| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
3Institute for Systematics and Population Biology, University of Amsterdam, Kruislaan 318, NL-1098 SM Amsterdam, The Netherlands; and 4Institute for Plant Genetics and Crop Plant Research IPK, Corrensstrasse 3,D-06466 Gatersleben, Germany
Received for publication February 25, 1998. Accepted for publication January 26, 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: adaptive radiation • Asteraceae • cpDNA RFLPs • indels • long-distance dispersal • Microseris • phylogeny • trnL(UAA)-trnF(GAA)
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), the genus Dendroseris on the Juan Fernandez Islands (Crawford et al., 1992
), and the genus Aeonium on the Canary islands (Lems, 1960
). These examples concern taxa noted for their conspicuous morphological variation that are supposed to have been evolved from one or a few individuals after long-distance dispersal from related, more uniform, genera on the continent. In contrast to the morphology, molecular diversification is generally less pronounced among the oceanic island relatives. This suggests sampling of genetic variation by a founding event followed by drift in the next few generations. Molecular evolutionary studies of the oceanic island taxa helped to understand mechanisms of evolution, origins of organismic lineages, and the genetic basis of adaptations (e.g., Schilling, Panero, and Eliasson, 1994
; Francisco-Ortega, Jansen, and Santos-Guerra, 1996
; Okada, Whitkus, and Lowrey, 1997
; Vargas, Baldwin, and Constance, 1998
). Apart from the oceanic island flora, examples of disjunct intercontinental distributions of plant genera exist, such as the disjunction of temperate herbs between the west coast of North America and southern South America (Carlquist, 1983
). Like the floras of the oceanic islands, these disjunctions are supposed to be the result of bird dispersals rather than of tectonic plate movements (Carlquist, 1983
). Little empirical data are available about the evolution of intercontinental dispersed species, but investigation of these taxa could add insights into the processes of adaptive radiation and speciation.
The Australian and New Zealand Microseris (Asteraceae, Lactuceae), M. lanceolata (Walp.) Sch.-Bip., and M. scapigera (Forst.) Sch.-Bip., provide a good opportunity to study patterns of adaptive radiation following intercontinental dispersal. This allotetraploid (2n = 4x = 36) perennial plant group finds its closest relatives in western North America (Chambers, 1955
; Wallace and Jansen, 1990) where six perennial and seven annual species of Microseris occur. The one remaining species of the genus, the annual M. pygmaea, occurs in Chile. Karyotypic and morphological features suggest an origin of the Australian and New Zealand Microseris by hybridization of a North American annual and perennial diploid species, followed by polyploidization and long-distance dispersal (Chambers, 1955
). This hypothesized origin suggests a single introduction into Australia or New Zealand. Its present distribution covers New Zealand, Tasmania, and southern Australia (Fig. 1), and various ecotypes exist (Table 1; Fig. 2). Marked adaptations are tubers to overcome summer drought ("murnong" or M; Gott, 1983
), vegetative propagation via shoots on horizontally outgrowing roots to resist winter-frozen mountain slopes ("alpine" or A), and waxy leaves to avoid evaporation near seashores ("coastal" or C). This morphological diversification is maintained in the greenhouse. Both self-compatible and self-incompatible breeding systems are present in the taxon (Table 1), and the Australian mainland harbors a rare autofertile "fine-pappus" ecotype (F; N. H. Scarlett, personal communication, La Trobe University, Melbourne, Australia). On the basis of morphological features, it has been suggested to include the Australian "fine-pappus" ecotype together with the Tasmanian and New Zealand plants in M. scapigera, and the Australian "alpine" and "murnong" ecotypes in M. lanceolata [B. V. Sneddon, personal communication, Victoria University of Wellington, New Zealand: a revision of Australian Microseris for the Flora of Australia, volumes 37 and 38, Asteraceae 1 and 2, in preparation (A. E. Orchard [Ed.], Australian Biological Resource Study, Canberra, Australia)].
|
|
|
). It is one of the 4–12 native genera of Lactuceae in Australia (A. E. Orchard, personal communication), and one of the 7–9 native genera in New Zealand (I. Breitwieser, Landcare research, Christchurch, New Zealand, personal communication). Tubers of the Australian lowland ecotype were called "murnong" by Victorian aborigines and are known to have been used as a staple food (Gott, 1983
). The current genetic structure of the "murnong" ecotype was analyzed with allozymes and shows a low detectable differentiation among populations, supposedly reflecting its original abundance and widespread distribution (Prober, Spindler, and Brown, 1998
).
A phylogenetic tree of Microseris based on restriction fragment length polymorphisms (RFLPs) in the chloroplast genome shows a strongly supported annual and perennial clade and places the two Australian accessions tested at the basis of the annual clade (Wallace and Jansen, 1990
). In a nuclear RFLP study of Microseris including six Australian mainland accessions, the Australian Microseris shows features of both the annual and perennial species (Van Houten, Scarlett, and Bachmann, 1993
). Both studies confirm the close relationship of the Australian Microseris to the North American species, while the cpDNA phylogeny suggests that an ancestral annual plant has been the maternal parent of the original hybrid. Similarity in achene, pappus, root, and inflorescence morphology renders the distinctive perennial M. borealis a likely candidate for the paternal parent (Chambers, 1955
). The nuclear RFLPs show the "fine-pappus" ecotype as least diverged from the North American species, which caused the authors to suggest that this ecotype might be representative for the earliest founders in Australia (Van Houten, Scarlett, and Bachmann, 1993
). The autofertile breeding system of the "fine-pappus" ecotype provides a likely explanation for successful establishment of Microseris into the Southern hemisphere after arrival of a single individual. This hypothesis would also imply a shift in breeding system from autofertility to self-incompatibility within the Australasian Microseris. It also suggests a derivation of Tasmanian and New Zealand plants from (an) Australian population(s).
