| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Population Biology |
The University of Georgia, Institute of Ecology, Savannah River Ecology Laboratory, Drawer E, Aiken, South Carolina 29802 USA
Received for publication August 23, 2004. Accepted for publication March 28, 2005.
ABSTRACT
Genetic and clonal diversity vary between two closely related cattail species (Typha angustifolia and T. latifolia) from Ukraine. This diversity was calculated from microsatellite data. Forty-eight percent of the total variation was partitioned between species, which formed distinct clusters in a dendrogram with no indication of hybrid populations. Typha angustifolia had higher heterozygosity at the species (Hes = 0.66) and population (Hep = 0.49) levels than did T. latifolia (Hes = 0.37 and Hep = 0.29, respectively). The higher number of alleles in T. angustifolia may be indicative of larger effective population sizes due to its higher seed production. Clonal diversity of T. angustifolia was lower than that of T. latifolia (Ng/Nr = 0.40 and 0.61, Simpson's D = 0.82 and 0.94, respectively). Correlations between clonal and genetic diversity were higher for T. latifolia than T. angustifolia, suggesting that the importance of factors and their interactions affecting this relationship are different for the two species. Latitudinal and longitudinal trends were not observed in either species despite the large sampling area. Population differentiation was relatively high with FST of 0.24 and 0.29 for T. angustifolia and T. latifolia, respectively. Weak isolation by distance was observed for T. latifolia but not for T. angustifolia.
Key Words: cattails • clonal structure • genetic diversity • microsatellites • Typha angustifolia • Typha latifolia • Ukraine
Asexual reproduction occurs widely in plants and is also found in lichens, fungi, and some animal groups (Stuefer et al., 2002
). Many plants use both sexual and clonal reproductive modes (Richards, 1986
). The extent to which these are used can vary widely within and among species. This variation in the mode of reproduction has significant consequences for the ecology, genetics, and evolution of plants (Eckert, 2002
). Clonal and population genetic diversities may be independent and influenced by different factors (Hangelbroek et al., 2002
). Several studies show that genetic variation in clonal plants can be similar to that of nonclonal plants (Hamrick and Godt, 1989
). This may partially explain the ability of widely distributed plants to adapt to a variety of environmental conditions. Although widespread plants are likely to have relatively high genetic diversity, geographic range can be a poor predictor of genetic structure (Loveless and Hamrick, 1984
; Hamrick and Godt, 1989
) as indicated by different levels of genetic diversity among species of widespread plants.
The breeding system is an important factor in determining population genetic structure. Many clonal plants have a mixed mating system with both outcrossing and selfing. Self-pollination can occur not only within the same flower (autogamy), but also between different ramets of the same genet (geitonogamy). Selfing, which mostly occurs through geitonogamy, may lead to inbreeding depression (Eckert, 2000
); this explains why progeny from outcrossing are about twice as fit as those from selfing (Eckert and Barrett, 1994
). A low frequency of outcrossing events may be sufficient to maintain genetic diversity and fitness in clonal plants.
Depending on the rates of seedling recruitment, clonal species can show one of two strategies: "initial seedling recruitment" (ISR) and "repeated seedling recruitment" (RSR, Eriksson, 1997
; Eckert, 2002
). These types of recruitment can determine genetic structure at the genet or ramet level (Travis et al., 2004
). Species with RSR consistently produce genets of different sizes, while new genets may become established only when open spaces are available for the offspring of ISR species. Cattails, which are primarily an ISR species, are emergent wetland plants that combine both sexual and vegetative modes of reproduction. Within an established cattail stand, seedlings may be unable to survive because of competition with older ramets (McNaughton, 1966
; Grace, 1985
). Disturbance can be important for preserving or introducing genetic variability in cattail populations by creating open spaces for seedling recruitment and growth resulting in more fragmented populations. Populations of different or even the same species may show very wide ranges of genetic and clonal diversities depending on different ecological and biological factors.
Different markers have been used to study genetic variation in widely distributed plants, including cattails. These markers reveal different levels of variability. Even though allozymes have been used successfully to study breeding structure of clonal plants (Ellstrand and Roose, 1987
; Widen et al., 1994
), almost no polymorphism was found in populations of T. latifolia and T. domingensis from the eastern United States (Mashburn et al., 1978
; Sharitz et al., 1980
). Allozymes allowed clear identification of Typha species (T. angustifolia, T. latifolia, and T. domingensis, Sharitz et al. 1980
) and showed that T. glauca is not an F1 hybrid between T. latifolia and T. angustifolia but has some intermediate and some fixed characteristics of both species. Typha glauca has also been described using morphological and biochemical characteristics as a species that resulted from introgression between the two species (Fassett and Calhoun, 1952
; Lee, 1975
). In contrast, a study using random amplified polymorphic DNA (RAPDs) reported T. glauca as an F1 hybrid (Marcinko Kuehn et al., 1999
). These studies were conducted on samples from different locations. The markers apparently give different interpretations of what constitutes an F1 hybrid or T. glauca varies across its range.
