Skip Navigation



AOBPreview published online on November 10, 2007
Annals of Botany, doi:10.1093/aob/mcm271
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
101/1/145    most recent
mcm271v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kao, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kao, R. H.
Agricola
Right arrow Articles by Kao, R. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Origins and Widespread Distribution of Co-existing Polyploids in Arnica cordifolia (Asteraceae)

Rebecca Hufft Kao*

University of California Santa Cruz, EE Biology, EMS A316, 1156 High St, Santa Cruz, CA 95064, USA

* For correspondence. Present address: Bioagricultural Sciences and Pest Management, Colorado State University, Fort Collins, CO 80523-1177, USA. E-mail rhkao{at}lamar.colostate.edu

Received: 4 February 2007    Returned for revision: 15 May 2007    Accepted: 10 September 2007   


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

Background and Aims: Polyploidy is a central force structuring genetic diversity in angiosperms, but its ecological significance and modes of origin are not fully understood. This work investigated the patterns of coexistence and molecular relatedness of polyploids in the perennial herb, Arnica cordifolia.

Methods: The local- and broad-scale distributions of cytotypes were analysed using flow cytometry. Samples were collected from both roadside and understorey habitats to test the hypothesis of niche separation between triploids and tetraploids. The nuclear rDNA internal transcribed spacer (ITS) and plastid rpl16 spacer, trnL intron plus trnL-trnF spacer and trnK 3' intron regions were sequenced.

Key Results: Broad-scale sampling established that both triploids and tetraploids were common throughout the range of the species, pentaploids were rare, and diploids were not found. Local-scale sampling revealed coexistence of both triploids and tetraploids within the majority of sites. Triploids and tetraploids were equally represented in the understorey and roadside habitat. Triploids were more variable than tetraploids, but both cytotypes shared polymorphisms in ITS.

Conclusions: Coexistence of cytotypes appears to be the norm in A. cordifolia, but habitat differentiation (roadside vs. understorey) is not supported as a coexistence mechanism. Molecular analyses supported multiple events creating triploids but revealed a lack of variation in the tetraploids. Additionally, sequence polymorphisms in ITS suggested a hybridization event prior to polyploidization.

Key words: Apomixis, Arnica cordifolia, flow cytometry, habitat differentiation, minority cytotype exclusion theory, polyploidy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Polyploidy is a major force in plant diversification, occurring in the history of at least 47–70% of angiosperms (Grant, 1981; Masterson, 1994; Soltis and Soltis, 1999; Otto and Whitton, 2000; De Bodt et al., 2005). Although polyploidy is common and seems to have advantages (Maceira et al., 1993; Otto and Whitton, 2000; Levin, 2001), there are reasons it should not be favoured. Persistence of newly created cytotypes may be limited by reproductive isolation and frequency-dependent minority cytotype exclusion (Thompson and Lumaret, 1992). Models of a new polyploid cytotype in competition with its parental diploid show that due to minority cytotype disadvantage, niche separation should be important for persistence and spread of the new cytotype (Levin, 1975; Fowler and Levin, 1984). Coexistence of cytotypes usually requires niche differentiation and highly localized spatial patterns of habitat differentiation between cytotypes (Thompson and Lumaret, 1992). Changes resulting from polyploidization can lead to differential habitat and resource use by cytotypes (e.g. Brochmann and Elven, 1992; Felber-Girard et al., 1996; Gauthier et al., 1998; Brochmann et al., 2004; Soltis et al., 2004; Paun et al., 2006). However, we still do not know to what extent new polyploids are adapted to novel ecological niches, how much reproductive isolation exists between new polyploids and their progenitors or the probability of establishment of a new polyploid in relation to its mode of origin (Ramsey and Schemske, 1998).

Arnica cordifolia (Asteraceae), an understorey herb, is a widely distributed and relatively mature polyploid complex and exhibits greatest cytological diversity in the Rocky Mountains (Wolf, 1980, 1987). Wolf (1980) proposed, based on only nine samples, that tetraploids of A. cordifolia usually occurred in the shaded understorey and triploids occurred in disturbed open habitats, particularly road banks. Apomixis and odd polyploidy are both thought to be means of preserving adaptively valuable hybrid products (Grant, 1981). The occurrence of triploids coupled with apomictic reproduction in A. cordifolia led Wolf (1980) to suggest that the occurrence of triploids was the result of stabilizing selection to preserve an advantageous hybrid product exploiting a new niche. The relationship between polyploid distributions and disturbance has long been noted anecdotally (Stebbins, 1942), but little work has been done to test this observation.

