Pl Syst Evol 271: 29–40 (2008) DOI 10.1007/s00606-007-0609-z Printed in The Netherlands Plant Systematics and Evolution Hybridisation processes in sympatric populations of pines Pinus sylvestris L., P. mugo Turra and P. uliginosa Neumann W. Wachowiak,1,2 W. Prus-Głowacki2 1 2 Institute of Dendrology, Polish Academy of Sciences, Kórnik, Poland Department of Genetics, Institute of Experimental Biology, Adam Mickiewicz University, Poznań, Poland Received 7 May 2007; Accepted 4 September 2007; Published online 4 December 2007
Springer-Verlag 2007 Abstract. Natural hybridisation was postulated between the closely related pine species Pinus sylvestris and the P. mugo complex, however no clear evidence on propagation of mature hybrids in nature has been documented so far. To test the hybridisation hypothesis we applied chloroplast DNA (cpDNA) markers and isozymes in the analyses of 300 individuals representing the variety of morphological forms in the sympatric populations of P. sylvestris, P. mugo and P. uliginosa at the peat bog complex in the Sudety Mts., Poland. Additionally, the haplotypes of paternally inherited cpDNA of 149 open pollinated progeny derived from seeds were compared to the haplotypes of parental trees to access the intensity and direction of contemporary hybridisation. The morphologically highly variable polycormic (multistemmed) hybrids between P. mugo and P. uliginosa were identified. The second group of hybrids was found among the monocormic (single-stemmed) P. sylvestris-like individuals carrying the cpDNA from P. mugo complex. Hybrids of P. sylvestris as a pollen donor and P. mugo or P. uliginosa as a mother were not found, either in the group of examined trees, or among the open pollinated progeny. The results indicate that numerous hybrids can exist in the sympatric population of the species studied and that gene flow can successfully proceed from P. mugo complex to P. sylvestris. Hybridisation and ecological selection seems to play a significant role in diversification and evolution of the investigated species. Keywords: P. sylvestris; P. mugo; P. uliginosa; hybridisation; molecular markers; reproductive barrier; sympatric population; speciation Natural hybridisation is recognised as an important process leading to diversification and adaptive evolution in plants and animal species (Lewontin and Birch 1966, Arnold 1997). There are many examples, which show that hybrid genotypes may have equivalent or even higher fitness as compared to parental species and can be favoured in a given environment. Even in case of initially reduced fertility or viability of hybrids from early generations, the gene flow can proceed in the populations leading to propagation of hybrids and speciation (Arnold et al. 1999). The genus Pinus is the largest in conifers and is divided into two monophyletic subgenera Correspondence: Witold Wachowiak, Institute of Dendrology, Polish Academy of Sciences, Porkowa 5, 62-035 Kórnik Poland; current address: Department of Genetics, University of Oulu, 3000, 90012 Oulu, Finnland e-mail: witold.wachowiak@oulu.fi 30 W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex including Haploxylon (subgenus Strobus) and Diploxylon (subgenus Pinus). It contains about one hundred species widely distributed in the northern hemisphere and some tropical and subtropical areas (Critchfield and Little 1971). Hybridisation in pines was detected in several species including P. halepensis and P. brutia in Turkey (Bucci et al. 1998), P. contorta and P. banksiana in Canada (Wagner et al. 1987), Pinus pumila and P. pentaphylla in Japan (Watano et al. 1996) or P. taeda and P. echinata in the USA (Chen et al. 2004). Natural hybridisation was postulated also between closely related Scots pine (P. sylvestris) and the taxa from P. mugo complex including dwarf mountain pine (P. mugo Turra) and peatbog pine (P. uliginosa Neumann) (Siedlewska 1994, Boratyński et al. 2003). P. sylvestris is the most widespread forest tree species in Europe and Asia whereas P. mugo is an endemic species typical to the mountain regions of Europe (Critchfield and Little 1971). P. uliginosa was described in the Central Sudetes (Neumann 1837) where it grows mainly on