Bot. Bull. Acad. Sin. (2003) 44: 113-122

Park et al. — A test of hybrid origin of Paraixeris koidzumiana

Allozyme variation in Paraixeris: a test for the diploid hybrid origin of Paraixeris koidzumiana (Compositae)

Ki-Ryong Park^1,*, Jae-Hong Pak², and Bong-Bo Seo²

¹Department of Biology, Kyung-Nam University, 449 Wolyoung-dong, Masan 631-701, Korea

²Department of Biology, College of Natural Sciences, Kyungpook National University, Daegu 702-701, Korea

(Received August 7, 2002; Accepted November 18, 2002)

Abstract. Variation in isozyme patterns of four Korean Paraixeris species was used to test the hypothesis that P. koidzumiana is a hybrid derivative of P. chelidoniifolia and P. denticulata, or alternatively, of the progenitor P. chelidoniifolia-P. sonchifolia pair. Eleven loci from eight enzymes were examined for 29 populations of P. koidzumiana, P. denticulata, P. chelidoniifolia, and P. sonchifolia. Although recent RAPD and morphological data supported the hybrid origin of P. koidzumiana, isozyme evidence did not. No evidence of additivity was found in P. koidzumiana at the loci that significantly differentiate P. sonchifolia from other species. In addition, P. koidzumiana populations did not combine the marker alleles differentiating P. denticulata from P. chelidoniifolia. Recent divergence between P. koidzumiana and P. chelidoniifolia is strongly supported by the high genetic identity value and sharing of the same high frequency alleles at most loci. Our isozyme data largely support a conclusion that the unidentified individuals found in species sympatry are hybrid derivatives.

Keywords: Hybridization; Isozyme; Paraixeris koidzumiana.

Introduction

Natural hybridization is common in flowering plants and provides an important source of increasing genetic variation, creating novel lineages via stabilization of the hybrid derivatives and breakdown or reinforcement of isolating barriers (Levin, 1979; Rieseberg and Ellstrand, 1993; Wolfe and Elisens, 1993; Arnold, 1992, 1997; Morrell and Rieseberg, 1998; Nielsen, 2000). However, due to sterility or limited fertility in hybrid individuals, and the lack of molecular markers to detect parental taxa, diploid hybridization has been assigned only minor evolutionary importance (Arnold, 1992; Wolfe et al., 1998). Several recent studies of plant speciation suggest that new diploid species might arise rapidly through hybridization between genetically divergent species (Ungerer et al., 1998). In this case, genetic additivity of allozyme data has been useful in determining parental species (Gallez and Gottlieb, 1982). However, the detection of ancient hybrid origin may be difficult due to the parental taxa and its derivatives having had a chance to accumulate new genetic variation (Morrell and Rieseberg, 1998). Although many cases of recombinational speciation in the Far East Asian Paraixeris group have been proposed based on cytological and morphological studies (Ono, 1946, 1950, 1951), few have been tested with molecular markers.

The Korean endemic Paraixeris koidzumiana Kitam. (Kitamura, 1942; Pak, 1991) is a diploid (2n = 10) herbaceous biennial, restricted to Mt. Chiri, while the closely

related P. chelidoniifolia, P. denticulata, and P. sonchifolia are also diploid, but widely distributed in Korea, Japan, and Manchuria (Pak and Kawano, 1990). Since Kitamura (1942) described P. koidzumiana from the specimen of Koidzumi collected from Mt. Chiri, Lee (1979) proposed that P. koidzumiana originated from hybridization between P. sonchifolia and P. chelidoniifolia. This hybrid origin was originally proposed because of the intermediate leaf morphology between the two species. Paraixeris koidzumiana resembles P. chelidoniifolia in having pinnately divided leaves, but is similar to P. sonchifolia in having margined petioles that clasp at the base. However, a cytological study by Pak (1991) did not give support to P. sonchifolia being one of the parental species of P. koidzumiana. Additionally, the difference between the flowering periods of P. sonchifolia and P. chelidoniifolia argues against their hybridization (Tae et al., 2001). A recent molecular study using RAPD data supports the hybrid origin of P. koidzumiana from P. denticulata (Houtt.) Nakai and P. chelidoniifolia (Tae et al., 2001). In addition, at Mt. Chiri, where the above three species grow sympatrically, plants were discovered with leaves and inner-involucre numbers that appeared to be intermediate between P. chelidoniifolia and P. denticulata. Canonical variate analysis on the morphological characters suggested that the individuals were hybrids derived from P. chelidoniifolia and P. denticulata (Pak, unpublished data). The objective of the study was to determine, using isozyme analysis, whether P. koidzumiana and putative hybrid populations were derived through hybridization among P. chelidoniifolia, P. denticulata, and P. sonchifolia.

