行政院國家科學委員會專題研究計畫 成果報告
台灣柯及柳葉柯(殼斗科) 之親緣地理及保育研究(3/3)
計畫類別： 個別型計畫
計畫編號： NSC92-2313-B-006-002-
執行期間： 92 年 08 月 01 日至 93 年 07 月 31 日 執行單位： 國立成功大學生物學系（所）
計畫主持人： 蔣鎮宇
報告類型： 完整報告
處理方式： 本計畫可公開查詢
中 華 民 國 93 年 11 月 2 日
Abstract:
Gene genealogy of the cpDNA atpB-rbcL noncoding spacer was reconstructed to assess the phylogeographic pattern of two closely related oaks, Lithocarpus formosanus (Skan) Hayata and L. dodonaeifolius (Hayata) Hayata (Fagaceae). High levels of nucleotide and haplotype diversities but low levels of genetic differentiation between species and among populations were detected. The result is consistent with the paraphyly of this cpDNA spacer in both species suggested by a
neighbor-joining analysis. On the contrary, RAPD fingerprints revealed limited ongoing gene flow and significant genetic
differentiation between species and among populations. Given low possibilities that seeds disperse across a long geographic range in the modern vegetation, high Nm values, estimates of number of migrants per generation deduced from the seed-carried organelle DNA marker are likely to represent historical migrations. A migrant-pool model explains the heterogeneous composition of the organelle DNA within populations and the low differentiation among populations. Ancestral polymorphic alleles, however, prolonged the lineage sorting period within populations and species. Given a relatively short evolutionary duration since isolation, high genetic heterogeneity has made the attainment of coalescence improbable.
本研究利用葉綠體atpB-rbcL noncoding spacer來重建兩近緣種台灣柯以及柳葉柯之 親緣關係，結果顯示該片段具有高度遺傳歧 異度，而且族群間具有低有低度遺傳分化，
然而從所重建之親緣樹狀圖顯示兩物種呈現
paraphyly親緣關係，但是利用RAPD分子指 紋技術則顯示族群間具有高度遺傳分化以及 有限基因交流。
Key words: Fagaceae, lineage sorting, Lithocarpus, phylogeography, Taiwan.
Introduction:
To understand phylogeographic patterns and the gene genealogy of species of
continental islands, oaks provide ideal material.
A great number of data are available for species of continents of Europe and North America (Whittemore & Schaal, 1991; Manos et al., 1999), and the fagaceous plants share similar breeding systems and demography. In Taiwan, Lithocarpus species are highly diverged, with 14 species recognized (Yang et al., 1997). Like many endemic species that survived glacial cycles on the island, a large number of relictual Lithocarpus species, such as L. castanopsisifolius (Hayata) Hayata and L.
konishii (Hayata) Hayata, are sporadically distributed with limited population sizes, partly due to biogeographic history and recent human disturbance (Lin, 1966). The taxonomy of Taiwan’s oaks has been well documented (Liao, 1996; Huang et al., 1999). Some species
complexes have long puzzled taxonomists, including Lithocarpus formosanus (Hayata) Hayata and L. dodonaeifolius (Hayata) Hayata.
Kudoˆ (1931: 387) recognized the latter within L. formosanus (as Synaedrys [: Lithocarpus]
formosana fo. dododaeifolia), and Li (1953) also suggested a conspecific relationship. In contrast, recent taxonomic treatments (Liao, 1996; Yang et al., 1997) recognized two
separate species.
Trees of 4–9 m in height from both taxa in Lithocarpus share entire leaf margins and a rounded leaf apex. The oblanceolate leaf shape of Lithocarpus dodonaeifolius is distinct from the elliptic leaf shape of L. formosanus. In addition, shorter petioles (4–8 mm), longer infructescence (3–5 cm), and tawny tomentose cupule bracts characterize L. dodonaeifolius, while longer petioles (10–13 mm), shorter infructescence (ca. 3 cm), and gray tomentose cupule bracts occur in L. formosanus.
Lithocarpus formosanus and L. dodonaeifolius are allopatrically distributed in southern Taiwan about 30 km apart. A single extant population of L. formosanus, consisting of no more than 100 plants, remains in the wild, although several scattered populations were previously recorded (cf. Lu, 1996). This population is distributed along the Nanjen Stream in the Kengting National Park. Two subpopulations of L. formosanus occurring along ridges about 400 m alt. are separated by the Nanjen stream. In contrast, three
populations of L. dodonaeifolius are distributed along the Central Mountain Range of the Taiwanese island: Mt. Weiliaoshan (with about 100 individuals, ca. 1200 m alt.),
Chingshuiying (with about 200 individuals, ca.