The complete lack of intraspecific variation at allozyme loci in the eastern United States (Sharitz et al., 1980
) was surprising given the results of McNaughton's (1966
, 1967
) studies, that showed ecotypic variation at morphological, physiological, and biochemical levels. McNaughton (1967)
suggested that this variation has a genetic basis, and Suda et al. (1977)
indicated that genotypic flexibility and plasticity may be critical for Typha's adaptation to stressful environments. Keane et al.'s (1999)
study was the first to show intraspecific variation among populations of T. latifolia using a variable number of tandem repeats (VNTRs). His heterozygosity estimates were the lowest detected for plants using VNTR loci. Microsatellites, which are co-dominant loci with high mutation rates of 10–3 to 10–4, are often used as genetic markers (Ellegren, 2000
). Microsatellite loci in T. angustifolia and T. latifolia have relatively high numbers of alleles (Tsyusko-Omeltchenko et al., 2003
) making them useful for study and comparisons of genetic and clonal diversity among populations between and within Typha species. Microsatellites may also be specific enough to detect populations of hybrid origins if they exist. Despite the high level of variation for microsatellites, there have been few studies of population structure of clonal plants using them (Reusch et al., 2000
).
Our primary objective was to compare clonal and genetic diversity for two Typha species that live in slightly different habitats in Ukraine but have different life history characteristics. We hypothesized that the species with greater reproductive potential, T. angustifolia, should have higher genetic and clonal diversity than the other species, T. latifolia. We compared the clonal and genetic diversities of the two species using the same microsatellite loci. If clonal and genetic diversities are independent, then there should be no correlation between them. Because the two species occurred over a wide geographic area, we also tested for latitudinal and longitudinal clines for both types of diversity. The relationships of differentiation among populations and correlations between pairwise genetic and geographic distances among intraspecific populations indicated the importance of dispersal and outbreeding among populations. Finally, we generated a tree including populations of both species to test for the occurrence of populations in which hybridization has altered genetic and clonal diversity.
MATERIALS AND METHODS
Study species
Typha latifolia and T. angustifolia are wetland plants with world-wide distributions from the Arctic Circle to about 30°S (Sculthorpe, 1967
). Typha is a perennial plant that grows successfully in a variety of habitats including fresh and brackish water, deep marshes, or shallow roadside ditches. When found together, the two species are often segregated by water depth with T. latifolia living in shallow and T. angustifolia deeper water. Typha latifolia is a much better competitor in shallow water and in dense stands due to its greater ability for light capture and tolerance for shading. When they occur together in shallow water, T. latifolia may completely replace T. angustifolia (Grace and Wetzel, 1981
). While reproducing through rhizomes, Typha forms dense stands where other plants are excluded. The rhizomes remain viable for 17–22 mo and maintain physiological connection between ramets for about two yr (Dickerman and Wetzel, 1985
). Typha latifolia produces more rhizomes than T. angustifolia, and therefore its vegetative reproduction may be more enhanced. Male and female inflorescences are located on the same shoot, and selfing is one of the reproductive modes. Selfing (autogamy and geitonogamy) occurs in up to 70% of the matings in Typha populations (Kratinger, 1975
). Seed production is extensive with 117 000– 268 000 small seeds from a single ramet. Typha angustifolia produces more seeds than T. latifolia, allowing increased opportunities for dispersal. Each fruit is equipped with hairs for wind dispersal and contains one seed, which is released as soon as the fruit touches water. Several conditions (warm temperature, moisture, shallow water, and decreased oxygen concentration) are critical for seed germination, and the percentage of successfully germinated seeds is low.
Sampling sites
We sampled 13 and 11 Ukrainian populations of T. angustifolia and T. latifolia, respectively (Fig. 1). Samples (N = 659) were collected within populations of both species at distances of at least 1 m or more apart during July–August of 2001 and 2002. The distances among populations varied from 25–700 km. Populations were chosen from five geographic regions: Transcarpathia, Cherkasska, Zhitomirska, Chernigovska, and Kievska. A geographic positioning system was used to determine latitude and longitude at each location. There were five sites in Transcarpathia (Borony, 49°6' N, 23°5' E; Batevo, 48°44' N, 21°58' E; Zarichevo, 48°46' N, 22°4' E; Glubokoe, 48°47' N, 22°8' E; Chop, 48°44' N, 21°59' E), two sites in Cherkasska (Buzivka, 50°20' N, 26°46' E; Uman', 50°13' N, 26°24' E), one site in Zhitomirska (Brusilov, 50°51' N, 28°17' E), three sites in Chernigovska (Mekshunovka, 51°13' N, 29°21' E; Gubichi, 51°10' N, 29°13' E; Chernigov, 51°08' N, 29°07' E), and five sites in Kievska (Kiev, 50°57' N, 28°36' E; Belaya Tserkov, 50°40' N, 27°45' E; Ivankovo, 51°12' N, 29°19' E; Atashev, 51°15' N, 29°28' E; and Paryshev, 51°16' N, 29°30' E). There were three sites with relatively small areas of approximately 40 m2 (Borony, Zarichevo, and Glubokoe) and three (Brusilov, Kiev, and Ivankovo) with large areas, up to 500 m2. Most of the sites were located in disturbed areas except Ivankovo, Brusilov, and Chernigov. The other sampling sites occupied areas of 200–300 m2. The Ivankovo, Atashev, and Paryshev populations of T. angustifolia were located close (35–40 km) to the failed Chornobyl nuclear reactor. Typha latifolia populations were located at least 90 km from Chornobyl. The species co-occurred at eight sites (Fig. 1).
|
. All loci were isolated from T. angustifolia. A Gensize Rox 500 ladder Genpak (Genetix, Boston, MA, USA) was mixed with PCR product, and the mixture was run on an ABI 377 sequencer (Applied Biosystems, Foster City, California, USA) to determine microsatellite allele sizes. Allele scoring was conducted using Genescan and Genotyper software (Applied Biosystems). Data were transformed into Genepop and Arlequin formats using a genotyper-genepop converter program (www.today.myip.org).