*Corresponding author. Tel: 82-55-249-2240; E-mail: park@kyungnam.ac.kr

Botanical Bulletin of Academia Sinica, Vol. 44, 2003

Materials and Methods

A total of 361 individuals from 29 populations, representing four Paraixeris species and four hybrid populations previously used in Canonical variate analysis by Pak (unpublished data), were examined for isozyme variation (Table 1). Voucher specimens were deposited at the Herbarium of Kyung-Nam University (KNUH). Soluble enzymes were isolated from the fresh leaf tissue of field-collected plants, ground in an extracting buffer containing 0.1 M tris-HCl, pH 7.5, 1 mM EDTA (tetrasodium salt), 10 mM MgCl₂, 10 mM KCl, 14 mM 2-mercaptoethanol, and 5-10 mg/ml solid polyvinylpolypyrrolidone (Gottlieb, 1981). Leaf extracts were centrifuged in 1.5 ml tubes and the supernatant absorbed onto wicks of Whatman 17 MM chromatography paper.

Eight enzymes were resolved on 12.5% starch gels utilizing two buffer systems. System I had an electrode buffer of 0.065 M L-histidine and 0.007 M citric acid, adjusted to pH 6.5 with NaOH. This gel buffer was a 1:3 dilution of the electrode buffer. System II consisted of an electrode buffer with 0.18 M tris, 0.1 M boric acid, and 0.004 M EDTA, pH 8.6. System I was used to resolve malate dehydrogenase (MDH), 6-phosphogluconate dehydrogenase (6PGD), glyceraldehyde 3-phosphate dehydrogenase (GA3PD), malic enzyme (ME), and phosphoglucomutase (PGM). System II was used to resolve the enzyme systems alcohol dehydrogenase (ADH), phosphoglucose isomerase (PGI), and triosephosphate isomerase (TPI). Enzyme-activity staining and agarose overlays generally followed the protocols of Soltis et al. (1983). Isozymes and allozymes were numbered sequentially and lettered alphabetically, beginning with the most anodal form. The BIOSYS-1 program (Swofford and Selander, 1981) was used to calculate mean number of alleles per locus (A), percentage of polymorphic loci (P), mean observed heterozygosity (H_o), and mean expected heterozygosity (H_e) within the populations studied. For the analysis of population dif

ferentiation Wright's (1965) F statistics were calculated. The F statistics include F_IS, an index of inbreeding, F_IT, the overall inbreeding coefficient, and F_ST, a measure of the genetic differentiation over subpopulations (Wright, 1965). A UPGMA phenogram was produced by input of Nei's (1978) genetic identity values into the BIOSYS-1 program. We investigated genetic variation among populations using principle components analysis (PCA) of allelic data matrix, carried out using NYSYS-pc (Rohlf, 1992).

Results

Eleven loci, coding for eight enzymes, were scored for 29 populations of four species of Paraixeris (Tables 1, 2). The number of isozymes detected in Korean Paraixeris groups (2n = 10) was similar to that reported for other diploid species (Weeden and Wendel, 1989).

Allele frequencies for 11 polymorphic loci were summarized for four species of Paraixeris and their hybrid populations (Table 2). The distribution of the highest-frequency alleles was consistent among four species in ADH-1 and GA3PD-1. Paraixeris sonchifolia had five high frequency or fixed alleles (GA3PD-2^d , ME-1^d, PGI-1^a, PGM-1^c, and TPI-1^c) which were either not found or had a low frequency in P. koidzumiana, P. chelidoniifolia, and P. denticulata. Paraixeris koidzumiana and P. chelidoniifolia had the same allele in highest frequency at most of the loci. ADH-2^c was the highest frequency allele in P. chelidoniifolia, but was absent from P. denticulata and was a low frequency allele in P. sonchifolia. Paraixeris denticulata and P. sonchifolia shared a diagnostic allele ADH-2^b, which is a low frequency allele found only in a population of P. chelidoniifolia.

