions (Gaut et al., 1992
). This may explain the low chloroplast sequence variation in Lophocereus and related columnar cacti relative to other plant taxa. The limited sequence variation could also be due to the relatively recent origin and radiation of the columnar cacti in North America. Within Lophocereus, sequence variation may have been further reduced as a result of genetic bottlenecks during periods of range expansion and contraction due to Pleistocene climatic change. Paleoecological analyses of plant remains preserved in packrat middens have been used to infer northward range expansion of senita from more southerly Pleistocene refugia within the last 2000–4000 yr (Peñalba and Van Devender, 1998
; Van Devender, in press
). Founding events coupled with the high coalescence rates of haploid genomes may, therefore, explain lack of variation within Lophocereus, especially in northern Baja California and Sonora.
The question remains whether the widespread L. schottii and the Baja endemic L. gatesii are in fact distinct species. If indeed they are, the lack of observed variation between their cpDNA could reflect late divergence of these taxa, slow chloroplast sequence evolution, and/or bottlenecks in the history of the genus. Alternatively, L. schottii and L. gatesii may constitute a single species. In this case, the observed morphological differences may reflect geographical variation within a single widespread species as opposed to the existence of evolutionarily independent lineages. Climatic effects on cactus morphology have previously been documented for Lophocereus (Felger and Lowe, 1967
) and for other columnar cacti in the Pachycereeae (Cornejo and Simpson, 1997
). More detailed genetic and ecological analyses of the transition zones between current taxonomic species and varieties will be required to resolve the phylogeny and systematics of the genus.
Phylogenetic analyses within tribe Pachycereeae
Relationships within the Pachycereinae
As shown in Fig. 1, our phylogeny strongly supported the sister relationship of Lophocereus and the hummingbird-pollinated P. marginatus. This is in agreement with Gibson and Horak (1978)
, who found that these two species share a very similar growth habit, external shoot morphology, and stem chemistry. In further agreement with Gibson and Horak (1978)
, we found that P. weberi is basal to the clade with Lophocereus and P. marginatus, and these three cacti are sister to a clade containing C. gigantea, P. pecten-aboriginum, P. pringlei, and P. grandis. Within this latter clade, however, our analyses indicate P. pringlei to be the basal species of Pachycereus, whereas P. pecten-aboriginum is basal to P. pringlei and P. grandis in the morphological phylogeny of Cornejo and Simpson (1997)
.
Constrained topologies
The positions of C. gigantea and N. tetetzo (or P. hollianus) in our phylogeny are not consistent with current taxonomic schemes (Buxbaum, 1961
; Gibson and Horak, 1978
; Hunt and Taylor, 1990
; Barthlott and Hunt, 1993
; Anderson, 2001
). For example, Gibson and Horak (1978)
placed Neobuxbaumia and Carnegiea in the subtribe Pachyereinae (Fig. 2B), but outside the Pachycereus phylad. In contrast, our data support the monophyly of C. gigantea and Pachycereus (and Lophocereus, Figs. 1, 2A). Thus, based on growth habit, vegetative morphology, and flower and fruit structure, Carnegiea appears to be more closely related to Neobuxbaumia, but when stem chemistry or cpDNA sequence data are used for comparison, Carnegiea appears to be more closely related to Pachycereus.
Cornejo and Simpson (1997)
converted Gibson and Horak's findings into a standard phylogenetic tree format (Fig. 2B), allowing us to test the likelihood of their phylogeny by forcing its topology onto our cpDNA data set. Constraining the phylogeny so that Carnegiea is positioned outside of Pachycereus and sister to Neobuxbaumia results in a significantly lower maximum likelihood value (see Table 1) relative to the topology in Fig. 1. Consistent with our analysis are the results of a study using cpDNA RFLP data (Cota and Wallace, 1997
), in which Carnegiea forms a separate clade with Lophocereus and P. marginatus, whereas Bergerocactus emoryi, Lemairocereus (Pachycereus) hollianus, and Neobuxbaumia euphorbioides form a second clade within the Pachycereinae.