In the present study we investigated the phylogenetic relationships among a wide range of Australian and New Zealand Microseris populations that represent the distribution range and various ecotypes and between the Australasian Microseris and the North American species of the genus. For this, we used RFLPs and trnL(UAA)-trnF(GAA) intergenic spacer variants in the chloroplast genome. The choice of chloroplast DNA (cpDNA) was based on its proven utility in molecular evolutionary studies (reviewed in Soltis, Soltis, and Doyle, 1992
), its freedom of complex sexual processes due to its uniparental inheritance, and its ease of analysis. Due to the conservative mode of evolution of cpDNA (Downie and Palmer, 1992
) and the low taxonomic level of our study, we used a fine-scale restriction site analysis with four-base enzymes in combination with fragment separation on polyacrylamide gels to detect sufficient informative RFLPs. We discuss the utility of the small indels obtained with this method for phylogenetic analysis at the intra- and interspecific level. TrnL-trnF regions were amplified and products of a representative set of plants were sequenced to search for additional cpDNA mutations. The phylogenetic results are compared with morphological, chloroplast, and nuclear DNA analyses to obtain a better understanding of the evolutionary history and adaptive radiation of Australian and New Zealand Microseris and to assess evolutionary processes following colonization of a continent.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). All populations and outgroup species were represented by two plants in the RFLP analysis and an additional 2–10 plants in the trnL(UAA)-trnF(GAA) length variant analysis. The annual species were chosen from populations most diverged for RAPD patterns and cpDNA RFLPs (Van Heusden and Bachmann, 1992a, b
; Roelofs and Bachmann, 1997
; Roelofs et al., 1997
). The trnL-trnF intergeneric spacers of eight ingroup and two outgroup accessions, representing the different variants, were sequenced.
|
), and the modifications by Roelofs and Bachmann (1995)
. In addition, either an RNase treatment was included or DNA was CsCl purified (Vlot et al., 1992
). Approximately 5 µg of total DNA were digested with each of the four-base recognizing restriction enzymes, HinfI, RsaI, and Tru1I, under conditions recommended by the manufacturer. Restriction fragments were resolved on 6% polyacrylamide gels (Sequagel-6, Biozyme, Landgraaf, The Netherlands) and electroblotted to Hybond NTM membranes (Amersham, Buckinghamshire, UK; Kreitman and Aguadé, 1986
). The cpDNA was examined using a SacI Lactuca cpDNA library that covers 96% of the genome (Jansen and Palmer, 1987
; Table 3). Probes were 32P random primed labeled (Feinberg and Vogelstein, 1983
), and blots were hybridized overnight at 65°C with one probe at a time. Subsequently, blots were washed twice for 5 min with 2x SSC at room temperature (Sambrook, Fritsch, and Maniatis, 1989
), twice for 30 min with 0.5x SSC/0.1% SDS at 65°C, and once with 0.5x SSC/0.1% SDS at room temperature. X-OMAT-AR5 films (Kodak, Driebergen, The Netherlands) were exposed at -70°C for 2–10 d with the use of an intensifying screen.
|
). The choice of method was, however, necessary to detect sufficient informative RFLPs, based on earlier results by Wallace and Jansen (1990)
using five- and six-base enzymes. In the ingroup, homology assessment of mutations was judged from the low number of synapomorphies found per enzyme/probe combination (one to five) in the generally conservative fragment patterns. Mutations were considered restriction site changes when both the undigested as well as the two fragments after digestion were detected. Mutations were regarded as indels when detected with more than one enzyme using the same probe. Hybridization with the trnL-trnF fragments aided in the homology assessment within the relatively variable 7.7-kb region (probe 11; Table 3) by eliminating variable bands that hybridized to trnL-trnF. Except for probes 10 and 11 (6.9- and 7.7-kb region; Table 3) mutations were not found in neighboring probe regions at the chloroplast genome, so that multiple scoring of identical mutations was avoided. Homology assessment of mutations in the ingroup vs. the outgroups, especially the more diverged North American perennial Microseris and Uropappus lindleyi (Wallace and Jansen, 1990
), was more difficult. When fragment patterns were too complex, mutations in the ingroup were scored as missing data in the outgroups. Consequently, the number of mutations on which rooting was based is reduced. In the case of complex patterns, also, fewer mutations were included among the outgroups than actually present. The data set is available on request from the first author.