The alleles observed in P. koidzumiana were not a combination of those observed in P. sonchifolia and P. chelidoniifolia. None of the five unique to high frequency alleles to P. sonchifolia were found in P. koidzumiana

Park et al. — A test of hybrid origin of Paraixeris koidzumiana

Botanical Bulletin of Academia Sinica, Vol. 44, 2003

Park et al. — A test of hybrid origin of Paraixeris koidzumiana

whereas most of the high frequency alleles in P. koidzumiana were found in P. chelidoniifolia. The diagnostic allele, ADH-2^b in P. denticulata and P. sonchifolia, was found in only a few individuals of a population of P. koidzumiana (Table 2). The low frequency, but unique alleles of P. denticulata, GA3PD-2^a, PGI-1^a,c and TPI-2^c,d were not found in P. koidzumiana.

Genetic identity values among populations within species were 0.933 for P. denticulata, 0.915 for P. koidzumiana, and 0.866 for P. chelidoniifolia (Table 3). Nei's (1978) genetic identities between species show P. koidzumiana to be more genetically similar to P. chelidoniifolia (0.882) than to P. denticulata (0.764) or P. sonchifolia (0.561).

Botanical Bulletin of Academia Sinica, Vol. 44, 2003

Measures of genetic variation within and among populations of four species and four hybrid populations are presented in Table 4. Among the four species, P. denticulata and P. sonchifolia showed higher values of A (1.80-1.90), P (62.91-63.60), H_o (0.251-0.428) and H_e (0.242-0.314) than P. koidzumiana or P. chelidoniifolia (Table 4). Most of the putative hybrid populations represented levels of isozyme polymorphism greater than those found in their parental species (Table 4). Genetic variation within and between populations was measured using F-statistics (Table 5). A UPGMA phenogram based on genetic identity values depicts the close genetic relationship among populations of P. chelidoniifolia and P. koidzumiana whereas the population of P. sonchifolia was clearly separated from the remaining species (Figure 1). Three putative hybrid populations (HYB26, HYB27 and HYB28) form a cluster with P. denticulata populations whereas hybrid population 29 is nested in the P. koidzumiana and P. chelidoniifolia populations. Principal components analyses (PCA) based on genetic data showed the populations of P. koidzumiana to be most similar to P. chelidoniifolia and P. sonchifolia to be distinct from all other species (Figure 2). The putative hybrid populations, HYB26 and HYB27 were most similar or overlapping to the populations of P. denticulata.

Discussion

Hybrid Origin of P. koidzumiana

Five species of Paraixeris have been reported from far eastern Asia (Kitamura, 1955) and, aside from P. yoshinoi, four of these occur in Korea. Paraixeris koidzumiana, an endemic species to Korea, is found in sunny and moist habitats, mainly at low elevations on Jiri Mountain. It is sympatric with P. chelidoniifolia and P. denticulata (Tae et al., 2001). Based on the leaf, bract, and flower morphology, P. sonchifolia and P. chelidoniifolia were proposed as the putative parents of P. koidzumiana (Lee, 1979). Recent cytological data strongly suggested the exclusion of P. sonchifolia as a putative parent of P. koidzumiana based on a unique satellite at the chromosome of P. sonchifolia (Pak and Kawano, 1990; Pak, 1991). Additionally, flowering periods strongly support the exclusion of P. sonchifolia as a parental species. Paraixeris sonchifolia blooms in late spring to early summer, but the

remaining species bloom between September and November.

Molecular analysis using RAPD markers provided support for the hybrid origin of P. koidzumiana (Tae et al., 2001). The specific RAPD markers from P. chelidoniifolia and P. denticulata also occur in P. koidzumiana. Considering the floret numbers, P. koidzumiana lies between P. chelidoniifolia and P. denticulata, suggesting a hybrid origin (Tae et al., 2001).