There are several modes of polyploidy which could create the triploids and tetraploids in this system. Apomicts have several reproductive pathways that could lead to the repeated production of new ploidies within populations (Van Dijk et al., 1999). One of the main pathways considered previously includes the creation of tetraploids in one step from diploids, with subsequent diploid–tetraploid matings forming triploids (Ramsey and Schemske, 1998). If triploid formation is limited in this manner, triploids should be derived from tetraploid lineages. Alternatively, triploids could be created directly from diploids, with subsequent diploid–triploid matings creating tetraploids (Ramsey and Schemske, 1998). The patterns of molecular variation for this triploid bridge theory should be the opposite of the previous mechanism, with tetraploids arising from triploid lineages.

While one would like to be able to observe the establishment of new polyploids, for most cases, only the products of that process can be observed. The distribution of cytotypes can give insight into the origins of polyploidy, the extent of reproductive isolation between cytotypes and the mechanisms maintaining spatial separation of cytotypes. Whereas other studies have focused on narrow zones of polyploidy overlap, the widespread distribution of cytotypes of A. cordifolia in the southern Rocky Mountains was studied to determine the frequency of coexistence of different cytotypes. Sequence data from nuclear (ITS1) and chloroplast (rpl16 spacer, trnL intron + trnL-trnF spacer and trnK 3' intron) regions were used to elucidate the origin of the different cytotypes and their relationship to each other. In addition, to find out if triploids are indeed restricted to disturbed or roadside habitats when they co-occur with tetraploids was evaluated by testing the hypothesis that habitat differentiation contributes to cytotype coexistence in this system.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Study system
Arnica (Asteraceae) is a rhizomatous perennial genus with yellow ray and disc flowers. Its diploid species are sexual and its polyploid species generally form asexual seeds (meiotic diplospory) (Barker, 1966). The base chromosome number for Arnica is 19 (Bocher and Larsen, 1955; Ornduff et al., 1963, 1967). Most of the diploids in the genus show regular bivalent formation during meiosis, but the triploids and most tetraploids have an irregular meiosis that is characterized by the presence at metaphase I of a variable number of univalents, bivalents, chains and equational chromosome divisions (Ornduff et al., 1967). The reports of such irregularities have been observed in apomicts in this genus, so all species in the genus with similar meiotic disturbances or having triploid or pentaploid chromosome numbers are likely apomictic as well (including A. cordifolia) (Ornduff et al., 1967). Apomixis in this species has recently been confirmed, with both pseudogamous and autonomous (more common) endosperm formation (Kao, 2007).

Arnica cordifolia is one of the most widely distributed species of Arnica, ranging from Alaska to New Mexico and California (Maguire, 1943). It is largely a woodland species and can be found in high meadows and coniferous forests at 1200–3000 m a.s.l. Prior to this study, the following was known about cytotype distribution: diploids and pentaploids were each restricted to three populations (Ornduff et al., 1967; Löve et al., 1971; Wolf, 1980), some triploid populations had been found in the Rocky Mountains (Wolf, 1980, 1987), and the tetraploids dominated, occurring throughout the range of the species (Ornduff et al., 1963, 1967; Wolf, 1980) (Fig. 1A).


Figure 1
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Distribution of Arnica cordifolia with historical ploidy information (taken from Ornduff et al., 1963, 1967; Löve et al., 1971; Wolf, 1980, 1987) (A), collection sites (B), and ploidy information from this study (C). Grey shading indicates species distribution. Circles represent cytotype samples: white = diploids, light grey = triploids, dark grey = tetraploids, black = pentaploids. Squares represent populations not used (unfilled) and used (filled) in sequencing analyses.

 
Collection sites
Samples were collected from 15 sites in Colorado, USA, where the highest frequency of populations with multiple cytotypes was known from prior work. Six additional sites were sampled from the western edge of the species range, in California, Oregon and Washington (Table 1 and Fig. 1B). The Pacific Northwest is thought to be the origin of the species and where diploids had been found previously (Fig. 1A). Habitat type (understorey or roadside) was noted, and, where possible, samples were collected from both habitats (Table 1) to compare the frequency of triploids and tetraploids in each habitat type.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Distribution of cytotypes from field-collected individuals of Arnica cordifolia in Colorado and the Pacific Northwest States

 
Cytotype determination
To establish initial cytotype information, chromosomes from cell squashes of root tips were counted following the protocol of Wolf (1980). Rhizomes were collected in the field in 2003 and transplanted to the University of California Santa Cruz greenhouse to be used for cell squashes. After establishing ploidy level via squashes, leaves from greenhouse plants of known ploidy were used to verify the DNA content using flow cytometry following the methods of Arumuganathan and Earle (1991). After this initial verification, all samples were cytotyped using flow cytometry on a Becton-Dickinson FACScalibur at the University of California Santa Cruz or a Beckman Coulter Cytomics FC500 at the University of Colorado Cancer Center. All leaves were collected in 2003–2005, stored on ice and analysed within 5 d of collection. The number of samples individually cytotyped per population varied between two and 218 (Table 1). An internal standard (Pisum sativum, 9·56 pg; Johnston et al., 1999) was used in all samples to calculate DNA content.