In the arrangement shown in Fig. 2C, we test the phylogenetic relationship published by Buxbaum (1961)
. Buxbaum placed Lophocereus and Carnegiea as sister taxa into the subtribe Stenocereinae and classified P. marginatus and P. weberi within Stenocereus. All remaining Pachycereus species were placed into the subtribe Pachyereinae. We used Buxbaum's phylogeny as a reference for constructing this alternative tree, though we maintained the nomenclature/classification system of Gibson and Horak. Our analysis of the tree topology constructed using Buxbaum's classification (Fig. 2C) shows that the likelihood (Table 1) of this constrained tree was lower than that of the Gibson and Horak tree and significantly lower than that of the unconstrained optimal tree shown in Figs. 1 and 2A. We conclude, therefore, that with the exception of S. thurberi, all species examined here are best treated as members of the subtribe Pachycereinae. Moreover, Lophocereus is very closely related and probably sister to the hummingbird-pollinated P. marginatus as suggested by Gibson and Horak (1978)
, and not sister to the bat-pollinated Carnegiea, as suggested by Buxbaum (1961)
. The robust phylogenetic positions of Lophocereus and C. gigantea within the Pachycereus phylad shows Pachycereus to be paraphyletic.
The phylogenetic position of P. hollianus and N. tetetzo in our tree may suggest that the genus Pachycereus needs to be reorganized to include Lophocereus, Carnegiea, Neobuxbaumia, and perhaps other species, into a monophyletic group. An alternative possibility, which is also supported by Gibson (1982)
, is that P. hollianus might not be a member of the genus Pachycereus. According to Gibson (1982)
, P. hollianus is the least specialized member of the Pachycereus group, and he speculates that it may in fact be "distinctive enough to be classified to its own genus." Furthermore, Gibson suggests that Carnegiea is closely related to Neobuxbaumia and these, in turn, to Mitrocereus. Clearly, further research that includes additional taxa is required to define the limits of the genus Pachycereus and to gain a better understanding of the relationships among genera in the Pachycereinae.
Evolution of pollination systems in the Pachycereeae and the Pachycereus phylad
Selection pressures induced by interactions of flower and pollinator have lead to reciprocal adaptations often referred to as "pollination syndromes." Knowledge of these syndromes can provide a useful tool for predicting the suite of visiting pollinators. Nonetheless, several morphological and physiological characters important for attracting pollinators are shared among flowers pollinated by New World bats, birds, and hawkmoths. Shared traits include relatively large flowers and quantities of nectar and blossoms positioned outside of the foliage where they are exposed to hovering pollinators (Faegri and van der Pijl, 1971
). The flowers of night-blooming bat and hawkmoth-pollinated species often attract nectar-feeding birds at dusk and in the morning (e.g., C. gigantea and S. thurberi; Fleming, Tuttle, and Horner, 1996
; Fleming et al., 2001
). Similarly, day-flying hawkmoths may compete with hummingbirds for nectar at bird blossoms (Faegri and van der Pijl, 1971
). These observations underscore the potential for overlap in attraction among bat, hummingbird, and hawkmoth flowers and are consistent with frequent evolutionary transitions among these pollination systems (van Helverson, 1993
).
Bat pollination is widespread in the tribe Pachycereeae and many species produce flowers ideally adapted for bat pollination (Gibson and Horak, 1978
). In a recent survey, 42 of the 70 species in the Pachycereeae were found to be exclusively or primarily bat-pollinated (Valiente-Banuet et al., 1996
). Given that bat pollination is the ancestral condition, selection pressures have in several cases led to pollination by smaller insects, hummingbirds, and hawkmoths. These shifts are primarily in the subtribe Stenocereinae (Gibson, 1982
). As indicated in Fig. 1, and given that all species and varieties of Lophocereus are a single species, then Lophocereus and its sister P. marginatus are the only two cacti in the Pachycereus phylad of which we are aware with derived, non-bat-pollination syndromes. Pachycereus weberi, which is sister to Lophocereus and P. marginatus in our phylogeny, is nocturnal, bat pollinated, and has funnelform flowers that are 8–10 cm in length (Gibson and Horak, 1978
).