Data analyses
Genetic diversity characteristics including percentage of polymorphic loci (Pp), mean allele number (MAN), average observed and expected heterozygosities (Ho and He, respectively), and variance of allele size for each population were calculated. Genepop version 3.1 (Raymond and Rousset, 1995
) was used for calculation of Ho and He, and RSTCALC (Goodman, 1997
) for mean allele size (MAS) and its variance. MAS and its variance are useful variables for species comparisons, because they may be indicative of differences in microsatellite mutation rates between species. There are positive correlations between allele size and mutation rate for many species (Ellegren, 2000
). The difference in size between two alleles at the same locus (allele span) is another important factor related to microsatellite mutation rates (Ellegren, 2000
). As the allele span increases, the variance of MAS becomes larger. Tests for deviations from Hardy-Weinberg expectations and for linkage disequilibrium were conducted using Genepop. Sequential Bonferroni corrections were applied to estimate significance where appropriate (Rice, 1988
).
Because cattail reproduction is primarily vegetative, several samples (ramets) can have the same genotype and belong to one genetic individual (genet). To differentiate between ramets and genets, we calculated expected frequencies of multilocus genotypes that occurred in more than one sample. These frequencies were used to calculate Pgen, the probability that two samples have the same genotype by chance (Reusch et al., 2000
). Ramets with the same genotype were considered one genet when Pgen < 0.05. Genetic characteristics were calculated for each population for both genets and ramets to detect differences between them. Comparisons between estimates based on ramet and genet data for each population were conducted using paired t tests. Three basic genotypic diversity characteristics were calculated for each population. The proportion of distinguishable genotypes was calculated as a ratio of number of genets (Ng) to that of ramets (Nr). Simpson's diversity index (D) corrected for sample size was calculated according to Pielou (1969)
. Fager's (1972)
evenness index (E) was used to estimate the distribution of genets in a population. To test whether populations with higher genetic diversities also have higher clonal diversities, we calculated correlations between these diversity estimates.
Individual Hoi, MANi, MASi, and its variance were calculated for each genet. These data were not normally distributed, had heteroscedastic variances, and none of the applied transformations (square root, log, or arcsine) solved these problems; therefore, the values were ranked, and Duncan's (1955)
multiple comparisons test was used on the ranked values to test for differences among populations. Wilcoxon rank sum tests (Sokal and Rohlf, 1995
) were used to compare genetic and genotypic characteristics between species and between sites where species co-occur with the sites containing a single species. The total number of alleles for each species was calculated. Alleles with frequency above 0.30 were considered common alleles. The number of common alleles shared between the two species was counted. Histograms of number of alleles for all populations for each species were constructed using 11 frequency intervals. Chi square was used to test whether the number of alleles per frequency class was the same in both species. The first two categories, which had the lowest allele frequencies, were sequentially removed from these analyses to determine whether distributional differences between the species were mostly due to these categories. Most statistical tests were conducted using SAS version 8.1 (SAS, 2000
).
Analyses of molecular variance (AMOVA; Excoffier et al., 1992
) were applied to partition variance between and within species using ARLEQUIN (Schneider et al., 2000
). An AMOVA accounts for either frequency or size differences between pairs of different haplotypes and was performed using both options (FST or RST, respectively). Only genet data were used to test for differentiation among populations. Regression analyses of genetic and genotypic diversity characteristics with latitude or longitude were conducted to test for clinal relationships. Genetic distances (FST and RhoST) were calculated using FSTAT (Goudet, 1995
) and RSTCALC (Goodman, 1997
), and matrices of pairwise genetic distances were constructed for each species. Geographic distances were determined from longitudinal and latitudinal data (Viard et al., 1997
). Correlations between genetic and geographic matrices were calculated with Mantel's (1967)
procedure using 10 000 permutations. Mantel tests were also used to calculate correlations between FST and RhoST matrices for each species. Cavalli-Sforza and Edwards (1967
; DC) distances were also calculated to examine relationships among populations. Five subprograms of PHYLIP, version 3.5 (Felsenstein, 1993
) were used to generate the tree. Each allele frequency matrix was resampled 1000 times with SEQBOOT; GENDIST was used to calculate DC's for each bootstrap matrix; unrooted neighbor-joining trees were generated with NEIGHBOUR. A consensus tree was created using CONSENSE and DRAWGRAM.
Evidence of population bottlenecks within both species was tested using Bottleneck, version 1.2.02 (Cornuet and Luikart, 1997
) and AGARst (Harley, 2001
). Indication of recent bottlenecks can be informative about the past population dynamics of the species. Distributions of individual heterozygosities expected from the observed allele frequencies for each population and locus were calculated with Bottleneck using the assumptions of three different mutation models: the infinite allele model (IAM), the stepwise mutation model (SMM), and the two-phase model (TPM). Wilcoxon signed-ranks tests were used in the program to detect deviations from expected heterozygosities. To detect reductions in population size, the mean ratio of the number of alleles to the range in allele size (M; Garza and Williamson, 2001
) for each locus was calculated with AGARst (Harley, 2001
). A recent bottleneck was indicated when M < 0.68. Statistical significance was indicated when P 
0.05.