|
), the modifications by Hombergen and Bachmann (1995)
, and an extra RNase treatment. TrnL-trnF intergenic spacers were amplified from 2 to 3 ng total DNA using the "e" and "f" primers of Taberlet et al. (1991)
and following their protocol. Either 0.05 units of Super Taq DNA Polymerase with Super Taq buffer (HT Biotechnology, Cambridge, UK) or, when products were sequenced, 2.75 units ExpandTM high-fidelity thermostable DNA polymerase with accompanying MgCl2 buffer (Boehringer Mannheim, Almere, The Netherlands) were used. Amplification products were resolved on 1.2% agarose gels and visualized by UV-light after staining with ethidium bromide. For DNA sequencing, trnL-trnF amplification products were purified from 1% low melting point agarose gels using the Geneclean II kit (Biolabs, Westburg, Leusden, The Netherlands). Fragments were either sequenced manually, using the direct CircumventTM Thermal Cycle dideoxy DNA sequencing kit (Biolabs), or automated on an ALFexpressTM (Pharmacia Biotech, Roosendaal, The Netherlands) after cloning into a Bluescript SK(+) vector. For cloning, T-vectors were constructed by digestion of a Bluescript SK(+) vector at the EcoRV site followed by incubation of 50 µg vector with 0.3 mmol/L dTTP and 0.05 units of Super Taq DNA Polymerase with Super Taq buffer for 3 h at 72°C. Similarly, purified fragments were equipped with a 3'-terminal Adenosine residue by incubating them with dATP. Ligation of 25 ng A-fragment with 100 ng T-vector was performed with T4 DNA ligase according to the instructions of the manufacturer. Plasmids were electrotransformed into E. coli strain HB101 using the Gene PulserTM (BioRad) apparatus, and transformants were plated onto IPTG/X-gal plates (Sambrook, Fritsch, and Maniatis, 1989
). Plasmid DNA was isolated from positive clones using the alkaline lysis mini preparation method (Sambrook, Fritsch, and Maniatis, 1989
) with an additional RNase, phenol/chloroform, and ethanol precipitation treatment. Final yield was 
75 µg of purified DNA in 120 µL H2O from which 12 µg were sequenced using the Cy5TM AutoReadTM sequencing kit (Pharmacia Biotech). Sequences were manually aligned to the tobacco trnL-trnF region (Shinozaki et al., 1986
), and secondary structures of tRNA-Phe molecules were checked by hand (Sprinzl et al., 1989
).
Phylogenetic analysis
Phylogenetic trees were calculated with PAUP 3.1.1 (Swofford, 1993
) using the heuristic search algorithm with TBR branch swapping, STEEPEST DESCENT and MULPARS options, and with 1000 random additions of taxa. Calculations were performed with both the restriction site and length mutations included, using one representative that showed identical mutations, and with all changes equally weighted. The calculations were performed twice, once with the annual Microseris as an outgroup (mutations 1–55; Table 3A) and the perennial Microseris and Uropappus lindleyi excluded, and once with U. lindleyi as an outgroup (mutations 1–77; Table 3A, B). Dollo analyses or separate analyses for restriction site data were unnecessary because site changes showed no homoplasy. Decay analyses (Bremer, 1988
; Soltis et al., 1993
) were performed to assess branch support. Bootstrap analyses (Felsenstein, 1985
) were omitted because there were many possible most parsimonious trees. This was mainly due to missing data in a few ingroup individuals that were not analyzed for all three enzyme/probe combinations, and outgroup accessions that were not scored for all mutations. An indication for bootstrap support of branches within the ingroup was obtained by an analysis using only M. elegans-1 as an outgroup.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), and in the region wherein the trnL-trnF region is located (probe 11). In addition to the mutations found in the Australian, New Zealand, and North American annual Microseris (see above), 22 mutations (numbers 56–77; Table 3B) were included for the perennial outgroups (Table 2) and Uropappus lindleyi. All but two of these 22 mutations were potentially phylogenetically informative. Two mutations were found to be variable in both the ingroup and perennial outgroups (numbers 32 and 41; Table 3), while no mutations were variable in the annual as well as the perennial Microseris.
TrnL(UAA)-trnF(GAA) intergenic spacer analysis
Amplification products of the trnL-trnF intergenic spacer of different Microseris accessions showed one of three variants: a long (451 bp; Fig. 4), short (289 bp), or double (289 and 365 bp) fragment. The three variants cosegregated with mutations 21 and 23 (Table 3), which was confirmed by hybridizing the total DNA blots using trnL-trnF fragments as a probe. The DNA sequences showed the long trnL-trnF fragment of Microseris to be 85% homologous to the corresponding sequence in tobacco (positions 49854-50291; Shinozaki et al., 1986
; Fig. 5). The sequences of the short trnL-trnF fragments, isolated from either the single short or double fragment variants (Figs. 4, 5), were identical and exhibited a deletion of 162 bp when compared to the long fragment. The longer fragment of the double fragment variant showed at its upstream part a sequence identical to these found for the short fragments, and at its downstream part an additional 76 bp that resembled a tandemly repeated trnF exon (positions 50232–50307; Shinozaki et al., 1986
). The 76 bp included 16 nucleotides of the 5'-trnF exon downstream of the "f " primer site, and 60 nucleotides of the 3'-exon finishing with a second "f" primer site. Different accessions of similar length variants showed minor to no nucleotide substitutions. Only the perennial M. borealis contained variation at positions 126 (Fig. 5), 234, and 334–339, the latter resulting in the loss of a Tru1I site. In addition, one of the Microseris lanceolata-2 accession, M2-b, showed variation at positions 29 and 377. Because sequence variation appeared to be virtually absent within the trnL-trnF spacer, sequencing of this region was limited to a few representatives, and nucleotide substitutions found were not used in the phylogenetic analysis.