1500 m alt.), and Dazen (with 9 individuals, ca.
600 m alt.). These three populations of L.
dodonaeifolius are isolated by distances between 20 km and 60 km. Ecologically, both species usually grow on wind-facing slopes, mixing with other species of Fagaceae and Lauraceae in tropical or subtropical forests.
In order to determine the level of ongoing gene flow between populations, RAPD
fingerprints, which are mostly amplified from the nuclear genome (Hawkins & Harris, 1998), were utilized to assess the extent of migration.
In the study, several objectives are pursued: (1) the partitioning pattern of cpDNA variation within and between species; (2) the possible migratory mode of Taiwan’s oaks over the geological history; (3) the coalescence process of cpDNA alleles within species and populations; and (4) the level of ongoing gene flow between populations.
Results:
In this study, the atpB-rbcL noncoding spacer of cpDNA in Lithocarpus formosanus and L. dodonaeifolius were PCR amplified and sequenced. Length of the atpB-rbcL spacer of cpDNA varied from 730 bp (isolate doda36) to 935 bp (isolate for4005). A total of 996 bp were aligned of which 201 sites (20.2%) excluding the sites with alignment gaps were polymorphic. This noncoding spacer was A-T rich (68.8%). All sequences were unique.
Nineteen haplotypes and 39 haplotypes of the cpDNA noncoding spacer were determined in L. formosanus and L. dodonaeifolius,
respectively, according to the DnaSP analyses.
Nucleotide diversity (Jukes & Cantor, 1964) of 0.08074 vs. 0.01149 and 0.06026 vs. 0.00702 was estimated in the above species,
respectively. In the largest population (Chingshuiying) of L. dodonaeifolius, nucleotide diversity as measured by pairwise estimates of the cp-DNA was higher than that of those smaller populations.
A neighbor-joining (NJ) tree was reconstructed based on the genetic distance among aligned sequences of the cpDNA. Four groups (A–D) were identified and supported by bootstrap significantly. Clades B and C were grouped further, although not significantly (bootstrap value 5 0.52). The monophyly of the cpDNA noncoding spacer in either L.
formosanus or L. dodonaeifolius was not suggested by the NJ analysis. Nevertheless, clade D consisted of individuals of L.
dodonaeifolius exclusively. Within clade A, most sequences of L. formosanus and L.
dodonaeifolius were mixed, except for a subcluster A1, which comprised sequences of the latter species only. Within the clade B, an unusually long branch leading to the for1502 (L. formosanus) was identified, which differed by 49 mutational changes from its closest sequence of for1501 (L. formosanus).
Discussion:
Lithocarpus formosanus and L.
dodonaeifolius seemed to possess higher levels of cpDNA haplotype diversity (19 and 39 haplotypes, respectively) than other plants, e.g., 13 cpDNA haplotypes in Beta vulgaris subsp.
maritima (Desplanque et al., 2000), 11 haplotypes in Argania (El Mousadik & Petit, 1996), 23 haplotypes in white oaks
(Dumolin-Lape`gue et al., 1997), and 13 haplotypes in Alnus (King & Ferris, 1998). The nucleotide diversity of these two Taiwanese oaks was also high, compared to that of
California pines (Hong et al., 1993). High level of genetic diversity in the Fagaceae (cf. Petit et
al., 1997; Dumolin-Lape`gue et al., 1997) is probably associated with their long
evolutionary history, which allows genetic variation to accumulate within lineages (cf.
Chiang & Schaal, 1999). Nevertheless, the higher haplotype diversity of cpDNA of Lithocarpus from Taiwan may be simply derived from different molecular techniques employed. Nucleotide sequencing usually detects a higher level of genetic variation than do the RFLP and PCR-RFLP techniques. In our analysis, according to the deduced restriction site map, only four chlorotypes (cpDNA polymorphisms) for 58 individual haplotypes could be identified. Within smaller populations of L. dodonaeifolius from
Weiliaoshan and Dazen, the number of major clades was even lower (with only two), while high levels of haplotype diversity and
nucleotide diversity were assessed based on the sequence variation. As stated by Desplanque et al. (2000), a substantial within-population diversity in natural plants should have existed (cf. McCauley, 1994; El Mousadik & Petit, 1996; Raspe, 1998). The detection of the existing variation surely depends not only on the conservative nature of the molecular marker itself, but also the sensitivity of the tools employed. In this study, the socalled
‘‘higher’’ level of genetic variation of L.
formosanus and L. dodonaeifolius turned out much lower than that of other species, when nucleotide sequences were transferred to RFLP data (data not shown). Such low level of chlorotype diversity, as expected, may be ascribed to their smaller population number and size.