RESULTS
Species (s) diversity level, calculated from the pooled data, was higher for T. angustifolia than T. latifolia (Pps = 0.89 vs. 0.65; Hes = 0.66 vs. 0.37; Hos = 0.50 vs. 0.25 and MANs = 11.64 vs. 6.22, respectively). Characteristics of within-population variability (He, Ho, MAN, mean allele size, and mean variance of allele size) calculated from ramet and genet data for both species are listed in Tables 1 and 2. Probabilities for multilocus genotypes occurring more than once (Pgen) were always <0.05, and this facilitated differentiation between genets and ramets. Intermingled clones that consisted of at least two identical ramets were frequently observed within stands. Both species demonstrated significant differences between data for ramets and genets for He and mean variance of allele size (P < 0.05), and T. latifolia also showed significant differences for Ho.
|
|
2 = 22.0, P < 0.05), but not T. latifolia (2 = 9.9, P > 0.05). Simpson's D was relatively high for nine populations of T. angustifolia (0.827–0.975) and for all populations of T. latifolia (0.842–0.981). Genotypes were spatially evenly distributed in all but one population (Batevo, E = 0.064) of T. latifolia and three populations (Gubichi, Kiev, and Batevo) of T. angustifolia (E = 0.042–0.065).
|
2 = 23; df = 7; P < 0.005, Fig. 2). Significant differences were also observed after data from the first and second frequency categories were removed from the analyses (2 = 17; df = 6; P < 0.01 and 2 = 22; df = 5; P < 0.001, respectively). There were 10 common alleles per species (the alleles with frequency above 0.30), and none of these alleles were shared between species. When tested for correlations between genetic and clonal diversities, T. latifolia showed significant positive relationships of He and MAN with proportion of distinguishable genotypes and Simpson's D (r2 varied from 0.39 to 0.62 and P < 0.05). MAN was positively correlated with Ng/Nr and D in T. angustifolia (r2 = 0.33–0.39 and P < 0.05).
|
More variance was partitioned between species (48% with FST and 76% with RST) than among or within populations of each species (Table 4). There was more variance in each species distributed within than among populations when FST or RST options of AMOVA were used. The percentage of variation within and among populations calculated with FST was different from those calculated with RST for both species (Table 4). RhoST and FST matrices correlated significantly in T. angustifolia (rm = 0.82 and P = 0.01) and T. latifolia (rm = 0.61 and P = 0.02). Pairwise FST and RhoST values varied from 0.07 to 0.55 for T. angustifolia (with averages of 0.24 for FST and 0.26 for RhoST) and from 0.06 to 0.41 for T. latifolia (with averages of 0.29 for FST and 0.22 for RhoST). There was a small but significant correlation between matrices of pairwise genetic (FST or RhoST) and geographic distances for T. latifolia (rm = 0.26 and P = 0.04) but not T. angustifolia (Fig. 3). Longitudinal and latitudinal clines were observed only for one characteristic (MAS, r2 = 0.45 and 0.46, respectively) in T. angustifolia, but T. latifolia failed to demonstrate any clines.
|
|
|
The two closely related Typha species show some similarity but also substantial differences in genetic and clonal diversities. Inbreeding occurs in both species, and the levels of clonal and genetic diversities vary greatly among populations of each species. Typha angustifolia has higher genetic but lower clonal diversity than T. latifolia. Detecting higher genetic diversity in T. angustifolia is expected due to ascertainment bias (Ellegren et al., 1995
). Population differentiation is relatively high in both species, but isolation by distance is detected only for T. latifolia. Neither species shows latitudinal or longitudinal clines, but different factors are likely to influence clonal and/ or genetic diversities in each species, because these diversities are more independent in T. angustifolia than T. latifolia.
Within-population variability
The amount of variation detected for microsatellite loci in both Typha species is higher than previously demonstrated for allozymes (Mashburn et al., 1978
; Sharitz et al., 1980
) or VNTR loci (Keane et al., 1999
). In T. latifolia, lower levels of variation were observed for VNTRs vs. microsatellites for MAN (1.28 vs. 2.63), proportion of polymorphic loci (0.16 vs. 0.66), and Ho (0.08 vs. 0.26). Various markers have different mutation rates and are expected to show differences in their polymorphism. However, mutation rates of VNTR loci should be similar to those of microsatellites or even higher, because unlike allozymes they are both noncoding, selectively neutral markers (Jeffreys et al., 1988
). The microsatellites used in this study were dinucleotides, whereas VNTRs in Keane's study contained tetranucleotide repeats. In addition, the samples in this and Keane's studies were collected from two different geographic regions (Ukraine and United States, respectively), and many more samples were taken from a larger geographic area in this study, so levels of variability may or may not vary between these regions. Microsatellites are thus good markers for examining genetic and clonal diversities in plants such as Typha in which allozymes showed almost no polymorphism.
Genotypic diversity of both Typha species is relatively high for the proportion of distinguishable genotypes (Ng/Nr = ca. 0.58) and Simpson's D (ca. 0.90) but not E (ca. 0.40) when compared to the allozyme results for 21 (Ellstrand and Roose, 1987
) and 47 (Widen et al., 1994
) plant species with asexual reproduction (Ng/Nr = 0.17 and 0.27; D = 0.62 and 0.75; and E = 0.68 and 0.75, respectively). Genotypic diversity of T. latifolia calculated for VNTR loci (Keane et al., 1999
) has a lower Ng/Nr (0.39) than that of microsatellite loci (0.61). The average Ng/Nr calculated from microsatellite data is lower for Typha than for Zostera marina (0.74, Reusch et al., 2000
) and Elymus athericus (0.92, Bockelmann et al., 2003
). Thus, two genotypic characteristics (Ng/Nr and E) calculated from microsatellite data for both Typha species were overall higher than VNTR's estimates, but Ng/Nr was lower than microsatellite estimates for other clonal plants probably because of Typha's reproductive characteristics. The over- and underestimations of Typha population genetic diversity may occur because of significant differences between estimates for ramets and genets, and this may be a problem for these types of comparisons.