Although the RAPD and morphological data suggested a hybrid origin for P. koidzumiana, our isozyme evidence did not support this hypothesis. The alleles observed in P. koidzumiana were not a combination of those observed in P. sonchifolia and P. chelidoniifolia or those in P. chelidoniifolia and P. denticulata. No evidence of additivity was found in P. koidzumiana at five isozyme loci that significantly differentiate P. sonchifolia from the remaining species. With one exception, the populations of P. koidzumiana did not combine the marker alleles that differentiate P. denticulata from P. chelidoniifolia. Although isozymes and RADPs are equally useful in identi

Park et al. — A test of hybrid origin of Paraixeris koidzumiana

fying parental species of putative hybrids by a comparison of electrophoretic profiles of the hybrid and its parents, the invalid assessment of band homology has been considered one of the major problems with RAPDs in systematic studies (Dowling et al., 1996; Edwards, 1998; Såstad et al., 1999).

Comparing within species identity values (I = 0.915 - 0.866), the high genetic identity value, 0.882, between P. koidzumiana and P. chelidoniifolia indicates that either relatively little genetic differentiation has occurred between them or that P. koidzumiana is a recent derivative of P. chelidoniifolia (Rieseberg et al., 1990; Soltis and Bloom, 1991; Wolfe and Elisens, 1993). Paraixeris koidzumiana is restricted to a small geographic area covering the national park at Jiri Mountain while P. chelidoniifolia is relatively widespread and commonly occurs in wet and open mountain areas and roadsides. The UPGMA and PCA

Figure 2. Three-dimensional model of numbered populations (Table 1) derived from principal components analysis of gene frequency data in Korean Paraixeris.

results are consistent with a recent divergence in the above two species. The phenogram illustrated a close genetic relationship between P. chelidoniifolia and P. koidzumiana and that the populations of P. koidzumiana and P. chelidoniifolia are intermixed (Figure 1). These data suggest that P. koidzumiana and P. chelidoniifolia may be a recent divergent species pair rather than a stabilized hybrid derivative as proposed by Lee (1979) and Tae et al. (2001). However, the hypothesis of progenitor-derivative species pair seems to be unreliable because a significant number of low frequency alleles occurring in P. koidzumiana were not found in P. chelidoniifolia.

Hybridization among Species-Pairs in Korean Paraixeris

Previous canonical variate analysis using morphological data, pollen stain-ability analysis, and pollen morphology (Pak, unpublished data) strongly suggest that individuals from populations 26, 27 and 28 found at road-sides at Bykso-ryung, were most likely hybrids of P .denticulata and P. chelidoniifolia. A low percentage of pollen stain ability (26.2%) in aniline blue and intermediate pollen grain size between the two parental species indicate that these individuals are hybrids. Our isozyme results revealed that the three populations shared a high genetic identity with P. denticulata. Population 27 had the most marker alleles of P. denticulata, but shared a marker allele (ADH-2^c = 0.2 ) with P. chelidoniifolia, indicating that this population may also have experienced past introgression in the direction from P. chelidoniifolia to P. denticulata. The most common alleles from the presumed hybrid population 29 were shared by both P. koidzumiana and P. chelidoniifolia. In addition, the Gagi Mountain population showed high genetic identities with these two species. The cluster analysis using electrophoretic results

Figure 1. UPGMA phenogram derived from Nei's genetic identity value among 29 populations representing four Korean Paraixeris species and four hybrid populations. Species abbreviation and population numbers refer to Table 1.

Botanical Bulletin of Academia Sinica, Vol. 44, 2003

showed that the population of Gagi Mountain actually clustered most closely with the populations of P. koidzumiana and P. chelidoniifolia.

Hybrid populations usually display levels of isozyme polymorphism greater than their putative parents (Morrell and Rieseberg, 1998). The proposed hybrid populations, 27 and 29, revealed greater values of genetic variation compared with the putative parental species (Table 4). Thus, hybridization among Korean Paraixeris species seems to be commonplace where the species are sympatric.

Genetic Variation in Korean Paraixeris

The mean values of A and P for four species of Korean Paraixeris were similar to those calculated by Hamrick and Godt (1989) for animal pollinated out-crossing species with annual habit. Comparing the genetic variation data from the Bonin Islands' Crepidiastrum species (Ito and Ono, 1990; P = 22.34, A = 1.24, H_e=0.056), the most closely related species to Paraixeris, Korean Paraixeris species show significantly higher values of genetic variation. This may be due to the relatively widespread distribution of Paraixeris species across Korea, Japan, and Manchuria. However, the Crepidiastrum species of the Bonin Islands should have experienced a bottleneck effect due to the small population size and restricted habitats (Ito and Ono, 1990).