Molecular analyses
Portions of leaf samples used in flow cytometry analyses were placed on DrieriteTM (WA Hammond Drierite CO, Ltd, Xenia, OH, USA). DNA was extracted using DNeasy plant extraction kits (Qiagen Inc., Valencia, CA, USA). PCR was performed in 25-µL reactions (12·6 µL water, 1 ng sample DNA, 2·5 µL 10 x reaction buffer, 1·5 µL MgCl2, 2·5 µL of each 10 M primer, 2 µL of 2·5 mM dNTP mix, and 0·4 µL Taq DNA polymerase). PCR products were cleaned with either Zymoclean Gel DNA Recovery kits (Zymo Research, Orange, CA, USA) or QIAquick PCR Purification Kits (Qiagen Inc.). Sequencing was carried out using ABI Prism BigDye 3·1 following Applied Biosystems (Foster City, CA, USA) conditions and chemistry, except using 1/5 volume reactions brought up to 10 µL All sequences were performed using Applied Biosystems Automated Sequencers (Models 377 and 3100).

Four regions were screened for variation using universal primers: ribosomal internal transcribed spacer region (ITS1), using ITS4 and ITS5 primers (White et al., 1990); rpl16 spacer (rpl16) using primers rpl16F71 and rpl16R1516 (Kelchner and Clark, 1997); trnL intron plus trnL-trnF spacer (together called trnL) using primers trnLc and trnFf (Taberlet et al., 1991); and trnK 3' intron (trnK) using primers matK8F and trnK2R (Steele and Vilgalys, 1994; Johnson and Soltis, 1995). The first region represents biparental DNA while the other three represent maternal DNA from chloroplasts. Only variable regions were used to sequence a larger number of individuals from the entire sampling distribution. For the two nonvariable regions (ITS1 and rpl16), ten and four samples were sequenced, respectively. Individuals from 16 sites were sequenced for trnK (104 individuals) and trnL(74 individuals). However, due to difficulties in sequencing the trnL region, only 72 individuals were sequenced for both regions (Table 2). Only these 72 individuals were used in subsequent analyses. The remaining individuals not represented in Table 2 but sequenced for trnK came from sites 2 (1 individual), 3 (5), 4 (1), 5 (1), 6 (1), 7 (2), 10 (1), 11 (3), 14 (8), 16 (1), 17 (3), 18 (2) and 19 (3). The additional individuals sequenced for trnL not represented in Table 2 came from sites 8 (1) and 13 (1).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Number of individuals of Arnica cordifolia for combined trnK 3' intron/trnL haplotypes (A–E) per cytotype per site

 
Eupatorium cannabinum (Asteraceae: Eupatorieae) was used as an outgroup to root my haplotype map (GenBank accession numbers: AB217695 and AB217692). Arnica is closely related to tribe Eupatorieae (Baldwin et al., 2002). Sequence editing and alignment was done using MEGA version 3·1 (Kumar et al., 2004). A neighbor-joining analysis was performed in PAUP 4·0b10 using the default settings (Sinauer Associates, Inc.; www.lms.si.edu/PAUP) to build a haplotype tree. Alternative phylogenetic hypotheses were compared using the Shimodaira–Hasegawa (SH) test (Shimodaira and Hasegawa, 1999) using PAUP with RELL (resampling estimated log-likelihood) optimization and 10 000 bootstrap replicates. An AMOVA was performed using Arlequin (Excoffier et al., 2005) on the combined haplotypes using the Colorado sites (excluding site 5 with only one individual). Gene diversity was calculated for each cytotype using Arlequin.

Statistical analyses
Statistical analyses were performed using Systat v10·2 (Systat Software Inc. 2002; http://www.systat.com). A paired t-test on the number of triploid vs. tetraploid individuals evaluated whether one cytotype was consistently dominant in the mixed cytotype sites. Count data were log transformed to improve normality prior to this analysis. For sites with both habitat types, a paired t-test was used to determine if triploids were more common in roadside than understorey habitat. For this analysis, the proportion of triploids in roadside habitat (number of triploids in roadside/total individuals sampled from roadside) was compared with the proportion in the understorey (number of triploids in understorey/total individuals sampled from understorey). To determine if triploids were more often found in roadside habitat than tetraploids, a t-test was used on the proportion of triploids (number of triploids in roadside/total number in roadside at that site) in all sites with roadside samples.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Cytotype distribution
The DNA content was 4·59 pg ± 0·25 for triploids, 6·49 pg ± 0·37 for tetraploids and 7·80 pg ± 0·38 for pentaploids. No diploids were found in any population. Tetraploids were common across the range, as previously thought (Table 1 and Fig. 1C). However, triploids were also surprisingly widespread, found in most sites and not just a few in the eastern edge of the range as suggested previously (Fig. 1A vs. Fig. 1C). Triploids made up 45 % of all plants sampled. Pentaploids were only found in four populations and were dominant (8 out of 8 samples) in only one site in California. The majority of sites (57·1 %) contained both triploids and tetraploids (Table 1 and Fig. 1C). When both triploids and tetraploids were present, no one cytotype was consistently dominant (paired t = 0·60, d.f. = 11, P = 0·56).