The degree to which plant species are sexual and asexual varies among species and even populations of the same species (Eckert et al., 2003
). A substantial difference in proportion of distinguishable genotypes was detected between estimates for two Saggitaria species that were based on the same allozymes (Edwards and Sharitz, 2003
). Both Typha species also showed large variation among populations in clonal diversity. In Widen et al.'s (1994)
survey of clonal plants, about 91% of populations were multiclonal and 48% showed some monoclonality (Stenström et al., 2001
). Different factors such as age and level of disturbance could have influenced clonal diversity of these populations. Clonal diversity varies with population age in Spartina alterniflora (Travis et al., 2004
). Disturbance may introduce new open spaces for seed recruitment and increase genetic and clonal diversity in ISR plants such as Typha. The amount of geitonogamy, which may lead to inbreeding depression (Eckert and Barrett, 1994
), is higher in larger than in smaller fragmented clones (Travis et al., 2004
), and the occurrence of the latter may be more frequent with an increase in disturbance level.
Significant differences in genetic diversity were also common among populations of each species. Individual genetic characteristics of T. angustifolia were lower in the eastern population (Mekshunovka) but higher in the northern populations (Ivankovo and Atashev). There were two southern populations of T. latifolia, Buzivka and Uman', with the highest and lowest values for genetic diversity, respectively. The two northern populations of T. angustifolia were located in close proximity to Chornobyl. Radiation-induced mutations may be partially responsible for the higher values of these populations (Tsyusko, 2004
). However, the third population (Paryshev), which was also located close to Chornobyl, did not show a similarly high level of genetic diversity. The T. latifolia Buzivka population had high genetic variability but was 250 km away from the failed reactor. The question of changes in population genetic diversity as a result of radiation exposure cannot be addressed here, because we have not included data for contaminated populations. However, an increase in microsatellite mutation rates was documented previously for plant populations growing at Chornobyl (Kovalchuk et al., 2000
).
Differences in genetic variability among Typha populations could also be due to their phylogeographic structure. However, there were no distinct clades associated with geographical regions observed for either species (Fig. 4). Clines were observed over longitude and latitude only for MAS in T. angustifolia, but this effect is probably due to chance because of the number of tests that were conducted and the large spatial gap seen in Fig. 1. The high interpopulational differences in genetic variability between adjacent populations of both species make it unlikely that geographical area can be used to explain these differences.
The high variability of certain populations like Buzivka, where both species co-occurred, could have been due to hybridization. However, both species are clearly separated on the dendrogram with a bootstrap value of 100 (Fig. 4), which indicates a lack of hybrid populations. In addition, there were no individuals that were heterozygous for the species-specific common alleles with frequencies over 0.30. Genetic diversities were not significantly higher when compared between sites where species co-occurred and sites with one species. However, because samples were taken only from plants with clearly identified species-specific morphological characteristics, the probability of collecting hybrids was low. Establishment of populations by a few founders from several populations with different gene frequencies followed by breeding among genets could also cause high genetic diversity. Different factors are responsible for changing genetic diversity, and spatial population heterogeneity seems to be the rule for many plants, including Typha.
Close inbreeding occurs commonly in Typha (McNaughton, 1966
). Inbreeding increases homozygote frequency relative to that expected with random mating (Hartl, 2000
). Significant homozygote excesses were observed in three populations of T. angustifolia and two of T. latifolia. Wahlund effect could also be responsible for the heterozygote deficiencies, but the distances between Typha samples within a stand were relatively small, and the existence of two subpopulations over this short distance is unlikely. Intermingled clones also occur frequently within stands. Inbreeding rather than Wahlund effect probably accounts for the deviations from Hardy-Weinberg equilibrium in our study. Without knowledge of the history of the populations, it is not possible to completely explain the differences in genetic and genotypic variation among Typha populations.
The two closely related Typha species showed similarity in evenness of their spatial genotypic distributions, but they differed significantly in all other individual and population genetic and genotypic characteristics. Typha angustifolia was more variable than T. latifolia for genetic but not clonal diversity. The correlations between genetic and clonal diversities were higher in T. latifolia than T. angustifolia. This suggests that clonal diversity may have more effect on genetic diversity of T. latifolia. Clonal and genetic diversities are independent and affected by different environmental factors in Carex and Potamogeton (Stenström et al., 2001
; Hangelbroek et al., 2002
). The situation may be similar for T. angustifolia, and this may explain its higher genetic but lower clonal diversity when compared to T. latifolia. Higher genetic variability of T. angustifolia is at least partially due to an ascertainment bias (Ellegren et al., 1995
): microsatellite alleles tend to be longer in the species for which the primers were first developed, which in our case was T. angustifolia (Tsyusko-Omeltchenko et al., 2003
). There may still be a difference in MAS and its variance between the species, suggesting higher mutations rates in T. angustifolia than T. latifolia. Differences among other characteristics of the two Typha species could also be important in explaining their genetic variability. The distributions of the number of alleles per frequency class were significantly different between the species. Typha angustifolia had more alleles, and many of them had low frequency (Fig. 2), which may indicate larger effective population sizes in this species than T. latifolia and is probably a consequence of their reproductive characteristics. Typha angustifolia has greater seed production and lower production of rhizomes than T. latifolia (McNaughton, 1966
), which may be important for dispersal and colonizing new habitats.