|
|
). The secondary structure of the tRNA-Phe molecule coded for by the 5'-exon showed a G-G mismatch at base pair 4–69 (positions 418 and 483) in the acceptor stem, while no mismatches occurred in the other three stems. The annealing of the acceptor stem of the tRNA-Phe molecule coded for by the 3'-exon is, due to the two substitutions found at positions 1 and 2, probably also disturbed. Apart from the three trnL-trnF variants obtained by polymerase chain reaction (PCR), the blots hybridized with the trnL-trnF fragments used as a probe showed a few other variants. One of these variants corresponded to mutation 24 (Table 3), whereas others were detected in three Microseris lanceolata accessions (A6-a and M14), the two perennial outgroups tested (Table 2), and Uropappus lindleyi. Lengths of the fragments involved indicated that these variants might represent independent duplications of the trnF exon, although this was not confirmed by the amplification products. Because the exact nature of these variants was unknown, they were excluded from the data matrix, while mutation 21 was scored as missing data for the accessions involved. Currently, the additional trnL-trnF variants are being investigated by sequencing the trnL(UAA)-trnV(UAC)/ndhJ region (Vijverberg and Bachman, in press).
The distribution of the three trnL-trnF variants is shown in Fig. 1. The short fragment, which exhibited the 162-bp deletion, is unique to Australian Microseris in which it was observed in populations from New South Wales, South Australia, and northeastern Victoria (
). Approximately one-third of the plants of four of the populations from New South Wales (M3, M7, M9, and M10, !; Fig. 1) showed the double fragment, which exhibited the duplicated trnF exon. All New Zealand, Tasmanian, and remaining Australian Microseris investigated, as well as all outgroup species, showed the long trnL-trnF fragment. Some populations from northeastern Victoria (A6, M14, M16, and M17, ;t1
;t1; Fig. 1), involving both the "alpine" (Fig. 2; Table 1) and "murnong" ecotypes, showed within-population variation for the long and short fragment.
Phylogenetic relationships within the ingroup
Cladistic analyses of 53 Australian and New Zealand Microseris populations and three annual outgroup species (Table 2), using 55 equally weighted length and restriction site mutations (Table 3A), resulted in 1120 most parsimonious trees of length 60, consistency index 0.92 (0.86 when autapomorphies were excluded), and retention index 0.96 (autapomorphies in- or excluded). The strict consensus tree of the search is shown in Fig. 6. A total of five homoplasious characters were found, of which three are plotted onto the tree (numbers 11, 18, and 29; Fig. 6), while the two others (numbers 41 and 46) supported branches not present in the strict consensus tree.
|
unpublished data). The ingroup accessions that were not included in the clade are all members of the second Southern hemisphere species of Microseris, M. scapigera, and comprise all Tasmanian and New Zealand populations as well as the autofertile or "fine-pappus" ecotype from the Australian mainland. Except for population F8nt, the New Zealand Microseris shared one unique insertion (number 4; Table 3). In addition, F8nt differed from the other New Zealand populations by two autapomorphies (numbers 7 and 17). The "fine-pappus" ecotypes from Australia and Tasmania (F1-F7nt) were least divergent from the North American Microseris.
Microseris lanceolata was divided into three clades on the basis of one or two chloroplast mutations each (Mln-1, -2, and -3; Fig. 6). Mln-1 (
; Figs. 1, 6) included populations from northern New South Wales and one from the vicinity of Melbourne. Plants of these populations resembled each other in their morphology, which is intermediate between the "murnong" and "alpine" ecotypes (Table 1). Mln-2 (; Fig 1) consisted of all populations of the "murnong" ecotype from New South Wales, northern Victoria, and South Australia, as well as population M15, and part of the individuals of populations M16 and M17 (and M14) from eastern Victoria. This clade included in addition part of the individuals of population A6 of the "alpine" ecotype, also from northeastern Victoria. Mln-2 was well characterized by the 162-bp deletion in the trnL-trnF intergenic spacer (number 23; Table 3), and some individuals within this clade contained the duplicated trnF exon (number 21). Mln-3 (; Fig. 1) comprised the remaining (individuals of the) Victorian populations, both "murnong" and "alpine" ecotypes. Within this last clade, population A7 of the "alpine" ecotype shared one insertion with its neighboring populations of the "murnong" ecotype (number 41; Table 3).
Position of the ingroup within the genus
Cladistic analyses of all Microseris species tested (Table 2) and Uropappus lindleyi used as an outgroup, using 77 equally weighted length and restriction site mutations (Table 3), resulted in 1680 most parsimonious trees of length 84, consistency index 0.92 (0.88 when autapomorphies were excluded), and retention index 0.96 (autapomorphies in- or excluded). The strict consensus tree, summarized for the ingroup topology shown in Fig. 6, is presented in Fig. 7. In addition to the five homoplasious characters mentioned in the previous paragraph, mutation 75 was homoplasious (Table 3). The results showed the genus Microseris to be defined by five chloroplast mutations (Fig. 7). The annual, Australian, and New Zealand Microseris form a monophyletic group on the basis of three mutations and were sister to the perennial Microseris that were defined by ten mutations. The monophyly of the annual species was not supported. Microseris douglasii-1 and M. elegans-2 shared mutations with each other rather than with the second accession of their species, resembling earlier results of Roelofs et al. (1997)
. The Californian M. laciniata-2 shared two mutations with M. borealis rather than with the Oregon accession of M. laciniata.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) and are therefore often omitted from phylogenetic analyses (e.g., Sytsma and Gottlieb, 1986
; Palmer et al., 1988
; Wallace and Jansen, 1990
). They have a tendency to cluster in "hot spots" (e.g., Downie and Palmer, 1992;
Hipkins et al. 1995
), which makes homology assignment difficult. In several studies, however, small indels have been proven useful for phylogenetic reconstruction of closely related species (e.g., Doebley, Ma, and Renfroe, 1987
; Soltis et al., 1989
; Soltis, Soltis, and Bothel, 1990
). Also, in a chloroplast study of Crassulaceae based on sequences of the trnL-trnF intergenic spacer, only one of 34 indels of length 3–20 bp was found to be homoplasious (Van Ham et al., 1994
). In our study, no "hot spot" regions for indels are recognized because few mutations are detected per enzyme/probe combination. Four of the five homoplasious mutations are <20 bp, whereas the fifth has an unknown length. Our results show that small indels can be useful for phylogeny reconstruction at the intra- and interspecific level, in particular when they are >20 bp.