peat bogs. The present distribution of Scots pine is a result of postglacial migration from several glacial refugia (Willis and Andel 2004, Cheddadi et al. 2006). It is supposed that recolonisation created the zones of secondary contacts between isolated local populations from ice free regions which survived the last glacial maximum with populations from southern refugia. As the ranges of P. mugo and P. sylvestris overlapped in some part of their distribution, hybridisation between the species was supposed to contribute to high diversification within P. mugo complex (Christensen 1987a). It was also suggested that P. uliginosa could result from ancient cross-pollination between P. sylvestris and the taxa from P. mugo complex (PrusGłowacki et al. 1998, Lewandowski et al. 2000). At present P. sylvestris and P. mugo have mostly allopatric distribution. Some sympatric populations of the species were reported on low lying peatbogs from post glacial period occupied by relict populations of P. mugo and surrounded by extensive forest stand of other conifers including Picea abies. The natural hybridisation between the species was studied in several populations, however the biometric studies were limited by the lack of diagnostic characters suitable for identification of hybrids. The estimates of hybridisation intensity based on anatomical and morphological techniques varies from rare formation of hybrids (Christensen and Dar 1997) to the formation of putative hybrid swarms (Staszkiewicz 1993). The hybridisation hypothesis was tested with the use of serological techniques and isozymes. Mixed traits of antigenic proteins as compared to putative parental species were found in the populations suggesting the possibility of hybridisation (Prus-Głowacki et al. 1981). No fixed differences in isozymes were found between the P. sylvestris and P. mugo complex. In the majority of the peat bog populations the allele frequencies in polycormic individuals were similar to those observed in dwarf mountain pine from continuous range, which suggested rather the low extent of hybridisation (Filppula et al. 1992, Neet-Sarqueda 1994, Odrzykoski 2002). No evidence of ongoing hybridisation was found in the studies applying RFLP markers (Filppula et al. 1992, Odrzykoski 2002). In the controlled crossing experiments, the hybridisation barriers between P. sylvestris and P. mugo were observed by Wachowiak et al. (2005a, 2006a), whereas Kormutak et al. (2005) successfully crossed the species in both directions. Recently, DNA markers of paternally inherited chloroplast DNA were described for P. sylvestris and P. mugo complex (Wachowiak et al. 2000, Wachowiak et al. 2006a). The markers were applied in hybridisation studies in two putatively hybridising populations of the species. The ongoing but very rare hybridisation was detected in P. sylvestris and P. uliginosa population but no evidence of the existence of hybrid trees was found (Boratyńska et al. 2003, Wachowiak et al. 2005b). The ongoing hybridisation was also detected in the sympatric population of P. sylvestris and P. mugo but only one hybrid tree was identified (Wachowiak et al. 2006b). The study questioned the existence of a hybrid swarm between the species in the investigated population. In the study, we applied DNA markers and isozymes to identify hybrids in the sympatric W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex population of P. sylvestris, P. mugo and P. uliginosa from the ‘‘Torfowisko pod Zieleńcem’’ reserve in Poland. The occurence of the three pine species in a very diverse habitat of the peat-bog complex gives unique opportunity for studying adaptive evolutionary processes involving natural hybridisation. Specifically, we asked the question if natural hybridisation takes place in this population and leads to propagation of hybrids trees? Then, if the observed patterns of hybridisation are consistent with our previous investigations in two populations of different species composition? And finally, what can be the evolutionary consequences of hybridisation in the studied group of taxa? We demonstrate here that hybridisation can proceed in natural population of P. sylvestris, P. mugo and P. uliginosa and may produce