Among-population variability
There was 48% of the total genetic variation distributed between the two species. Within each species, more variability of differences in haplotype frequency was distributed within (ca. 75%) than among populations (ca. 25%). The microsatellite variation among Typha populations was similar to that of other clonal plants such as Zostera marina (29%; Reusch et al., 2000
), but higher than estimates for Elliottia racemosa (18%; Godt and Hamrick, 1999
) or Elymus athericus (14%; Bockelmann et al., 2003
). The high variation among Typha populations is probably due to their high selfing rates and to extensive vegetative reproduction (Keane et al., 1999
), that reduce gene flow among populations.
Decreased gene flow has been documented for selfing plants (Hamrick and Godt, 1989
; Schoen and Brown, 1991
) and animals (Jarne and Charlesworth, 1993
; Viard et al., 1997
). Most pairwise population-genetic distances in both Typha species were relatively high, as expected with limited gene flow. The FST values could have been underestimated depending on the level of polymorphism of the microsatellite loci used (Hedrick, 1999
). In T. angustifolia six loci had heterozygosity estimates larger than 0.8, and in T. latifolia there were three loci with such high heterozygosities (Tsyusko-Omeltchenko et al., 2003
). Because the level of differentiation cannot exceed the level of homozygosity estimated with the same markers (Hedrick, 1999
), the FST values were closer to saturation in T. angustifolia than T. latifolia. The occurrence of bottlenecks was higher in the first than in the second species, and population differentiation may have increased substantially because of the bottlenecks (Hedrick, 1999
). Thus, care should be taken while interpreting population differentiation, because it may depend on the level of the marker's polymorphism, recent bottlenecks, and constraints on allele size and back mutations (Nauta and Wessing, 1996
; Hedrick, 1999
).
The combination of population differentiation and gene flow normally produces positive correlation between genetic and geographic distances (Reusch et al., 2000
; Oleksyk, 2001
; Bockelmann et al., 2003
), which is used as evidence for isolation by distance (IBD). The correlation between genetic and geographic distances, although significant, was relatively low for T. latifolia and even lower and nonsignificant for T. angustifolia (Fig. 3). Higher gene flow is expected among populations located close to each other when IBD occurs, and it decreases as the distance among them increases. The lack of correlation between genetic and geographic distances in T. angustifolia is unexpected given the species biology: T. angustifolia because of higher seed production presumably has more founder events than does T. latifolia. Founder events interacting with asexual reproduction may be one of the major causes for creating more scatter about the trend with geographic distance. The positive slope of these relationships is similar for both species, but the scatter of the points is larger in T. angustifolia than in T. latifolia (Fig. 3). The IBD observed in T. latifolia may also be partially explained by chance variation among the smaller number of its populations included in the analysis. Relatively isolated founding populations that expand through asexual reproduction can produce high differentiation even among adjacent Typha populations.
In conclusion, the genetic and clonal diversities are higher than found in previous studies of Typha using allozymes and VNTRs. There was some similarity between the two Typha species: both of them showed substantial variation in levels of genetic and clonal diversities and high differentiation among populations. However, the two species formed two separate groups on the dendrogram without any indication of hybrid populations. Different factors are probably responsible for the observed spatial heterogeneity in each species as indicated by stronger correlations between genetic and clonal diversities in T. latifolia when compared to those of T. angustifolia. Both types of diversities seem to be more independent in the latter species, which may explain its higher genetic but lower clonal diversity. Higher genetic diversity is expected in T. angustifolia due to its greater reproductive potential but can be partly due to ascertainment bias in the loci used. The higher number of alleles and more low-frequency alleles in T. angustifolia may, however, also be an indication of larger effective population sizes in this species than in T. latifolia. Spatial heterogeneity is relatively high among populations of both species, and isolation by distance was observed only for T. latifolia. Latitude and longitude do not seem to be affecting genetic and/ or clonal diversities of either of the species despite the large areas that were sampled in Ukraine.
FOOTNOTES
1 The authors thank I. Bilanin, J. Goryanaya, I. Chizhevskij, and A. Shulga for assistance with sampling; M. Bondarkov for support at the International Radioecology laboratory in Ukraine; V. Omeltchenko for help with data management; T. Oleksyk, J. Hamrick, and C. Dallas and two anonymous reviewers for valuable suggestions and recommendations; and J. Unrine and M. Wilson for statistical advice. This work is part of a Ph.D. dissertation in the Interdisciplinary Toxicology Program through the Institute of Ecology of The University of Georgia. The research was supported by the Environmental Remediation Sciences Division of the Office of Biological and Environmental Research, U.S. Department of Energy, through Financial Assistance Award No. DE-FC09-96-SR18546 to the University of Georgia Research Foundation and a Sigma Xi grant. 