We suggest three possible scenarios that may have led to the evolution of moth pollination in Lophocereus and to hummingbird pollination in P. marginatus. In the first, the transition from a bat-pollinated common ancestor may have occurred independently in these two lineages: the lineage leading to P. marginatus evolved smaller, diurnal, hummingbird-pollinated flowers, whereas in the lineage leading to Lophocereus, flowers remained nocturnal but experienced a reduction in size and nectar production, apparently associated with adaptation to moth-pollination. The second possibility is the sequential evolution of derived pollination syndromes in Lophocereus and P. marginatus from a bat-pollinated common ancestor. Under this hypothesis, an initial transition to hummingbird pollination would have served as a preadaptation to moth pollination in Lophocereus. The evolution of reduced flower size, reduced nectar production, and diurnal anthesis may have been followed by a reversal to nocturnal anthesis. Hummingbirds are known to be significant pollinators of primarily bat-pollinated species, at least in some years (Fleming, Tuttle, and Horner, 1996
; Sahley, 1996
) and under this scenario would have been an active selective agent on the floral biology of the bat-pollinated ancestor. The alternative hypothesis is that the transition to moth pollination occurred first and served as a preadaptation to hummingbird pollination in P. marginatus. Under this scenario, the evolution of reduced flower size and nectar production would have preceded the evolution of diurnal anthesis. The transition from a highly specialized obligate pollination mutualism to a more generalized hummingbird-pollinated system seems perhaps the less parsimonious of these two alternative scenarios; however, there is molecular phylogenetic support for this evolutionary sequence in Yucca with the hummingbird-pollinated taxon Hesperaloe nesting within an obligately moth (Tegeticula)-pollinated clade (Yucca including Y. whippplei; Bogler, Neff, and Simpson, 1995
; Clary and Simpson, 1995
). The third possible scenario is the evolution of hummingbird- and moth-pollinated lineages from a common ancestor with a more generalized pollination system. Under this hypothesis, a transition from bat pollination to a generalized pollination system, perhaps involving various insect and hummingbird species, served as a preadaptation for the evolution of the more specialized pollination systems in Lophocereus and P. marginatus.
In summary, our analyses failed to reveal cpDNA variation in Lophocereus. This result is consistent with the conspecificity of taxa within the genus, however further genetic and ecological studies will be required to resolve the phylogeny of Lophocereus. Our analyses strongly support the close relationship of Lophocereus and P. marginatus. These results suggest that the shift from bat pollination in the Pachycereinae likely represents a single evolutionary event and did not arise independently in Lophocereus.