many fertile hybrids competing with parental species. High intensity of hybridisation accompanied by ecological selection seems to be meaningful for the evolution of the sympatric populations of the analysed taxa. Materials and methods Study area and sampling. Plant material was collected at the ‘‘Torfowisko pod Zieleńcem’’ reserve (called hereafter Zieleniec reserve) which is the largest peat bog complex in the Sudety Mountains, the southwest part of Poland. The formation of peat started about 9.000 to 7.500 years ago and at present the reserve covers the area of about 156 ha. P. sylvestris are found mostly on dryer part of the peat bog growing in close vicinity of the taxa from P. mugo complex including P. mugo Turra and P. uliginosa Neumann. 300 individuals representing the phenotypic forms observed at Zieleniec reserve were collected from the area of the entire peat bog. These included individuals classified as P. sylvestris (85 in total), P. mugo (37), P. uliginosa (66) and 112 oligo- and polycormic (multistemmed) individuals of atypical morphology, which could not be classified to either of the above taxa. Selected phenotypic traits, i.e. growth form, bark colour of the upper part of trunk and main branches, colour and shape of needles and setting angle of conelet from the previous year were used for preliminary taxonomic classification. Samples of one-year old twins including winter buds were collected from selected trees. Additionally, 149 31 open pollinated seeds from seven trees were analysed. Mixed pool of seeds from a few cones was analysed separately for each individual. The seeds were germinated for two weeks and the seedlings were used for further analyses. The seeds were derived from one P. sylvestris, one P. uliginosa, two P. mugo and three individuals identified in the course of the analysis as hybrids. DNA extraction and cpDNA markers application. The needles of mature trees (ca. 100 mg of fresh material) and the whole two-weeks old seedlings were used for DNA extraction following the CTAB (cetyltrimethylammonium bromide) protocol (Wachowiak et al. 2006a). Species diagnostic to P. sylvestris and P. mugo cpDNA haplotypes were defined with the use of two DNA markers. One of them represents single nucleotide restriction site polymorphism in the trnL-trnF region (Wachowiak et al. 2000). It can be detected with the use of PCRRFLP method and DraI restriction enzyme which leads to undigested PCR product for P. sylvestris (haplotype S) and digested (two bands) for P. mugo (haplotype M). The sequence analysis of this region in P. uliginosa indicated its identity to P. mugo (Wachowiak et al. 2005b). PCR-amplification was carried out in a total volume of 15 ll containing about 10 ng of template DNA, 2.5 mM MgCl2, 100 lM of each of dNTP, 0.2 lM each of primer and 0.25 U Taq polymerase (Fermentas, Lithuania) with the respective 1x PCR buffer following the cycle profile and primers as previously reported (Wachowiak et al. 2000). The PCR products (10 ll) were subjected to the over night restriction analyses at 37C. After digestion, the samples were separated in 2% agarose gel, stained with ethidium bromide and analysed under UV light. The second species-diagnostic DNA marker originated from the chloroplast microsatellite region Pt41093 (Vendramin et al. 1996). Teufel (unpublished) found that the length differences in this region varied from 86 to 92 bp (> 86) for P. mugo and from 78 to 82 bp (< 82) for P. sylvestris and thus clearly distinguishes between the two species. This result was further confirmed in the analyses of individuals from controlled crosses (Wachowiak et al. 2006a). P. uliginosa length variation of Pt41093 microsatellite region is within the range for P. mugo. PCR-amplification was carried out in a total volume of 25 ll containing about 20 ng of template DNA, 2.5 mM MgCl2, 100 lM of each dNTP, 0.2 lM of each primer and 0.25 U of Taq polymerase with the respective 1x PCR buffer (Fermentas, Lithuania). PCR was run in a 32 W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex Personal Cycler (MJ Research, USA). The PCR products were separated in a 8% polyacrylamide gel (39:1 acrylamide:bisacrylamide, Sigma), stained with ethidium