According to the shared cpDNA alleles of the atpB-rbcL noncoding spacer, the speciation of Lithocarpus dodonaeifolius and L.
formosanus may be recent. Like other fagaceous plants, migration via long-range seed dispersal (cf. Petit et al., 1997) of the ancestral populations of the Taiwan’s oaks became possible due to the dramatic change of vegetation (cf. Chiang & Hong, 1999) during deglaciation. Many novel niches were then available for plants that survived after species extinction. Seeds from different resource populations may have migrated into the refugia and settled subsequently. Such migration may have increased the heterogeneity of cpDNA composition within populations. Extinction and re-colonization regulated by geological events, however, were thought to enhance genetic differentiation among populations (cf. Wright, 1977) owing to founder events resulting from the colonization by a small number of
surviving individuals. Recently, Wade and McCauley (1990) further suggested that the results of extinction/re-colonization on genetic differentiation among populations depend on the number of founders as well as the level of heterogeneity of genetic composition within populations according to coalescence theory.
When the colonist size is small and the genetic heterogeneity is low, genetic differentiation among populations will be reached fast via stochastic processes alone.
Despite the relatively small population size in both Lithocarpus species, no
coalescence has been achieved within populations. To counter the genetic diversity loss within small populations due to genetic
drift, it is likely that the genetic composition of ancestral populations, even species, prior to deglaciation as well as succeeding colonizing populations was highly heterogeneous. The four chlorotypes noted for these two
Lithocarpus species in Taiwan may have existed long before the speciation event. In addition, recurrent genetic recombination within the chloroplast intergenic spacer may have increased the heterogeneity within species as well. Given small population sizes, the low level of genetic differentiation among
Lithocarpus populations at the cpDNA
noncoding spacer region is possibly ascribed to a short duration (since the last deglaciation) for coalescence.
At the intraspecific level, patchy structure of local populations (e.g., Aquilegia, Strand et al., 1996; Fagaceae, Petit et al., 1997;
Dumolin-Lape`gue et al., 1999) or geographic subdivision between long isolated populations (e.g., northern and southern populations of Liriodendron tulipifera L., Sewell et al., 1996;
Sarmathic–Baltic and Alpine–Central European populations of Picea abies (L.) H.
Karst., Vendramin et al., 2000) have been documented in many tree species of continents.
Low level of organelle DNA differentiation among local populations was detected in Lithocarpus dodonaeifolius and L. formosanus as well as other plants of continental islands, such as Cycas taitungensis Shen et
al. (Huang et al., 2001), Amorphophallus (cf.
Chiang & Peng, 1998), Michelia formosana (Kanehira) Masamune (Lu et al., 2002) of Taiwan as well as Japanese Abies (Tsumura &
Suyama, 1998). Due to the limited area and
available habitats of the island, the population size of Taiwan’s oaks is effectively smaller than that of continental species. Under near neutrality, a long period of lineage sorting for cpDNA chlorotypes in small populations may be likely ascribed to high levels of heterogeneity in genetic composition.
Interestingly, greater genetic variation has been noted in southern continental refugia (such as Italy, the Balkans, and the Iberian Peninsula) than in northern populations (e.g., Central Europe) (European oaks, cf. Dumolin- Lape`gue et al., 1997, 1998; European Alnus, King & Ferris, 1998; Fagus, Demesure et al., 1996; and Japanese Abies, Tsumura & Suyama, 1998). The continent-island discrepancy may be associated with the fact that Taiwan, which is straddled across today’s subtropics to tropics, is much more south than most areas of
European and American continents
geographically. Taiwan may have provided more fitting habitats for surviving plants during the glacial maximum than mainland refugia.
Reference:
Avise, J. C. 1999. Phylogeography: The History and Formation of Species. Harvard Univ. Press, Cambridge, Massachusetts.
Bennett, K. D. 1990. Milankovitch cycles and their effects on species in ecological and evolutionary time. Paleobiology
16: 11–21.