2 Author for correspondence (tsyusko{at}srel.edu
)
LITERATURE CITED
Bockelmann A. C. T. B. H. Reusch R. Bijlsma P. Bakker 2003 Habitat differentiation vs. isolation-by-distance: the genetic population structure of Elymus athericus in European salt marshes. Molecular Ecology 12: 505-515[CrossRef][Medline]
Cavalli-Sforza L. L. A. W. F. Edwards 1967 Phylogenetic analysis: models and estimation procedures. American Journal of Human Genetics 19: 233-257
Cornuet J. M. G. Luikart 1997 Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001-2014[ISI]
Dickerman J. R. G. Wetzel 1985 Clonal growth in Typha latifolia: population dynamics and demography of ramets. Journal of Ecology 73: 535-552[CrossRef][ISI]
Duncan D. B. 1955 Multiple range and multiple F tests. Biometrics 11: 1-42
Eckert C. G. 2000 Contributions of autogamy and geitonogamy to self-fertilization in a mass-flowering, clonal plant. Ecology 81: 532-542[CrossRef][ISI]
Eckert C. G. 2002 The loss of sex in clonal plants. Evolutionary Ecology 15: 501-520[CrossRef][ISI][Medline]
Eckert C. G. S. C. H. Barrett 1994 Inbreeding depression in partially self-fertilizing Decodon verticillatus (Lythraceae): population genetic and experimental analyses. Evolution 48: 952-964[CrossRef][ISI]
Eckert C. G. K. Lui K. Bronson P. Corradini A. Bruneau 2003 Population genetic consequences of extreme variation in sexual and clonal reproduction in an aquatic plant. Molecular Ecology 12: 331-344[CrossRef][Medline]
Edwards A. L. R. R. Sharitz 2003 Clonal diversity in two rare perennial plants: Sagittaria isoetiformis and Sagittaria teres (Alismataceae). International Journal of Plant Sciences 164: 181-188[CrossRef]
Ellegren H. 2000 Microsatellite mutations in the germline: implications for evolutionary inference. Trends in Genetics 16: 551-558[CrossRef][ISI][Medline]
Ellegren H. C. R. Primmer B. C. Sheldon 1995 Microsatellite evolution: directionality or bias in locus selection. Nature Genetics 11: 360-362[CrossRef][ISI][Medline]
Ellstrand N. C. M. L. Roose 1987 Patterns of genotypic diversity in clonal plant species. American Journal of Botany 74: 123-131[CrossRef][ISI]
Ericksson O. 1997 Clonal life histories and the evolution of seed recruitment. In H. de Kroon and J. van Groenendael [eds.], The ecology and evolution of clonal plants, 211–226. Buckhuys, Leiden, Netherlands
Excoffier L. P. E. Smouse J. M. Quattro 1992 Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479-491[Abstract]
Fager E. W. 1972 Diversity: a sampling study. American Naturalist 106: 293-310[CrossRef][ISI]
Fassett N. C. B. Calhoun 1952 Introgression between Typha latifolia and T. angustifolia. Evolution 6: 367-379
Felsenstein J. 1993 PHYLIP, version 3.5. University of Washington, Seattle, Washington, USA
Garza J. C. E. G. Williamson 2001 Detection of reduction in population size using data from microsatellite loci. Molecular Ecology 10: 305-318[CrossRef][Medline]
Godt M. J. W. J. L. Hamrick 1999 Population genetic analysis of Elliottia racemosa (Ericaceae), a rare Georgia shrub. Molecular Ecology 8: 75-82
Goodman S. J. 1997 R-ST Calc: a collection of computer programs for calculating estimates of genetic differentiation from microsatellite data and determining their significance. Molecular Ecology 6: 881-885[CrossRef]
Goudet J. 1995 FSTAT, version1.2. A computer program to calculate F-statistics. Journal of Heredity 86: 485-486
Grace B. J. 1985 Juvenile vs. adult competitive abilities in plants: size-dependence in cattails (Typha). Ecology 66: 1630-1638[CrossRef][ISI]
Grace B. J. R. G. Wetzel 1981 Habitat partitioning and competitive displacement in cattails (Typha): experimental field studies. American Naturalist 118: 463-474[CrossRef][ISI]
Hamrick J. L. M. J. W. Godt 1989 Allozyme diversity in plant species. In A. H. D. Brown, M. T. Clegg, A. L. Kahler and B. S. Weir [eds.], Plant population genetics, breeding, and genetic resources, 43–63. Sinauer, Sunderland, Massachusetts, USA
Hangelbroek H. H. N. J. Ouborg L. Santamaria K. Schwenk 2002 Clonal diversity and structure within a population of the pondweed Potamogeton pectinatus foraged by Bewick's swans. Molecular Ecology 11: 2137-2150[CrossRef][Medline]
Harley E. H. 2001 AGARst. A programme for calculating allele frequencies, GST and RST from microsatellite data, version 2. University of Cape Town, Cape Town, South Africa
Hartl D. L. 2000 A primer of population genetics. Sinauer, Sunderland, Massachusetts, USA
Hedrick P. W. 1999 Perspective: highly variable loci and their interpretation in evolution and conservation. Evolution 53: 313-318[CrossRef][ISI]
Jarne P. D. Charlesworth 1993 The evolution of the selfing rate in functionally hermaphrodite plants and animals. Annual Review of Ecological Systematics 24: 441-466[CrossRef][ISI]
Jeffreys A. J. N. J. Royle V. Wilson Z. Wong 1988 Spontaneous mutation rates to new length alleles at tandem-repetitive hypervariable loci in human DNA. Nature 332: 278-281[CrossRef][Medline]