|
FOOTNOTES |
|---|
2 Present address: Department of Botany, Iowa State University, Ames, Iowa 50011 USA
3 Author for reprint requests (Tel.: (319) 335-1977; Fax: (319) 335-1069; dbhattac{at}blue.weeg.uiowa.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Arias S. T. Terrazas K. M. Cameron 2001 Phylogeny of Pachycereus (Cactaceae). American Journal of Botany, Abstract 523
Barthlott W. D. R. Hunt 1993 Cactaceae. In K. Kubitzki, J. G. Rohwer, and V. Bittrich [eds.], The families and genera of vascular plants, vol. II, 161–197. Springer-Verlag, Berlin, Germany
Bogler D. J. J. L. Neff B. B. Simpson 1995 Multiple origins of the yucca–yucca moth association. Proceedings of the National Academy of Sciences, USA 92: 6864-6867
Britton N. L. J. N. Rose 1909 The genus Cereus and its allies in North America. Contributions from the United States National Herbarium 12: 413-437
Brouat C. L. Gielly D. McKey 2001 Phylogenetic relationships in the genus Leonardoxa (Leguminosae: Caesalpinioideae) inferred from chloroplast trnL intron and trn L-trn F intergenic spacer sequences. American Journal of Botany 88: 143-149
Brown J. M. O. Pellmyr J. N. Thompson R. G. Harrison 1994 Mitochondrial DNA phylogeny of the Prodoxidae (Lepidoptera: Incurvarioidea) indicates rapid ecological diversification of yucca moths. Annals of the Entomological Society of America 87: 795-802
Buxbaum F. 1961 Die Enwicklungslinien der Tribus Pachyereae-F. Buxb. (Cactaceae-Cereoideae), Botanische Studien, Gustav Fischer-Verlag, Jena, Germany
Clary K. H. B. B. Simpson 1995 Systematics and character evolution of the genus Yucca L. (Agavaceae): evidence from morphology and molecular analyses. Boletin de la Sociedad Botanica de México 56: 77-88
Cornejo D. O. B. B. Simpson 1997 Analysis of form and function in North American columnar cacti (Tribe Pachycereae). American Journal of Botany 84: 1482-1501[Abstract]
Cota J. H. R. S. Wallace 1997 Chloroplast DNA evidence for divergence in Ferocactus and its relationships to North American columnar cacti (Cactaceae: Cactoideae). Systematic Botany 22: 529-542[CrossRef][Web of Science]
Demesure B. N. Sodzi R. J. Petit 1995 A set of universal primers for amplification of polymorphic non-coding regions of mitochondrial and chloroplast DNA in plants. Molecular Ecology 4: 129-131[Medline]
Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15
Dumolin-Lapegue S. B. Demesure S. Fineschi V. L. Corre 1997 Phylogeographic structure of white oaks throughout the European continent. Genetics 146: 1475-1487[Abstract]
El Mousadik A. R. J. Petit 1995 Chloroplast DNA phylogeography of the argan tree of Morocco. Molecular Ecology 5: 547-555
Faegri K. L. van der Pijl 1971 The principles of pollination ecology. Pergamon, Braunschweig, Germany
Fain S. R. L. D. Druehl D. L. Baillie 1988 Repeat and single copy sequences are differentially conserved in the evolution of kelp chloroplast DNA. Journal of Phycology 24: 292-302[Web of Science]
Felger R. S. C. H. Lowe 1967 Clinal variation in the surface-volume relationships of the columnar cactus Lophocereus schottii in northwestern Mexico. Ecology 48: 530-536[CrossRef][Web of Science]
Felsenstein J. 1981 Evolutionary trees from DNA sequences: maximum likelihood approach. Journal of Molecular Evolution 17: 368-376[CrossRef][Web of Science][Medline]
Fleming T. H. J. N. Holland 1998 The evolution of obligate pollination mutualisms: senita cactus and senita moth. Oecologia 114: 368-375[CrossRef][Web of Science]
Fleming T. H. C. T. Sahley J. N. Holland J. D. Nason J. L. Hamrick 2001 Sonoran Desert columnar cacti and the evolution of generalized pollination systems. Ecological Monographs 71: 511-530
Fleming T. H. M. D. Tuttle M. A. Horner 1996 Pollination biology and the relative importance of nocturnal and diurnal pollinators in three species of Sonoran Desert columnar cacti. Southwest Naturalist 41: 257-269