The widespread species P. denticulata and P. sonchifolia have a higher genetic variation than the restricted, endemic species P. koidzumiana. These data are consistent with the previous hypothesis that the geographic range had a significant effect on genetic variation at species and population level (Frankel et al., 1995). A geographically restricted species such as P. koidzumiana may have a chance to experience high levels of inbreeding, genetic drift, and strong directional selection toward genetic uniformity (Karron, 1991).

The relatively low levels of genetic variation and the high genetic differentiation among populations (F_ST=0.402) of P. chelidoniifolia are exceptional. Unknown recent history of the populations and some of the attributes presented by P. chelidoniifolia, such as extensive clonal growth and lack of efficient seed dispersal mechanisms, should result in low genetic variation and a high genetic divergence among populations.

Acknowledgments. This work was supported by Korean Research Foundation Grant (KRF-99-015-DI0083).

Literature Cited

Arnold, M.L. 1992. Natural hybridization as an evolutionary process. Ann. Rev. Ecol. Syst. 23: 237-261.

Arnold, M.L. 1997. Natural Hybridization and Evolution. Oxford University Press, New York.

Dowling, T.E., C. Moritz, J.D. Palmer, and L.H. Rieseberg. 1996. Nucleic acids III: Analysis of fragments and restriction sites. In D.M. Hillis, C. Moritz and B.K. Mable (eds.), Molecu

lar Systematics, Second Edition, Sinauer Associates, Sunderland, Mass., pp. 407-514.

Edwards, K.J. 1998. Randomly amplified polymorphic DNAs (RAPDs). In A. Karp, P.G. Isaac and D.S. Ingram (eds.), Molecular Tools for Screening Biodiversity, Kluwer Academic Publishers, Dordrecht, pp. 171-179.

Frankel, O.H., A.D.H. Brown, and J.J. Burdon. 1995. The Conservation of Plant Biodiversity. Cambridge University Press, Cambridge.

Gallez, G.P. and L.D. Gottlieb. 1982. Genetic evidence for the hybrid origin of the diploid plant Stephanomeria diegensis. Evolution 36: 1158-1167.

Gottlieb, L.D. 1981. Electrophoretic evidence and plant populations. Prog. Phytochem. 7: 1-46.

Hamrick, J.L. and M.J.W. Godt. 1989. Allozyme diversity in plant species. In A.D.H. Brown, M.T. Clegg, A.L. Kahler and B.S. Weir (eds.), Plant Population Genetics, Breeding, and Genetic Resources, Sinauer, Sunderland, Mass., pp. 43-63.

Ito, M. and M. Ono. 1990. Allozyme diversity and the evolution of Crepidiastrum (Compositae) on the Bonin (Ogasawara) Islands. Bot. Mag. Tokyo 103: 449-459.

Karron, J.D. 1991. Patterns of genetic variation and breeding systems in rare plant species. In D.A. Falk and K.E. Holsinger (eds.), Genetics and Conservation of Rare Plants, Oxford University Press, Oxford.

Kitamura, S. 1942. Expositiones plantarum novarum orientali-Asiaticarum VII. Acta Phytotax. Geobot. 11: 120-133.

Kitamura, S. 1955. Compositae Japonicae. Mem. Coll. Sci. Kyoto Univ. Ser. B. 22: 77-126.

Lee, T.B. 1979. Illustrated flora of Korea. Hyangmoonsa, Seoul.

Levin, D.A. 1979. Part I: the role of hybridization in evolution, editor's comments. In D.E. Levin (ed.), Hybridization an evolutionary perspective, Dowden, Hutchinson & Ross, Inc., Stroudsburg, Pennsylvania.

Morrell, P.L. and L.H. Rieseberg. 1998. Molecular tests of the proposed diploid hybrid origin of Gilia achilleifolia (Polemoniaceae). Amer. J. Bot. 85: 1439-1453.

Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583-590.

Nielsen, L.R. 2000. Natural hybridization between Vanilla claviculata (W. Wright) Sw. and V. barbellata Rchb.f. (Orchidaceae): genetic, morphological, and pollination experimental data. Bot. J. Linn. Soc. 133: 285-302.