Bossart, J. L. & D. P. Prowell. 1998. Genetic estimates of population structure and gene flow: Limitations, lessons and new
directions. TREE 13: 202–206.
Chiang, T. Y. 2000. Lineage sorting
accounting for the disassociation between chloroplast and mitochondrial lineages in oaks of southern France. Genome 43: 1090–
1094.
---& K. S. Hong. 1999. Genetic diversity of Quercus species during the postglacial periods. Pp. 431–443 in Y. S. Lin (editor), Proceedings of 1999 Biodiversity Congress.
The Council of Agriculture, Taipei, Taiwan.
---& C. I. Peng. 1998. Phylogeography of the endemic plants in Taiwan. Pp. 148–155 in S. D. Yang (editor), Proceedings of Conservation of Endemic Species. Research Institute of Taiwan Endemic Species.
Nantou, Taiwan.
---& B. A. Schaal. 1999. Phylogeography of ten North American Hylocomium splendens based on nr-DNA ITS sequences. Molec.
Ecol. 8: 1037–1042.
---&--- . 2000a. Molecular evolution and phylogeny of atpB-rbcL noncoding spacer of the chloroplast DNA in the
Hylocomiaceae (mosses, Order Hypnales).
Bot. Bull. Acad. Sin. 41: 85–92.
Latta, R. G. & J. B. Mitton. 1997. A comparison of population differentiation across four classes of gene marker in limber pine (Pinus flexilis James). Genetics 146:
1153–1163.
Li, H. L. 1953. Taxonomic notes on the
Fagaceae of Formosa. Bull. Torrey Club 80:
317–324. Liao, J. C. 1996. Fagaceae. Pp.
51–123 in Editorial Committee of the Flora of Taiwan (editor), Flora of Taiwan, 2nd Ed.
Taipei. Lin, C. C. 1966. An outline of Taiwan’s Quaternary geohistory with a special discussion of the relation between natural history and cultural history in
Taiwan. Bull.
Dept. Archaeol. Anthropol. 23: 7–44.
Manen, J. F. & A. Natali. 1995. Comparison of the evolution of ribulose-1, 5-biphosphate carboxylase (rbcL) and atpB-rbcL
noncoding spacer sequences in a recent plant group, the tribe Rubieae (Rubiaceae). J.
Molec. Evol. 41: 920–927.
Manos, P. S., J. J. Doyle & K. C. Nixon. 1999.
Phylogeny, biogeography, and processes of molecular differentiation in Quercus subgenus Quercus (Fagaceae). Molec.
Phylogenet. Evol. 12: 333–349.
McCauley, D. E. 1994. Contrasting the distribution of chloroplast DNA and allozyme polymorphism among local populations of Silene alba: Implications for studies of gene flow in plants. Proc. Natl.
Acad. Sci. U.S.A. 91: 8127–8131.
Wade, M. J. & D. E. McCauley. 1990.
Extinction and recolonization: Their effects on the genetic differentiation of local populations. Evolution 42: 995–1005.
Whittemore, A. T. & B. A. Schaal. 1991.
Interspecific gene flow in sympatric oaks.
Proc. Natl. Acad. Sci. U.S.A. 88:
2540–2544.
Williams, J. G. K., A. R. Kubelik, K. J. Liak, J.
A. Rafalski & S. V. Tingey. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucl.
Acids Res. 18: 6531–6535.
Wolf, P. G., R. A. Murray & S. D. Sipes. 1997.
Speciesindependent, geographical
structuring of chloroplast DNA haplotypes in a mountain herb Ipomopsis
(Polemoniaceae). Molec. Ecol. 6: 283–291.
Wright, S. 1977. Evolution and the Genetics of
Populations, Vol. 3: Experimental Results and Evolutionary Deductions. Chicago Univ.
Press, Chicago.
Yang, Y. P., H. Y. Liu & S. Y. Lu. 1997.
Manual of Taiwan Vascular Plants, Vol. II.
The Council of Agriculture, Taipei.
Yesodi, V., S. Izhar, H. Hauschner, Y. Tabib &
N. Firon. 1997. Homologous recombination involving cox2 is responsible for a mutation in the cmS-specific mitochondrial locus of Petunia. Molec. Gen. Genet. 255: 106– 114.
Young, N. D. 1998. Pacific coast Iris species delimitation using three species definitions:
Biological, phylogenetic and genealogical.
Biol. J. Linn. Soc. 63: 99–120.