Molecular relatedness
No variation was found in ITS1 or rpl16 sequences (GenBank accession numbers: EF104922 and EF104921). For all ten samples sequenced, there were five polymorphic base pairs in ITS1. One base pair location was A/G and the other four were C/T heterozygotes.

One hundred and four individuals were successfully sequenced for trnK. The trnK 3' intron had five haplotypes (EF104925–EF104929). Due to a 2-bp indel, the size of the region varied between 518 and 520 bp per haplotype. Each haplotype is only 1 bp different from another haplotype. Tetraploids (34 individuals) were all one haplotype, and the two sequenced pentaploids were another unique haplotype. Triploids showed the greatest haplotype diversity, with four different haplotypes. The most frequent triploid haplotype (42 individuals and also the tetraploid haplotype) is the only haplotype in Southern Colorado and in the Pacific Northwest, the presumed origin of this species. Northern Colorado, with five haplotypes, shows the most diversity.

Only two haplotypes were found for the second variable region, trnL, for 74 individuals. Only a 1-bp difference(C–T transition) was found between the two haplotypes in a 748-bp region (EF104923 and EF104924). Similar to the results for trnK, one haplotype was found throughout the sample range, but the second haplotype was only found in northern Colorado. Combining trnL and trnK gave five unique haplotypes (A–E, Table 2 and Fig. 2A). One triploid haplotype in the trnK region is not represented here. Molecular diversity was only found in Colorado in the northern populations sampled (Fig. 3). Only one tetraploid individual was different from the others (Table 2 and Fig. 2A), with a 1-bp difference in trnL (matching the otherwise triploid-only trnL haplotype).


Figure 2
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. (A) Map of combined trnK 3' intron/trnL haplotypes –E) of Arnica cordifolia. Each line represents a 1-bp difference. Shading corresponds to cytotype and circle size is relative to the number of samples. (B) Neighbor-joining tree of five combined plastid haplotypes of Arnica cordifolia, rooted with Eupatorium cannabinum. Cytotype is shown to the right of each haplotype and bootstrap percentages higher than 50 are shown below the lines.

 


Figure 3
View larger version (127K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 3. Combined haplotype frequencies for trnK 3' intron/trnL for Arnica cordifolia triploids in Colorado. The size of the circles indicates the number of samples.

 
The neighbor-joining analysis of the combined haplotype data showed D and E as the most derived (Fig. 2B). The relationships of the haplotypes in Fig. 2B were similar to Fig. 2A, except here they were rooted with the outgroup E. cannabinum, showing haplotype A as the least derived haplotype of A. cordifolia. Although the SH test showed this to be the best phylogenetic hypothesis (–lnL = 2001·09), it was statistically no better an explanation of the data than four other phylogenetic hypotheses (these included a non-informative tree, two with tetraploids ancestral and two with pentaploids ancestral, all –lnL = 2006·87, P = 0·12). Gene diversity was higher for triploids (H = 0·491) than tetraploids (H = 0·077). More variation was partitioned within (74·2%) than among (25·8%) populations. The overall differentiation among populations was Fst = 0·258 (P = 0·016).

Habitat partitioning
Both tetraploids and triploids were found in understorey and roadside locations (Table 1). For sites with both understorey and roadside samples, triploids and tetraploids were found in each habitat type at most of the sites. Triploids were not more common than tetraploids on roadsides(t = 0·01, d.f. = 26, P = 0·99). Likewise, triploids were not more common in the roadside than in the understorey (t = 0·63, d.f. = 10, P = 0·54).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Whereas most other studies have focused on the co-occurrence of diploids and polypoids (Petit and Thompson, 1999; Husband et al., 2002; Meirmans et al., 2003; Baack, 2005; Paun et al., 2006), this system appears to lack diploids. Triploids and tetraploids were widespread and common throughout the range of Arnica cordifolia, with pentaploids being rare and diploids not found in any of the areas sampled. Both triploids and tetraploids were found equally among roadside and understorey habitat. A common, shared haplotype between triploids and tetraploids was found throughout the range sampled. In Colorado, where the majority of sampling was done, molecular diversity was found only in triploids and in northern Colorado.

Ploidy origins
Arnica is thought to have originated in the arctic or subarctic regions of western North America (Maguire, 1943), with A. cordifolia diploids widespread prior to the Pleistocene (Wolf and Denford, 1984). Wolf (1980) suggested that hybridization during and after the Pleistocene explains the high diversity of cytotypes in the Rockies. The high levels of molecular diversity found in this same region could be due to such hybridization events. Conversely, it could be that this species originated in the Rocky Mountains so these populations have had longer to diversify and create more triploid lineages. If the latter is true, that would mean that haplotype C was simply better at dispersing and colonizing new areas, which is why this haplotype is found at both ends of the species distribution. Currently, it is not possible to differentiate between these two theories.