The fact that many more indels than restriction site changes are found contrasts with most other studies (e.g., Soltis, Soltis, and Milligan, 1992
; Gielly and Taberlet, 1994
; Mes, van Brederode, and ‘t Hart, 1996
; Sang, Crawford, and Stuessy, 1997
). Only four of the RFLPs detected within the ingroup and annual outgroups are recognized as site changes from which three involve the AT-rich Tru1I sequence (TTAA; Table 3A). Apparently, the RFLPs of <20 bp that would have mostly remained undetected in conventional Southern blotting using six-base enzymes and agarose gels are indels rather than changes in closely linked restriction sites. The results suggest that evolution of the chloroplast genome, at least within Microseris, primarily occurs by small indels at the lower taxonomic levels. The phylogenetic utility of small indels in the Australasian Microseris indicates that a fine-scale restriction site method using four-base enzymes and polyacrylamide gels can be chosen in studies at the intra- and interspecific levels when other methods lack sufficient variation.
TrnL(UAA)-trnF(GAA) variation
A deletion of 162 bp in the trnL-trnF spacer, not detected in any of the North American Microseris or in the reference species tobacco, is present in populations from the Australian mainland (Mln-2, ; Figs. 1, 5, 6). A subset of these plants also exhibit a duplicated trnF exon (!; Fig. 1). Although deletions in intergenic spacers are quite common (e.g., Downie and Palmer, 1992
), duplications of entire tRNA genes are rarely reported (e.g., Tsai and Strauss, 1989
) and have not yet been observed in the plastids of Asteraceae. For the chloroplast genomes of grasses and gymnosperms, however, a number of partially duplicated tRNAs (Quigley and Weil, 1995; Howe et al., 1988; Tsudzuki et al., 1994; Hipkins et al., 1995) and pseudo-tRNA genes (Hiratsuka et al., 1989; Shimada and Sugiura, 1989) are known. The mechanism by which the tRNA gene duplications arose is unknown, but it has been speculated that their secondary structure or transcription might be involved (Howe et al., 1988;
Hipkins et al., 1995
). Also, entire tRNAs as well as dispersed repeats that are segments of tRNA genes are mentioned as possible substrates for recombination (e.g., Wolfe, 1988
). Manual inspection of the sequence of the trnF exon shows no internal repeats or "recombinogenic" (Howe et al., 1988
) parts. Repeated sequences are in general known for their susceptibility to recombination and slippage misrepair (e.g., Tsai and Strauss, 1989
; Wolfson, Higgins, and Sears, 1991
; Hipkins et al., 1995
). Accordingly, the tandemly repeated trnF exon might be involved in DNA rearrangements. This may also explain the additional trnL-trnF variants found within the chloroplast genomes of Microseris.
It is unknown whether one or both of the trnF exons code for functional tRNA-Phe molecules. Sequences of the exons indicate that the annealing of the acceptor stem is disturbed in both corresponding tRNA-Phe's. Due to these mismatches, the two exons may represent pseudogenes, which raises the question whether a functional trnF gene is present at all. The nucleotide substitutions at the downstream end of the 5'- (Fig. 5) and upstream part of the 3'-trnF exon suggest that this part of the sequence has been involved in the duplication. Because the remaining sequence of the 5'-exon is unchanged, this exon probably represents the original trnF gene, while the 3'-exon should then be the duplicated one. Possibly, the tRNA-Phe coded for by the 5'-exon is still functional despite the G-G mismatch in its acceptor stem.
According to the distribution of the two apomorphic trnL-trnF variants (Fig. 1), the present distribution of the variants tentatively indicates a spread of Microseris from Victoria to the north and west. Interestingly, the populations that show variation for the presence and absence of the 162-bp deletion concern both the "alpine" (Fig. 2) and "murnong" ecotypes. This probably indicates migration of Microseris between populations of either ecotype, possibly associated with introgression, or parallel evolution of similar ecotypes. Independent deletions of the 162 bp in the two ecotypes is considered less likely because this mutation shows no homoplasy (Fig. 6).
Evolutionary history of Australian and New Zealand Microseris
The chloroplast phylogenies (Figs. 6, 7) show the Australian and New Zealand Microseris as a monophyletic group, supporting a single, or at most a few closely spaced in time, colonizing event(s) into Australia or New Zealand. A single origin was anticipated on the basis of its supposed mode of origin (Chambers, 1955
), and is strongly supported by the uniform allotetraploid karyotype found within all members of the taxon (Sneddon, 1977
). The Australian and New Zealand Microseris are closely related to the North American and Chilean annual Microseris and more diverged from the North American perennial species of the genus (Fig. 7). This confirms the results of a cpDNA study by Wallace and Jansen (1990)
and supports their suggestion that an ancestral annual plant has been the maternal parent of the original hybrid. In contrast to the results of Wallace and Jansen (1990)
, our data are inconclusive about the monophyly of the annual Microseris, but instead show the Australasian Microseris to be natural.