Gaut B. S. S. V. Muse W. D. Clark M. T. Clegg 1992 Relative rates of nucleotide substitution at the rbcL locus of monocotyledonous plants. Journal of Molecular Evolution 35: 292-303[CrossRef][Web of Science][Medline]
Gibson A. C. 1982 Phylogenetic relationships of Pachycereeae. In J. S. F. Barker and W. T. Starner [eds.], Ecological genetics and evolution: the cactus–yeast–Drosophila model system, 3–16. Academic Press, New York, New York, USA
Gibson A. C. K. E. Horak 1978 Systematic anatomy and phylogeny of mexican columnar cacti. Annals of the Missouri Botanical Garden 65: 999-1057[CrossRef][Web of Science]
Gibson A. C. K. C. Spencer R. Bajaj J. L. McLaughlin 1986 The ever-changing landscape of cactus systematics. Annals of the Missouri Botanical Garden 73: 532-555[CrossRef][Web of Science]
Gilbert D. G. 1992 SeqApp, a biosequence editor and analysis application (available from http://iubio.bio.indiana.edu/soft/molbio/seqapp/)
Hartmann S. J. D. Nason D. Bhattacharya 2001 Extensive ribosomal DNA genic variation in the columnar cactus Lophocereus. Journal of Molecular Evolution 53: 124-134[Web of Science][Medline]
Herre E. A. 1996 An overview of studies on a community of Panamanian figs. Journal of Biogeography 23: 593-607[CrossRef][Web of Science]
Herre E. A. N. Knowlton U. G. Mueller S. A. Rehner 1999 The evolution of mutualisms: exploring the paths between conflict and cooperation. Trends in Ecology and Evolution 14: 49-53
Herre E. A. C. A. Machado E. Bermingham J. D. Nason D. M. Windsor S. S. McCafferty W. Van Houten K. Bachmann 1996 Molecular phylogenies of figs and their pollinator wasps. Journal of Biogeography 23: 521-530[CrossRef][Web of Science]
Hillis D. M. J. P. Huelsenbeck 1992 Signal, noise, and reliability in molecular phylogenetic analyses. Journal of Heredity 83: 189-195
Holland J. N. T. H. Fleming 1999a Geographic and population variation in pollinating seed-consuming interactions between senita cacti (Lophocereus schottii) and senita moths (Upiga virescens). Oecologia 121: 405-410[CrossRef][Web of Science]
Holland J. N. T. H. Fleming 1999b Mutualistic interactions between Upiga virescens (Pyralidae), a pollinating seed-consumer, and Lophocereus schottii (Cactaceae). Ecology 80: 2074-2084[Web of Science]
Hunt D. R. N. P. Taylor 1990 The genera of the Cactaceae: progress towards consensus. Bradleya 8: 85-107
Levy F. J. Antonovics J. E. Boynton N. W. Gillham 1996 A population genetic analysis of chlorplast DNA in Phacelia. Heredity 76: 143-155
Lindsay G. 1963 The genus Lophocereus. Cactus and Succulent Journal 35: 176-192
Machado C. A. E. A. Herre S. McCafferty E. Bermingham 1996 Molecular phylogenies of fig pollinating and non-pollinating wasps and the implications for the origin and evolution of the fig–fig wasp mutualism. Journal of Biogeography 23: 531-542[CrossRef][Web of Science]
Maddison W. P. D. R. Maddison 1992 MacClade. Sinauer, Sunderland, Massachusetts, USA
Maskas S. D. M. B. Cruzan 2000 Patterns of intraspecific diversification in the Piriqueta caroliniana complex in southeastern North America and the Bahamas. Evolution 54: 815-827[CrossRef][Web of Science][Medline]
Mayer M. S. P. S. Soltis 1994 The evolution of serpentine endemics: a chloroplast DNA phylogeny of the Streptanthus glandulosus complex (Cruciferae). Systematic Botany 19: 557-574[CrossRef][Web of Science]
Mohanty A. J. P. Martin I. Aguinagalde 2000 Chloroplast DNA diversity within and among populations of the allotetraploid Prunus spinosa L. Theoretical and Applied Genetics 100: 1304-1310[CrossRef][Web of Science]
Nicolosi E. Z. N. Deng A. Gentile S. La Malfa G. Continella E. Tribulato 2000 Citrus phylogeny and genetic origin of important species as investigated by molecular markers. Theoretical and Applied Genetics 100: 1155-1166[CrossRef][Web of Science]