Based on the haplotype tree (Fig. 2B), the tetraploids and pentaploids appear to have arisen from the triploids. This provides support for the triploid bridge hypothesis (Ramsey and Schemske, 1998; Yamauchi et al., 2004), suggesting that triploids mating with diploids allowed for the production and spread of tetraploids. One explanation for the origin of these three cytotypes is the creation of triploids from diploids, with subsequent cross-matings creating tetraploids and pentaploids. This would mean that diploids were once widespread (as thought previously by Wolf, 1980) but are now either rare or extinct since they were not found in this study. Asexual species having a broader distribution than their sexual counterparts is a common phenomenon (Bierzychudek, 1985). The lack of diploids found in the present study (and rareness in prior studies) suggests that these asexual polyploids have been more successful at recolonization post-glaciation than their presumed diploid progenitors, as has also been shown in other systems (e.g. Thompson and Whitton, 2006). However, the SH test could not distinguish among this and alternative phylogenetic hypotheses, probably due to the limited amount of variation found among the cytotypes.

Since both variable regions were from plastids, the variation found represents the plastid donor (presumably maternal) lineages in these plants (in addition to the high probability that all of the individuals arose via apomixis). This sheds some light on the maternal parent was giving rise to each line of cytotypes. It appears that all tetraploids are from one maternal lineage, whereas triploids appear to come from four different lineages and pentaploids from yet another lineage. This provides support for multiple origins of triploids but only a single origin of tetraploids in the regions sampled. However, it is possible that tetraploids and triploids have arisen repeatedly within each site but the data here are not resolved enough to assess this. Too few pentaploids have been sampled to say definitively if there have been multiple pentaploid events. Haplotype B (otherwise known only from California pentaploids) was also found from three individuals of unknown ploidy in two additional northern Colorado sites (data not shown). Like the multiple triploid genotypes found here, multiple genotypes of the apomictic polyploid Ranunculus carpaticola have been found, but these were distributed mainly among populations, suggesting that the limited variation found within populations was probably due to mutation rather than colonization (Paun et al., 2006).

Distribution of genetic diversity
Generally, genetic variation has been found to be higher in tetraploids than diploids (Soltis and Soltis, 1993), but some studies have found asexual polyploids to be less genetically diverse than their diploid progenitors (e.g. Paun et al., 2006). However no previous comparisons have been made between triploids and tetraploids. Triploids of A. cordifolia show higher levels of diversity than tetraploids. Similarly, high levels of diversity in asexual triploids have been found in Apios americana (Joly and Bruneau, 2004). The spatial structure of the triploid diversity suggests that there might be a historical barrier between northern and southern Colorado, such that the northern sampled region is between two historically isolated regions. Other species show similar breaks in Colorado (butterflies, DeChaine and Martin, 2005; red squirrels, Wilson et al., 2005). Separated glacial refugia have been offered as an explanation for these other species. Arnica cordifolia is thought to be an old species and part of the Arcto-Tertiary flora (Wolf and Denford, 1984), so it is quite possible that fragmentation due to glacial refugia is responsible for the distribution of variation in A. cordifolia as has been suggested for these other species. More extensive sampling north of Colorado could help clarify this.

Evidence for hybridization and asexuality
Sequence polymorphisms were found for ITS1 in all samples of A. cordifolia. The ITS region often undergoes rapid concerted evolution, resulting in sequence homogenization (Baldwin et al., 1995). Non-homogenization can be caused by recent hybridization events or due to ancient hybridization events forming asexual lineages (Baldwin et al., 1995). Another possible reason for the sequence polymorphism is mutational divergence of alleles in this asexual lineage (Birky, 1996). Of the six other Arnica species for which ITS1 has been sequenced (out of 30 species in the genus), there is no clear evidence that A. cordifolia was the result of a hybridization event between any of them (Table 3). ITS sequence polymorphism has been used to support hybridization in several systems (e.g. Guggisberg et al., 2006). Since all of the triploids and tetraploids of A. cordifolia sampled have the same sequence polymorphism, it is probable that a hybridization event occurred prior to polyploidization (or at least all current polyploids are derived from the same hybrid ancestor).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Nucleotides at five loci of ITS1 for A. cordifolia and six other Arnica species

 
The higher degree of variation in triploids than tetraploids, which should also be asexual (Ornduff et al., 1967), requires explanation. One possibility is that the variation means they are more unstable, such that mistakes occur more often during replication (Comai, 2005). However, given the evidence for asexuality in this system, it is likely that the higher diversity in triploids is due to them being much older than the tetraploids or to multiple origin events. Finding this much variation is surprising given that these are plastid regions and the variable populations are so close together. There is only one individual, in northern Colorado, which has the D haplotype combination, having the trnK 3' intron haplotype of C and the trnL haplotype of E (Fig. 2A).