bromide and analysed under UV light. The haplotype analyses. Both PCR-RFLP and microsatellite markers were applied to determine the haplotypes of mature trees. The data were compared to the phenotype of each individual. PCR-RFLP marker was applied in the analyses of open pollinated progeny derived from individuals classified on the basis of morphological traits as P. sylvestris (26 seedlings from one tree), P. mugo (28 seedlings from two trees), P. uliginosa (30 seedlings from one tree) and from three individuals tentatively classified as P. sylvestris but discovered to carry the cpDNA haplotype of P. mugo complex (65 seedlings). The species-diagnostic cpDNA haplotypes of progeny were compared to the haplotypes of mother trees. The results of haplotype analyses were compared to the outcomes of the previous studies of trees and an open pollinated progeny from the sympatric population of P. sylvestris and P. uliginosa (Wachowiak et al. 2005b) and P. sylvestris and P. mugo (Wachowiak et al. 2006b). Isozyme analyses. Out of 300 individuals genotyped at two marker loci five groups of trees were selected for isozyme studies (Table 1). These Table 1. included 34 P. sylvestris individuals (PUZ), 28 P. mugo (PMZ), 32 P. uliginosa (PUZ), a group of 29 P. sylvestris with cpDNA of P. mugo complex (P. sylvestris-like H1) and a group of 30 morphologically variable, oligo- and polycormic individuals, which could not be phenotypically classified as a pure species (Polycormic H2). Electrophoresis in starch gel was used for isozymes studies following the separation, staining procedures and genetic interpretation of the results as described by Odrzykoski (2002). All samples were genotyped at 10 enzymatic loci including: 6-phosphogluconate dehydrogenase (6PGD – E.C. 1.1.1. 44), malate dehydrogenase (MDH – 2 loci – E.C. 1.1.1.37), glutamate dehydrogenase (GDH – E.C. 1.4.1.3), shikimate dehydrogenase (SHDH – 2 loci – E.C. 1.1.1.25), diaphorase (DIA – E.C. 1.6.99), glutamateoxalacetic transaminase (GOT – 3 loci – E.C. 2.6.1.1). Allelic variants 6PgdB2 and MdhC2 were previously found to be more frequent in P. mugo in comparison to P. sylvestris, and called semi-diagnostic by Odrzykoski (2002). The allele frequency was compared between the five groups and with the allele frequency of pines studied by Wachowiak et al. (2006b). They included three samples of polycormic (BP 5, BP 7, BP 9) and three Location of populations and the sample size of present and the reference isozyme studies No. Species /sample N Location 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 P. sylvestris PSZ P. mugo PMZ P. uliginosa PUZ P. sylvestris-like H1a Polycormic H2b P. mugo DB P. mugo ZT P. mugo ST P. mugo BP5c P. mugo BP7c P. mugo BP9c P. sylvestris BM1d P. sylvestris BM2d P. sylvestris BM3d P. sylvestris ZF P. sylvestris PN20 34 28 32 29 30 51 40 56 64 87 42 49 50 72 101 53 Torfowisko pod Zieleńcem, Sudety Torfowisko pod Zieleńcem, Sudety Torfowisko pod Zieleńcem, Sudety Torfowisko pod Zieleńcem, Sudety Torfowisko pod Zieleńcem, Sudety Dubrawiska, Tatra Mts. _ Zółta Turnia, Tatra Mts. Stawy Toporowe, Tatra Mts Bór na Czerwonem, Nowy Targ Bór na Czerwonem, Nowy Targ Bór na Czerwonem, Nowy Targ Bór na Czerwonem, Nowy Targ Bór na Czerwonem, Nowy Targ Bór na Czerwonem, Nowy Targ Puszcza Zielonka, Poznań PN-20 seed orchard, Olsztyn a Reference Mts Mts Mts Mts Mts Present study Present study Present study Present study Present study Odrzykoski 2002 Odrzykoski 2002 Odrzykoski 2002 Odrzykoski 2002 Odrzykoski 2002 Wachowiak et al. 2006b Odrzykoski 2002 Odrzykoski 2002 Wachowiak et al. 2006b Myczko 2001 Odrzykoski 2002 P. sylvestris-like hybrids with cpDNA from P. mugo complex; boligo- and polycormic multistemmed individuals of atypical morphology, which on the base of cpDNA haplotypes and isozymes studied were concluded to represent hybrids between P. mugo and P. uliginosa; cpolycormic pines assumed to represent mostly pure P. mugo; dmonocormic pines assumed to represent pure P. sylvestris W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex monocormic (BM 1, BM 2, BM 3) pines from different regions from sympatric population of P. sylvestris and P. mugo from Bór na Czerwonem peat bog and the