Within the chloroplast phylogeny of the Australian and New Zealand Microseris (Fig. 6) we found a basal polytomy. This polytomy results from a lack of mutations rather than of homoplasy, suggesting rapid radiation early in the history of the taxon. "Hard" polytomies, i.e., the fixed attachments of the polytomous node to its descendant nodes (Maddison, 1989
), are indicative of the process of multiple-speciation. In this process, populations would originate via founder events followed by drift that in turn samples the initial genetic variation in the next generations (e.g., Okada, Whitkus and Lowrey, 1997
). Subsequently, the genetic variation within populations declines while differentiation between populations increases. "Hard" polytomies have been reported for highly diversified, relatively closely genetically related taxa on oceanic islands (e.g., Baldwin, Kyhos, and Dvorak, 1990
; Crawford et al., 1993
; Sang et al., 1994
; Mes, Van Brederode, and ‘t Hart, 1996
). Our results indicate that the process of adaptive radiation has occurred similarly within the continental Australasian Microseris as is known for the oceanic island taxa. Due to the "hard" polytomy, a better resolution of the basal relationships within the Australian and New Zealand Microseris is unlikely to be achieved on the basis of additional chloroplast mutations. In addition, the extensive sample size of our study does not support a biased estimate of phylogenetic relationships as a result of missing clade-specific mutations.
Because there is a basal polytomy (Fig. 6), the Australian mainland, Tasmania, or one of the islands of New Zealand are equally likely places for arrival of the founder population after long-distance dispersal from western North America. The chloroplast data show the autofertile "fine-pappus" ecotypes of Australia (F1-F7nt; Table 1; Fig. 6) to be the least diverged from the North American Microseris. This was also indicated by a nuclear DNA study (Van Houten, Scarlett, and Bachmann, 1993
; see introduction). When chloroplast and nuclear DNA are congruent, the chloroplast data support the suggestion that the "fine-pappus" ecotype might be closest to the earliest founders of Microseris in Australia. On the other hand, the results could imply that the mutation rate is reduced in both genomes of the Australian "fine-pappus" ecotypes. In a study of the Hawaiian silversword alliance (Baldwin et al., 1991
) it was demonstrated that the ancestor of the taxon overcame the breeding barrier of self-incompatibility, supposedly by the introduction of at least two individuals. This shows that autofertility in the founder is not per se needed to colonize new areas. In summary, the data are inconclusive about the closest relatives to the earliest founder of Microseris in the Southern hemisphere.
The chloroplast types (Fig. 6) confirm the monophyly of Microseris lanceolata as it was delimited by Sneddon (unpublished data). Within M. lanceolata three clades are recognized that correspond more with geographical distribution than with morphological entities (Figs. 1, 2, 6; Tables 1, 2). Each clade contains at least one population of both the "alpine" and "murnong" ecotypes, and a few populations of either ecotype are polymorphic for Mln-2 and -3 chloroplast types. In addition, one population of the "alpine" ecotype (A7) shares a chloroplast mutation with its neighboring populations of the "murnong" ecotype (number 41; Table 3). The incongruencies between the morphology and chloroplast data indicate that dispersal or introgression has occurred between the populations of different ecotypes within M. lanceolata or that there has been parallel evolution of morphological adaptations. The presence of intermediate morphologies between the "alpine" and "murnong" ecotypes (populations A1nt-A4nt, A8nt, and M20nt; Table 2) also indicates that hybridization or parallel evolution may have occurred among these ecotypes. Due to the incongruencies with the morphology, the chloroplast data do not support a subdivision of M. lanceolata into species or subspecies that comprise the "alpine" and "murnong" ecotypes, respectively, as was earlier suggested by Sneddon (1977)
. Nuclear markers will have to be examined to see whether, for example, the "alpine" adaptations have originated more than once, and whether or not there is a zone of introgression between populations of the "murnong" and "alpine" ecotypes in northeastern Victoria (Fig. 1).
The chloroplast phylogeny (Fig. 6) is inconclusive about the monophyly of Microseris scapigera (Sneddon, unpublished data). The results indicate that M. scapigera might be further subdivided into two or more (sub)species, e.g., one including the Australian and Tasmanian populations of the "fine-pappus" ecotype, one that contains all but one of the New Zealand populations, and a third that comprises the remaining New Zealand population (Fig. 6). The division of the Australasian Microseris into two species was mainly based on morphology and crossability data (Sneddon, 1977
, personal communication). Within this classification the inclusion of the Australian "fine-pappus" ecotype in M. scapigera was uncertain because it showed a low fertility in crosses with the New Zealand members of M. scapigera. The crossability data, morphology (Table 1; Fig. 2), and breeding system, distinguish the "fine-pappus" ecotype of the Australian mainland (F1-F5; Table 2) from the "coastal" one of New Zealand (C1-C4), as do the chloroplast types (Fig. 6). The chloroplast types of the "fine-pappus" ecotypes from the Australian mainland and Tasmania (F6nt, F7nt) are identical, while their morphologies (Table 1) are largely similar, suggesting a close relationship between these populations. The chloroplast DNAs of the New Zealand "fine-pappus" ecotypes (F8nt, F9nt) are more diverged from those of the Australian and Tasmanian ones, and this divergence is partly supported by their morphology (Table 1). The data indicate that F9nt is more similar to the New Zealand "coastal" ecotype than to the other "fine-pappus" populations, while F8nt is different from both the "coastal" and "fine-pappus" ecotypes. Our results are inconclusive about the direction of the interisland dispersals, and it is unclear whether the discrepancy between F8nt and the other New Zealand populations implies that these islands were colonized twice.