Cytotype coexistence
The coexistence of cytotypes, predominantly triploids and tetraploids, appears to be common in A. cordifolia, and most sites studied were of mixed ploidy. Tetraploids were widespread as found previously, but triploids were also common and found throughout the range of the species. When the two occurred together, their relative frequencies did not suggest that one cytotype consistently predominated. The determination of dominance within a site could therefore be due to random factors, such as chance colonization (e.g. Kliber and Eckert, 2005). However, there could be environmental differences among sites that could lead to the dominance of one cytotype over the other.

In addition to investigating the mechanism of the broad-scale distribution and origins of multiple cytotypes in A. cordifolia, this work tested the theory put forth by Wolf (1980) that triploids are being selectively maintained in disturbed habitat. Differences in habitat use between cytotypes might allow for more diversity. Within sites, both triploids and tetraploids were commonly found in both understorey and roadside habitats. This contradicts Wolf's theory that triploids are being maintained primarily in roadside habitats due to their competitive inferiority in understorey habitat. There is no evidence here that roadside-understorey differences are important to habitat partitioning between cytotypes. However, there could be other environmental factors not explored here that are important to the maintenance of multiple cytotypes within sites (Kao, 2006).

Conclusions
This work not only helps elucidate the relationship between cytotypes, but it also suggests that these are distinct lineages. Contrary to previous theories as to the origins of cytotypes in A. cordifolia (Wolf, 1987), this work supports the creation of tetraploids and pentaploids from triploids. A more broad-scale survey of samples from throughout the range of A. cordifolia, instead of just the edges of its range, will help clarify the molecular relationships among ploidies and the distribution of molecular diversity within this species.

This is the first work to explore the relationships and distributions of widespread cytotype coexistence. Unlike previous findings of multiple cytotypes being maintained along a narrow overlap of cytotype distributions, here the coexistence occurs throughout the species range. Additionally, there is no clear-cut habitat differentiation between cytotypes that can explain this common coexistence. Many previously studied polyploid species are sexual, despite many polyploid systems having some form of asexual reproduction. Here, apomixis in A. cordifolia might be more representative of the distributions of and relationships between other polyploids in nature.


   ACKNOWLEDGEMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
LITERATURE CITED
 
This material is based upon work funded by the NSF under Grant No. 0508633 and grants from the Science, Technology, Engineering, Policy, and Society (STEPS) Institute. Rocky Mountain National Park samples were collected under permit # ROMO-2005-SCI-0031. Special thanks to the Hodges laboratory and the University of Colorado Cancer Center for use of their facilities for flow cytometry and sequencing work and to G. Thornton, T. Anderson, S. Yang, R. Halvorson, J. Sanderson and J. Arnett for sample collection. Thanks to I. M. Parker, G. Pogson, G. Gilbert, A. Blair and two anonymous reviewers for advice on this manuscript.


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

    Arumuganathan K, Earle ED. Estimation of nuclear DNA content of plants by flow cytometry. Plant Molecular Biology (1991) 9:229–233.

    Baack EJ. Ecological factors influencing tetraploid establishment in snow buttercups (Ranunculus adoneus, Ranunculaceae): minority cytotype exclusion and barriers to triploid formation. American Journal of Botany (2005) 92:1827–1835.[Abstract/Free Full Text]

    Baldwin BG. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: An example from the Compositae. Molecular Phylogenetics and Evolution (1992) 1:3–16.[CrossRef][Medline]

    Baldwin BG, Sanderson MJ, Porter JM, Wojciechowski MF, Campbell CS, Donoghue MJ. The ITS region of nuclear ribosomal DNA – a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden (1995) 82:247–277.[CrossRef][Web of Science]

    Baldwin BG, Wessa BL. Origin and relationships of the tarweed-silversword lineage (Compositae-Madiinae). American Journal of Botany (2000) 87:1890–1908.[Abstract/Free Full Text]

    Baldwin BG, Wessa BL, Panero JL. Nuclear rDNA evidence for major lineages of helenioid Heliantheae (Compositae). Systematic Botany (2002) 27:161–198.[Web of Science]

    Barker WW. Apomixis in the genus Arnica (Compositae) (1966) Seattle: University of Washington. PhD Thesis.

    Bierzychudek P. Patterns in plant parthenogenesis. Experientia (Basel) (1985) 41:1255–1264.[CrossRef]

    Birky CW. Heterozygosity, heteromorphy, and phylogenetic trees in asexual eukaryotes. Genetics (1996) 144:427–437.[Abstract]

    Bocher TW, Larsen K. Chromosome studies on some European flowering plants. Botanisk Tidsskrift (1955) 52:125–131.