reference pure P. mugo and P. sylvestris (Odrzykoski 2002). P. mugo from Tatra Mts. (Poland) originated from Dubrawiska (DB), _ Zółta Turnia (ZT) and Wy_zni Staw Toporowy peat bog (ST). Samples of P. sylvestris come from Puszcza Zielonka in Poland (Myczko 2001) and from the seed orchard PN-20 that contains trees from northern Poland (Odrzykoski 2002). The list of reference population samples is presented in Table 1. GenAlex software was used to calculate allelic frequencies and Nei’s (1978) genetic distances between all groups. The genetic distances were used to conduct cluster analysis in MEGA 3 using the Unweighted Pair Group Method with Arithmetic Mean. Results Identification of hybrids. The results of cpDNA haplotypes analysis of 300 trees are summarised in Table 2. No other haplotypes than previously described for P. sylvestris and P. mugo complex (P. mugo and P. uliginosa) were observed in the examined group of individuals. The whole group of 37 P. mugo individuals, 66 P. uliginosa individuals and 112 oligo- or polycormic individuals of differentiated morphology had 33 cpDNA haplotypes diagnostic for P. mugo complex. Among the 85 individuals tentatively classified as P. sylvestris only 50 displayed cpDNA haplotypes typical for the species. The remaining 35 P. sylvestris-like individuals carried cpDNA haplotypes diagnostic for P. mugo complex (P. mugo and P. uliginosa). The cpDNA haplotypes of an open pollinated progeny are presented in Table 3. The hybrid seedlings with species diagnostic cpDNA haplotypes discordant with the haplotype of a parental tree were detected for the P. sylvestris progeny as well as for the P. sylvestris-like individuals but carrying the P. mugo cpDNA. The results of previous cpDNA haplotype studies of trees and open-pollinated progeny in sympatric populations of P. sylvestris/P. uliginosa and P. sylvestris/ P. mugo are also summarised in Table 2 and Table 3. Isozyme analysis. The frequencies of the most common alleles at 10 studied loci among the groups of individuals are presented in Table 4. Two semidiagnostic alleles for P. mugo (6PgdB2 and MdhC2) were the most frequent among P. mugo and P. uliginosa from Zieleniec reserve, three P. mugo populations from the reference group and among the Table 2. The cpDNA haplotypes and number of hybrid trees identified among the selected taxa from Zieleniec reserve and the results from the reference studies in a sympatric populations of P. sylvestris and P. mugo complex (P. mugo and P. uliginosa). Hybrids were found among the P. sylvestris-like individuals and oligo- and polycormic individuals as revealed from cpDNA and isozyme studies. M – haplotypes species diagnostic for P. mugo complex, S – for P. sylvestris No Taxa/pines 1 2 3 4 5 6 7 8 9 P. sylvestris P. mugo P. uliginosa Polycormic H2 P. uliginosa P. uliginosa P. sylvestris Polycormic Monocormic Location Number of trees Torfowisko pod Zieleńcem 85 Torfowisko pod Zieleńcem 37 Torfowisko pod Zieleńcem 66 Torfowisko pod Zieleńcem 112 The Stołowe Mts. 32 Low Silesian Pinewood 28 Low Silesian Pinewood 8 Bór na Czerwonem 42 Bór na Czerwonem 72 cpDNA Number Reference haplotypes of hybrids M S 35 37 66 112 32 28 0 42 1 50 0 0 0 0 0 8 0 71 35 0 0 112 0 0 0 0 1 Present study Present study Present study Present study Wachowiak et Wachowiak et Wachowiak et Wachowiak et Wachowiak et al. al. al. al. al. 2005b 2005b 2005b 2006b 2006b 34 W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex Table 3. The cpDNA haplotypes and number of hybrids identified among the open pollinated progeny derived from selected individuals (taxa) from Zieleniec reserve and the results from the reference studies. Hybrid seedlings of F1 generation were found among the progeny from P. sylvestris and P. uliginosa. Hybrids of further than F1 generations were produced by hybrid individuals of P. sylvestris-like phenotype. M – haplotypes species diagnostic for P. mugo complex, S – for P. sylvestris Taxa/pines Location Number cpDNA Number Reference of progeny haplotypes of hybrids P. sylvestris Zieleniec reserve 26 P. mugo Zieleniec reserve 28 P. uliginosa Zieleniec reserve 30 P. sylvestris-like H1 Zieleniec reserve 65 P. uliginosa Low Silesian Pinewood 487 P. sylvestris