Keane B. S. Pelican G. P. Toth M. K. Smith S. H. Rogstad 1999 Genetic diversity of Typha latifolia (Typhaceae) and the impact of pollutants examined with tandem-repetitive DNA probes. American Journal of Botany 86: 1226-1238
Kovalchuk O. Y. E. Dubrova A. Arkhipov B. Hohn I. Kovalchuk 2000 Wheat mutation rate after Chernobyl. Nature 407: 583[CrossRef][Medline]
Krattinger K. 1975 Genetic mobility in Typha. Aquatic Botany 1: 57-70
Lee D. W. 1975 Population variation and introgression in North American Typha. Taxon 24: 633-641[CrossRef]
Loveless M. D. J. L. Hamrick 1984 Ecological determinants of genetic structure of plant populations. Annual Review of Ecology and Systematics 15: 65-95
Mantel N. A. 1967 The detection of disease clustering and a generalized regression approach. Cancer Research 27: 209-220[ISI][Medline]
Marcinko Kuehn M. J. E. Minor B. N. White 1999 An examination of hybridization between the cattail species Typha latifolia and Typha angustifolia using random amplified polymorphic DNA and chloroplast DNA markers. Molecular Ecology 8: 1981-1990[CrossRef][Medline]
Mashburn S. J. R. R. Sharitz M. H. Smith 1978 Genetic variation among Typha populations of the southeastern United States. Evolution 32: 681-685[CrossRef][ISI]
McNaughton S. J. 1966 Ecotype function in the Typha community-type. Ecological Monographs 36: 297-325[CrossRef]
McNaughton S. J. 1967 Photosynthetic system II: racial differentiation in Typha latifolia. Science 156: 1363
Nauta M. J. F. J. Wessing 1996 Constraints on allele size at microsatellite loci: implications for genetic differentiation. Genetics 143: 1021-1032[Abstract]
Oleksyk T. K. 2001 Distribution and effects of radioactive contamination in rodent populations from Chornobyl. Ph.D. dissertation, The University of Georgia, Athens, Georgia, USA
Pielou E. C. 1969 An introduction to mathematical ecology. Wiley-Interscience, New York, New York, USA
Raymond M. F. Rousset 1995 GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal of Heredity 86: 248-249
Reusch T. B. W. T. Stam J. L. Olsen 2000 A microsatellite-based estimation of clonal diversity and population subdivision in Zostera marina, a marine flowering plant. Molecular Ecology 9: 127-140[CrossRef][Medline]
Rice W. R. 1988 Analyzing tables of statistical tests. Evolution 43: 223-225
Richards A. J. 1986 Plant breeding systems. George Allen & Unwin, London, UK
SAS. 2000 SAS/STAT user's guide, version 8.1. SAS Institute, Cary, North Carolina, USA
Schneider S. D. Roessli L. Excoffier 2000 Arlequin, version 2.000: a software population genetics data analysis. Genetics and Biometry Laboratory, University of Geneva, Geneva, Switzerland
Schoen D. J. A. H. D. Brown 1991 Intraspecific variation in population gene diversity and effective population size correlates with the mating system in plants. Proceedings of the National Academy of Sciences, USA 88: 4494-4497
Sculthorpe C. D. 1967 The biology of aquatic vascular plants. Edward Arnold, London, UK
Sharitz R. R. S. A. Wineriter M. H. Smith E. H. Liu 1980 Comparison of isozymes among Typha species in the eastern United States. American Journal of Botany 67: 1297-1303[CrossRef][ISI]
Sokal R. R. F. J. Rohlf 1995 Biometry. Freeman, New York, New York, USA
Stenström A. B. O. Jonsson I. S. Jónsdóttir T. Fagerström M. Augner 2001 Genetic variation and clonal diversity in four clonal sedges (Carex) along the Arctic coast of Eurasia. Molecular Ecology 10: 497-513[CrossRef][Medline]
Stuefer J. F. B. Erschamber H. Huber J.-I. Suzuki 2002 The ecology and evolutionary biology of clonal plants: an introduction to the proceedings of Clone-2000. Evolutionary Ecology 15: 223-230
Suda J. R. R. R. Sharitz D. O. Straney 1977 Morphological aberrations in Typha populations in post-thermal aquatic habitat. American Journal of Botany 64: 570-575[CrossRef][ISI]
Travis S. E. C. E. Proffitt K. Ritland 2004 Population structure and inbreeding vary with successional stage in created Spartina alterniflora marshes. Ecological Applications 14: 1189-1202[CrossRef][ISI]
Tsyusko O. V. 2004 Radiation and genetics of cattail populations from Chornobyl. Ph.D. dissertation, The University of Georgia, Athens, Georgia, USA
Tsyusko-Omeltchenko O. V. N. A. Schable M. H. Smith T. C. Glenn 2003 Microsatellite loci isolated from narrow-leaved cattail Typha angustifolia. Molecular Ecology Notes 3: 535-538[CrossRef][ISI]
Viard F. F. Justy P. Jarne 1997 The influence of self-fertilization and population dynamics on the genetic structure of subdivided populations: a case study using microsatellite markers in the freshwater snail Bulinus truncatus. Evolution 51: 1518-1528[CrossRef][ISI]
Widen B. N. Cronberg M. Widen 1994 Genotypic diversity, molecular markers and spatial distribution of genets in clonal plants. In C. M. Soukupova, C. Marshall, T. Hara, and T. Herben [eds.], Plant clonality, biology, and diversity, 139–157. Opulus Press, Uppsala, Sweden
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