    Brochmann C, Elven R. Ecological and genetic consequences of polyploidy in arctic Draba (Brassicaceae). Evolutionary Trends in Plants (1992) 6:111–124.[Web of Science]

    Brochmann C, Brysting AK, Alsos IG, Borgen L, Grundt HH, Scheen AC, et al. Polyploidy in arctic plants. Biological Journal of the Linnean Society (2004) 82:521–536.[CrossRef][Web of Science]

    Comai L. The advantages and disadvantages of being polyploid. Nature Reviews Genetics (2005) 6:836–846.[CrossRef][Web of Science][Medline]

    De Bodt S, Maere S, Van de Peer Y. Genome duplication and the origin of angiosperms. Trends in Ecology & Evolution (2005) 20:591–597.[CrossRef][Web of Science][Medline]

    DeChaine EG, Martin AP. Historical biogeography of two alpine butterflies in the rocky mountains: broad-scale concordance and local-scale discordance. Journal of Biogeography (2005) 32. 1943–1956.

    Excoffier L, Laval G, Schneider S. Arlequin ver. 3·0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online (2005) 1:47–50.[Medline]

    Felber-Girard M, Felber F, Buttler A. Habitat differentiation in a narrow hybrid zone between diploid and tetraploid Anthoxanthum alpinum. New Phytologist (1996) 133:531–540.[CrossRef][Web of Science]

    Fowler NL, Levin DA. Ecological constraints on the establishment of a novel polyploid in competition with its diploid progenitor. American Naturalist (1984) 124:703–711.[CrossRef][Web of Science]

    Gauthier P, Lumaret R, Bedecarrats A. Ecotype differentiation and coexistence of two parapatric tetraploid subspecies of cocksfoot (Dactylis glomerata) in the Alps. New Phytologist (1998) 139:741–750.[CrossRef][Web of Science]

    Grant V. Plant speciation (1981) New York, NY: Columbia University Press.

    Guggisberg A, Mansion G, Kelso S, Conti E. Evolution of biogeographic patterns, ploidy levels, and breeding systems in a diploid-polyploid species complex of Primula. New Phytologist (2006) 171:617–632.[Web of Science][Medline]

    Husband BC, Schemske DW, Burton TL, Goodwillie C. Pollen competition as a unilateral reproductive barrier between sympatric diploid and tetraploid Chamerion angustifolium. Proceedings of the Royal Society of London – Series B: Biological Sciences (2002) 269:2565–2571.[Abstract/Free Full Text]

    Johnson LA, Soltis DE. Phylogenetic inference in Saxifragaceae sensu-stricto and Gilia (Polemoniaceae) using matK sequences. Annals of the Missouri Botanical Garden (1995) 82:149–175.[CrossRef][Web of Science]

    Johnston JS, Bennett MD, Rayburn AL, Galbraith DW, Price HJ. Reference standards for determination of DNA content of plant nuclei. American Journal of Botany (1999) 86:609–613.[Abstract/Free Full Text]

    Joly S, Bruneau A. Evolution of triploidy in Apios americana (Leguminosae) revealed by genealogical analysis of the histone h3-d gene. Evolution (2004) 58:284–295.[Web of Science][Medline]

    Kao R. The coexistence of polyploids in Arnica cordifolia (Asteraceae) (2006) Santa Cruz: University of California. PhD Thesis.

    Kao R. Asexuality and the coexistence of cytotypes. New Phytologist (2007) 174:764–772.

    Kelchner SA, Clark LG. Molecular evolution and phylogenetic utility of the chloroplast rpl16 intron in Chusquea and the Bambusoideae (Poaceae). Molecular Phylogenetics and Evolution (1997) 8:385–397.[CrossRef][Web of Science][Medline]

    Kliber A, Eckert CG. Interaction between founder effect and selection during biological invasion in an aquatic plant. Evolution (2005) 59. 1900–1913.

    Kumar S, Tamura K, Nei M. Mega3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics (2004) 5:150–163.[Abstract/Free Full Text]

    Levin DA. Minority cytotype exclusion in local plant populations. Taxon (1975) 24:35–43.[Medline]

    Levin DA. 50 years of plant speciation. Taxon (2001) 50:69–91.[CrossRef][Web of Science]

    Löve A, Löve D, Kapoor BM. Cytotaxonomy of a century of Rocky Mountain orophytes. Arctic and Alpine Research (1971) 3:139–165.

    Maceira NO, Jacquard P, Lumaret R. Competition between diploid and derivative autotetraploid Dactylis glomerata L. from Galicia: implications for the establishment of novel polyploid populations. New Phytologist (1993) 124:321–328.[CrossRef][Web of Science]

    Maguire B. A monograph of the genus Arnica. Brittonia. (1943) 4:386–510.