Conclusions
The phylogenetic relationships of 53 populations of Australian and New Zealand Microseris were investigated using RFLPs and trnL(UAA)-trnF(GAA) length variants in the chloroplast genome. The results (Figs. 6, 7) indicate that the evolutionary processes that occurred within this continental taxon after long-distance dispersal from western North America were similar to those found in studies of oceanic island taxa (e.g., Baldwin, Kyhos, and Dvorak, 1990
; Crawford et al., 1992
; Francisco-Ortega, Jansen, and Santos-Guerra, 1996
; Okada, Whitkus, and Lowrey, 1997
). The occurrence of "hard" basal polytomies in the most parsimonious trees indicates that the taxon has early and rapidly radiated. The basal polytomies leave questions about the place of arrival of the first individual(s), the subsequent (interisland) dispersals, the direction of the shift in breeding system, and the order in which different ecotypes arose unresolved. The "hard" polytomies and the extensive sample size suggest that a better resolution of basal relationships will not be achieved on the basis of additional chloroplast mutations. The data are inconclusive about the monophyly of the Australasian species Microseris scapigera (sensu Sneddon, unpublished data), while the monophyly of the other species, M. lanceolata (sensu Sneddon, unpublished data), is confirmed. Within the latter, the distribution of RFLPs in the chloroplast genome suggests that different ecotypes have similar chloroplast types. In order to discriminate between likely explanations for this contradiction, such as introgression and parallel evolution of similar adaptations, expanded nuclear DNA investigations are needed.
|
|
|
FOOTNOTES |
|---|
3 Author for correspondence (e-mail: vijverberg{at}bio.uva.nl
).
5 E-mail: bachmann{at}ipk-gatersleben.de
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
———, ———, ———, and G. D. Carr. 1991 Chloroplast DNA evidence for a North American origin of the Hawaiian silversword alliance (Asteraceae). Proceedings of the National Academy of Sciences, USA 88: 1840–1843.
Bremer, K. 1988 The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795–803.[CrossRef][ISI]
Carlquist, S. 1983 Intercontinental dispersal. Sonderbdaen De des Naturwissenschaftlichen Vereins in Hamburg 7: 37–47.
Chambers, K. L. 1955 A biosystematic study of the annual species of Microseris. Contributions from the Dudley Herbarium 4: 207–312.
Crawford, D. J., T. F. Stuessy, M. B. Cosner, D. W. Haines, M. O. Silva, and M. Baeza. 1992 Evolution of the genus Dendroseris (Asteraceae: Lactuceae) on the Juan Fernandez Islands: evidence from chloroplast and ribosomal DNA. Systematic Botany 17: 676–682.[CrossRef][ISI]
———, ———, ———, ———, and ———. 1993 Ribosomal and chloroplast DNA restriction site mutations and the radiation of Robinsonia (Asteraceae; Senecioneae) on the Juan Fernandez Islands. Plant Systematics and Evolution 184: 233–239.[CrossRef][ISI]
Doebley, J. F., D. P. MA, and W. T. Renfroe. 1987 Insertion/deletion mutations in the Zea chloroplast genome. Current Genetics 11: 617–624.[CrossRef][ISI][Medline]
Downie, S. R., and J. D. Palmer. 1992 Use of chloroplast DNA rearrangements in reconstructing plant phylogeny. In P. S. Soltis, D. E. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants, 14–35. Chapman and Hall, New York, NY.
Ebes, H. 1988 The florilegium of Captain Cook's first voyage to Australia and New Zealand 1768–1771, 161, Plate 494 (original collection by Banks and Solander is preserved in Sydney, National Museum of Australia), Sotheby's Australia Pty. Ltd., Melbourne, Australia.
Feinberg, A. P., and B. Vogelstein. 1983 A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Analytical Biochemistry 132: 6–13.[CrossRef][ISI][Medline]
Felsenstein, J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791.[CrossRef][ISI]
Francisco-Ortega, J., R. K. Jansen, and A. Santos-Guerra. 1996 Chloroplast DNA evidence of colonization, adaptive radiation, and hybridization in the evolution of the Macaronesian flora. Proceedings of the National Academy of Sciences, USA 93: 4085–4090.
Gielly, L., and P. Taberlet. 1994 The use of chloroplast DNA to resolve plant phylogenies: noncoding versus rbcL sequences. Molecular Biology and Evolution 11:769–777.
Gott, B. 1983 "Murnong"—Microseris scapigera: a study of a staple food source of Victorian aborigines. Australian Aboriginal Studies 2: 2–18.
Hipkins, V. D., K. A. Marshall, D. B. Neale, W. H. Rottmann, and S. H. Strauss. 1995 A mutation hotspot in the chloroplast genome of a conifer (Douglas-fir: Pseudotsuga) is caused by variability in the number of direct repeats from a partially duplicated tRNA gene. Current Genetics 27: 527–579.