Ono, H. 1946. Intergenetic hybridization in Cichorieae, VII. Cytology of F₄ individuals of Paraixeris denticulata x Crepidiastrum platyphyllum. Cytologia 14: 158-173.

Ono, H. 1950. Intergeneric hybridization in Cichorieae, VIII. Morphological and fertility of F₄ individuals of Paraixeris deticulata x Crepidiastrum platyphyllum. Jap. J. Genet. 25: 167-176.

Ono, H. 1951. Intergeneric hybridization in Cichorieae, IX. Considerations on cytological and morphological features of F₄ individuals of Paraixeris denticulata x Crepidiastrum platyphyllum with special reference to fertility. Jap. J. Genet. 26: 156-173.

Pak, J.-H. 1991. Karyomorphology of Youngia koidzumiana (Compositae; Lactuceae). Kor. J. Plant Tax. 21: 117-121.

Pak, J.-H. and S. Kawano. 1990. Biosystematic studies on the

Park et al. — A test of hybrid origin of Paraixeris koidzumiana

genus Ixeris and its allied genera (Compositae-Lactuceae). III. Fruit wall anatomy and karyology of Crepidiastrum and Paraixeris, and their taxonomic implications. Acta Phytotax. Geobot. 41: 109-128.

Rieseberg, L.H., R. Carter, and S. Zona. 1990. Molecular tests of the hypothesized hybrid origin of two diploid Helianthus species (Asteraceae). Evolution 44: 1498-1511.

Rieseberg, L.H. and N.C. Ellstrand. 1993. What can molecular and morphological markers tell us about plant hybridization? Crit. Rev. Plant Sci. 12: 213-241.

Rohlf, F.J. 1992. NTSYS-pc: numerical taxonomy and multivariate analysis system (version 1.70). Exeter Software, New York.

Såstad, S.M., H.K. Stenøien, and K.I. Flatberg. 1999. Species delimitation and relationships of the Sphagnum recurvum complex (Bryophyta)-as revealed by isozyme and RAPD markers. Syst. Bot. 24: 95-107.

Soltis, P.S. and W.L. Bloom. 1991. Allozyme differentiation between Clakia nitens and C. speciosa (Onagraceae). Syst. Bot. 16: 399-406.

Soltis, D.E., C.H. Haufler, D.C. Darrow, and G.J. Gastony. 1983. Starch gel electrophoresis of ferns: a complication of glinding buffers, gel and electrode buffers, and staining schedules. Am. Fern J. 73: 10-27.

Swofford, D.L. and R.B. Selander. 1981. BIOSYS-1: a FORTRAN program for the comprehensive analysis of electro

phoretic data in population genetics and systematics. J. Hered. 72: 281-283.

Tae, K., D.S. In, and K. Shin. 2001. Assessment of random amplified polymorphic DNA (RAPD) for determining the origin of Youngia koidzumiana (Compositae). J. Plant Biol. 44: 61-64.

Ungerer, M.C., S.J.E. Baird, J. Pan, and L.H. Rieseberg. 1998. Rapid hybrid speciation in wild sunflowers. Proc. Natl. Acad. Sci. USA 95: 11757-11762.

Weeden, J.F. and N.F. Wendel. 1989. Genetics of plant isozymes. In D.E. Soltis and P.S. Soltis (eds.), Isozymes in Plant Biology, Portland, Dioscorides Press, pp. 46-72.

Whang, S.S., K. Choi, R.S. Hill, and J.H. Pak. 2002. A morphometric analysis of infraspecific taxa within the Ixeris chinensis complex (Asteraceae, Lactuceae). Bot. Bull. Acad. Sin. 43: 131-138.

Wolfe, A.D. and W.J. Elisens. 1993. Diploid hybrid speciation in Penstemon (Scrophulariaceae) revisited. Amer. J. Bot. 80: 1082-1094.

Wolfe, A.D., Q. Xiang, and S.R. Kephart. 1998. Diploid hybrid speciation in Penstemon (Scrophulariaceae). Proc. Natl. Acad. Sci. USA 95: 5112-5115.

Wright, S. 1965. The interpretation of population structure by F-statistics with special regard to systems of mating. Evolution 19: 395-420.

Botanical Bulletin of Academia Sinica, Vol. 44, 2003