    Masterson J. Stomatal size in fossil plants: evidence for polyploidy in majority of angiosperms. Science (1994) 264:421–424.[Abstract/Free Full Text]

    Meirmans PG, Vlot EC, Den Nijs JCM, Menken SBJ. Spatial ecological and genetic structure of a mixed population of a sexual diploid and apomictic triploid dandelions. Journal of Evolutionary Biology (2003) 16:343–352.[CrossRef][Web of Science][Medline]

    Ornduff R, Raven PH, Kyhos DW, Kruckeberg AR. Chromosome numbers in Compositae. III. Senecioneae. American Journal of Botany (1963) 50:131–139.[CrossRef][Web of Science]

    Ornduff R, Mosquin T, Kyhos DW, Raven PH. Chromosome numbers in Compositae. VI. Senecioneae. II. American Journal of Botany (1967) 54:205–213.[CrossRef][Web of Science]

    Otto SP, Whitton J. Polyploid incidence and evolution. Annual Review of Genetics (2000) 34:401–37.[CrossRef][Web of Science][Medline]

    Paun O, Greilhuber J, Temsch EM, Hörändl E. Patterns, sources and ecological implications of clonal diversity in apomictic Ranunculus carpaticola (Ranunculus auricomus complex, Ranunculaceae). Molecular Ecology (2006) 15:897–910.[Medline]

    Petit C, Thompson JD. Species diversity and ecological range in relation to ploidy level in the flora of the Pyrenees. Evolutionary Ecology (1999) 13:45–66.[CrossRef][Web of Science]

    Ramsey J, Schemske DW. Pathways, mechanisms, and rates of polyploid formation in flowering plants. Annual Review of Ecology and Systematics (1998) 29:467–501.[CrossRef][Web of Science]

    Shimodaira H, Hasegawa M. Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Molecular Biology and Evolution (1999) 16:1114–1116.[Web of Science]

    Soltis DE, Soltis PS. Molecular data and the dynamic nature of polyploidy. Critical Reviews in Plant Sciences (1993) 12:243–273.[CrossRef][Web of Science]

    Soltis DE, Soltis PS. Polyploidy: recurrent formation and genome evolution. Trends in Ecology & Evolution (1999) 14:348–352.[CrossRef][Web of Science][Medline]

    Soltis DE, Soltis PS, Tate JA. Advances in the study of polyploidy since plant speciation. New Phytologist (2004) 161:173–191.[CrossRef][Web of Science]

    Stebbins GLJ. The genetic approach to problems of rare and endemic species. Madroño (1942) 6:241–272.

    Steele KP, Vilgalys R. Phylogenetic analyses of Polemoniaceae using nucleotide-sequences of the plastid gene matK. Systematic Botany. (1994) 19:126–142.

    Taberlet P, Gielly L, Pautou G, Bouvet J. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology (1991) 17:1105–1110.[CrossRef][Web of Science][Medline]

    Thompson JD, Lumaret R. The evolutionary dynamics of polyploid plants: origins, establishment and persistence. Trends in Ecology & Evolution (1992) 7:302–306.[CrossRef][Web of Science]

    Thompson SL, Whitton J. Patterns of recurrent evolution and geographic parthenogenesis within apomictic polyploid Easter daises (Townsendia hookeri). Molecular Ecology (2006) 15:3389–3400.[CrossRef][Medline]

    Van Dijk PJ, Tas ICQ, Falque M, Bakx-Schotman T. Crosses between sexual and apomictic dandelions (Taraxacum). II. Heredity (1999) 83:715–721. The breakdown of apomixis.[CrossRef][Web of Science][Medline]

    White T, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal ribosomal rna genes for phylogenetics. In: Pcr protocols. A guide to methods and applications—Innis M, Gelfand D, Sninsky J, White J, eds. (1990) New York: Academic Press. 315–322.

    Wilson GM, Den Bussche RA, McBee K, Johnson LA, Jones CA. Intraspecific phylogeography of red squirrels (Tamiasciurus hudsonicus) in the central Rocky Mountain region of North America. Genetica (2005) 125:141–154.[CrossRef][Web of Science][Medline]

    Wolf SJ. Cytogeographical studies in the genus Arnica (Compositae: Senecioneae). I. American Journal of Botany (1980) 67:300–308.[CrossRef][Web of Science]

    Wolf SJ. Cytotaxonomic studies in the genus Arnica (Compositae. Rhodora (1987) 89. Senecioneae). 391–400.[Web of Science]

    Wolf SJ, Denford KE. Taxonomy of Arnica (Compositae) subgenus Austromontana. Journal of the New England Botanical Club (1984) 86:239–309.

    Yamauchi A, Hosokawa A, Nagata H, Shimoda M. Triploid bridge and role of parthenogenesis in the evolution of autopolyploidy. American Naturalist (2004) 164:101–112.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
101/1/145    most recent
mcm271v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kao, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kao, R. H.
Agricola
Right arrow Articles by Kao, R. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?