Pellmyr O. J. Leebens-Mack 1999 Forty million years of mutualism: evidence for Eocene origin of the yucca–yucca moth association. Proceedings of the National Academy of Sciences, USA 96: 9178-9183
Pellmyr O. J. N. Thompson J. M. Brown R. G. Harrison 1996 Evolution of pollination and mutualism in the yucca moth lineage. American Naturalist 148: 827-847[CrossRef][Web of Science]
Peñalba M. C. T. R. V. Devender 1998 Cambios de vegetacíon y clima en Baja California, México, durante los últimos 20,000 años. Geología del Noroeste 2: 21-23
Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818
Rasplus J.-Y. C. Kerdelhue I. Le Clainche G. Mondor 1998 Molecular phylogeny of fig wasps Agaonidae are not monophyletic. Comptes Rendus de L'Academie des Sciences Serie III Sciences de la Vie 321: 517-527[CrossRef]
Sahley C. T. 1996 Bat and hummingbird pollination of an autotetraploid columnar cactus, Weberbauerocereus weberbaueri (Cactaceae). American Journal of Botany 83: 1329-1336[CrossRef][Web of Science]
Shimodaira H. M. Hasegawa 1999 Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biology and Evolution 16: 1114-1116[Web of Science]
Small R. L. J. A. Ryburn R. C. Cronn T. Seelanan J. F. Wendel 1998 The tortoise and the hare: choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. American Journal of Botany 85: 1301-1315
Soltis D. E. M. S. Mayer P. S. Soltis M. Edgerton 1991 Chloroplast DNA variation in Tellima grandiflora (Saxifragaceae). American Journal of Botany 78: 1379-1390[CrossRef][Web of Science]
Soltis D. P. Soltis R. Kuzoff T. Tucker 1992 Geographic structuring of chloroplast DNA genotypes in Tiarella trifoliata (Saxifragaceae). Plant Systematics and Evolution 181: 203-216[CrossRef][Web of Science]
Swofford D. L. 2001 PAUP* V4.0b8. phylogenetic analysis using parsimony (*and other methods), version 4.0. Smithsonian Institution, Washington, D.C., USA
Taberlet P. L. Gielly G. Pautou J. Bouvet 1991 Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105-1110[CrossRef][Web of Science][Medline]
Valiente-Banuet A. M. D. C. Arizmendi A. Rojas-Martinez L. Dominguez-Canseco 1996 Ecological relationships between columnar cacti and nectar-feeding bats in Mexico. Journal of Tropical Ecology 12: 103-119
Van Devender T. R. In press Environmental history of the Sonoran Desert. In T. H. Fleming and A. Valiente Banuet [eds.], Evolution, ecology and conservation of columnar cacti and their mutualists. University of Arizona Press, Tucson, Arizona, USA
van Helverson O. 1993 Adaptations of flowers to the pollination by glossophagine bats. In W. Bartholott, C.M. Naumann, K. Schmidt-Loske, and K.-L. Schuchmann [eds.], Animal–plant interactions in tropical environments, 41–59. Museum Alexander Koenig, Bonn, Germany
Wallace R. S. 1995 Molecular systematic study of the Cactaceae: using chloroplast DNA variation to elucidate cactus phylogeny. Bradleya 13: 1-12
Weiblen G. D. 2000 Phylogenetic relationships of functionally dioecious (Moraceae) based on ribosomal DNA sequences and morphology. American Journal of Botany 87: 1342-1357
Wiebes J. T. 1979 Co-evolution of figs and their insect pollinators. Annual Review of Ecology and Systematics 10: 1-12
This article has been cited by other articles:
|
|
 |
|
  J. Shaw, E. B. Lickey, J. T. Beck, S. B. Farmer, W. Liu, J. Miller, K. C. Siripun, C. T. Winder, E. E. Schilling, and R. L. Small The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis Am. J. Botany, January 1, 2005; 92(1): 142 - 166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Shaw and R. L. Small Addressing the "hardest puzzle in American pomology:" Phylogeny of Prunus sect. Prunocerasus (Rosaceae) based on seven noncoding chloroplast DNA regions Am. J. Botany, June 1, 2004; 91(6): 985 - 996. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