Hiratsuka, J., H. Shimada, R. Whittier, T. Ishibashi, M. Sakamoto, M. Mori, C. Kondo, Y. Honji, C-R. Sun, B-Y. Meng, Y-Q. LI, A. Kanno, Y. Nishizawa, A. Hirai, K. Shinozaki, and M. Sugiura. 1989 The complete sequence of the rice (Oryza sativa) chloroplast genome: intermolecular recombination between distinct tRNA genes accounts for a major plastid DNA inversion during the evolution of the cereals. Molecular and General Genetics 217: 185–194.
Hombergen, E-J., and K. Bachmann. 1995 RAPD mapping of three QTLs determining trichome formation in Microseris hybrid H27 (Asteraceae: Lactuceae). Theoretical and Applied Genetics 90: 853–858.[ISI]
Howe, C. J., R. F. Barker, C. M. Bowman, and T. A. Dyer. 1988 Common features of three inversions in wheat chloroplast DNA. Current Genetics 13: 343–349.[CrossRef][ISI][Medline]
Jansen, R. K., and J. D. Palmer. 1987 Chloroplast DNA from lettuce and Barnadesia (Asteraceae): structure, gene localization, and characterization of a large inversion. Current Genetics 11: 553–564.
Kreitman, M., and M. Aguadé. 1986 Genetic uniformity in two populations of Drosophila melanogaster as revealed by filter hybridization of four-nucleotide-recognizing restriction enzyme digests. Proceedings of the National Academy of Sciences, USA 83: 3562–3566.
Lems, K. 1960 Botanical notes on the Canary Islands. II. The evolution of plant forms in the islands: Aeonium. Ecology 41: 1–47.
Maddison, W. 1989 Reconstructing character evolution on polytomous cladograms. Cladistics 5: 365–377.[ISI]
Mes, T. H. M., J. Van Brederode, and H. ‘T Hart. 1996 Origin of the woody Macaronesian Sempervivoideae and the phylogenetic position of the east African species of Aeonium. Botanica Acta 109: 441–506.[ISI]
Okada, M., R. Whitkus, and T. K. Lowrey. 1997 Genetics of adaptive radiation in Hawaiian and Cook islands species of Tetramolopium (Asteraceae; Astereae). I. Nuclear RFLP marker diversity. American Journal of Botany 84: 1236–1246.[Abstract]
Palmer, J. D., R. K. Jansen, H. J. Michaels, M. W. Chase, and J. R. Manhart. 1988 Chloroplast DNA variation and plant phylogeny. Annals of the Missouri Botanical Gardens 75: 1180–1206.[CrossRef][ISI]
Prober, S. M., L. H. Spindler, and A. H. D. Brown. 1998 Conservation of the grassy white box woodlands: effects of remnant population size on genetic diversity in the allotetraploid herb Microseris lanceolata. Conservation Biology 12:1279–1290.
Quigley, F., and J. H. Weil. 1985 Organization and sequence of five tRNA genes and of an unidentified reading frame in the wheat chloroplast genome: evidence for gene rearrangements during evolution of chloroplast genomes. Current Genetics 9: 495–503.[CrossRef][ISI][Medline]
Roelofs, D., and K. Bachmann. 1995 Chloroplast and nuclear DNA variation among homozygous plants in a population of the autogamous annual Microseris douglasii (Asteraceae, Lactuceae). Plant Systematics and Evolution 196: 185–194.[CrossRef][ISI]
———, and ———. 1997 Comparison of chloroplast and nuclear phylogeny in the autogamous annual Microseris douglasii (Asteraceae, Lactuceae). Plant Systematics and Evolution 204: 49–63.[CrossRef][ISI]
———, J. Van Velzen, P. Kuperus, and K. Bachmann. 1997 Molecular evidence for an extinct parent of the tetraploid species Microseris acuminata and M. campestris (Asteraceae, Lactuceae). Molecular Ecology 6: 641–649.[CrossRef][Medline]
Saghai-Maroof, M. A., K. M. Soliman, R. A. Jorgensen, and R. W. Allard. 1984 Ribosomal DNA spacer-length mutations in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proceedings of the National Academy of Sciences, USA 81: 8014–8018.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989 Molecular cloning, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Sang, T., D. J. Crawford, S.-C. Kim, and T. F. Stuessy. 1994 Radiation of the endemic genus Dendroseris (Asteraceae) on the Juan Fernandez Islands: evidence from sequences of the ITS regions of nuclear ribosomal DNA. American Journal of Botany 81: 1494–1501.[CrossRef][ISI]
———, ———, and T. F. Stuessy. 1997 Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae). American Journal of Botany 84: 1120–1136.[Abstract]
Schilling, E. E., J. L. Panero, and U. H. Eliasson. 1994 Evidence from chloroplast DNA restriction site analysis on the relationships of Scalesia (Asteraceae: Heliantheae). American Journal of Botany 81: 248–254.[CrossRef][ISI]
Shimada, H., and M. Sugiura. 1989 Pseudogenes and short repeated sequences in the rice chloroplast genome. Current Genetics 16: 293–301.[CrossRef][ISI][Medline]
Shinozaki, K., M. Ohme, M. Tanaka, Et Al. 1986 The complete nucleotide sequence of the tobacco chloroplast genome: its gene organization and expression. EMBO Journal 5: 2043–2049.[ISI][Medline]
Sneddon, B. V. 1977 A biosystematic study of Microseris subgenus Monermos (Compositae: Cichorieae). Ph.D. diss