Low Silesian Pinewood 329 Polycormic Bór na Czerwonem 43 Monocormic Bór na Czerwonem 22 polycormic pines from Bór na Czerwonem reserve, which in the study by Wachowiak et al. (2006b) were considered to represent mostly pure P. mugo. Both alleles were also the most frequent among the group of polycormic individuals from Zieleniec reserve, whereas the group of P. sylvestris-like individuals with cpDNA of P. mugo complex had the higher frequency of only MdhC2 allele. Contrary, P. sylvestris from Zieleniec reserve similarly to monocormic pines from Bór na Czerwonem reserve (considered by Wachowiak et al. (2006b) as a pure P. sylvestris) and the two reference P. sylvestris populations had the most frequent allele 6PgdB1 and MdhC1. In the remaining loci the most frequent alleles were shared between the analysed groups. The exception was allele Gdh1, which was more frequent in P. uliginosa from Zieleniec reserve _ and P. mugo from Zółta Turnia and the allele Sdh2 more frequent in P. mugo from Zieleniec reserve and the group of polycomic pines from this area. Genetic distances between the groups are presented in Table 5 and the relationships between populations are further demonstrated on a dendrogram (Fig. 1). P. mugo and P. uliginosa from Zieleniec reserve cluster together (DN = 0.002) and they show also very close genetic distance to polycormic pines from this M S 15 28 30 59 480 7 43 17 11 0 0 6 7 322 0 5 15 0 0 65 7 7 0 17 Present study Present study Present study Present study Wachowiak et Wachowiak et Wachowiak et Wachowiak et al. al. al. al. 2005b 2005b 2006b 2006b area carrying the cpDNA of P. mugo complex (DN = 0.003, DN = 0.005). The polycormic pines show close genetic distance to the reference allopatric populations of P. mugo (populations 6– 8; DN = 0.025, DN = 0.033 and DN = 0.064, respectively) and to the P. mugo from Bór na Czerwonem reserve (population 9–11; DN = 0.039, DN = 0.031 and DN = 0.045, respectively). Contrary, this group of polycormic pines show very high genetic distance to P. sylvestris from Zieleniec reserve (DN = 0.153), to the reference P. sylvestris from allopatric populations (average DN = 0.137) and to the group of P. sylvestris from Bór na Czerwonem reserve (average DN = 0.129). A genetic distance among the subspecies is higher or equal to 0.05 (Nei 1987). P. sylvestris-like individuals from Zieleniec reserve, which had the cpDNA of P. mugo complex, form a separate cluster with monocormic pines including P. sylvestris from Zieleniec reserve and the reference P. sylvestris populations. The genetic distance between the P. sylvestris-like individuals and P. sylvestris from Zieleniec reserve is 0.030 which is similar to the average genetic distance between P. sylvestris-like individuals and the remaining populations (12–16) of P. sylvestris (average DN = 0.031). This is lower as for P. sylvestrislike individuals and the reference P. mugo (pop- .950 .787 1 .598 .655 .922 .875 1 .625 .672 .964 .869 .927 .702 .846 .932 .898 .983 .756 .813 .938 .873 .968 .797 .800 1 1 1 .737 .566 .647 .500 .500 .922 .922 .951 .725 .755 .950 .938 1 .888 .838 .600 .719 .798 .746 .675 .920 .900 1 .704 .640 .913 .910 .993 .664 .549 .962 .830 .972 .585 .642 .613 .868 .650 .827 .619 .714 .568 .614 .714 .830 .700 .658 .900 .613 .633 .967 .950 1 .931 .762 1 .804 1 .654 .775 .554 .603 .554 .889 .944 .960 .944 .794 .500 .500 .919 .952 .940 .893 .773 .588 .853 .956 .779 1 .516 .720 .882 .800 .500 .500 .823 .516 .956 .735 .786 .500 .500 .982 .536 .825 1 .900 .950 .931 .990 .700 1 .625 1 .646 1 .574 SDH-B DIA-C GOT-A GOT-B GOT-C SDH-A GDH MDH-A MDH-C 6PGD-B 1 2 1 1 2 1 2 1 2 1 1 1 1 1 .750 .828 .844 .970 .640 .984 .750 .625 .994 .662 .992 10 9 8 7 6 5 4 3 2 1 .710 .820 .694 .764 .943 .784 .955 .725 .979 .715 .642 .624 .700 .952 .688 .630 .710 16 15 14 11 12 13 P. sylvestris P. mugo Taxa from Zieleniec reserve Locus Table 4. Frequency of the most common alleles at ten loci in 16 populations from Table 1. In bold – frequency of semidiagnostic alleles for P. mugo W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex 35 ulations 6–8, average DN = 0.062), but similar to P. mugo and P. uliginosa from Zieleniec reserve (DN = 0.043 and DN = 0.039, respectively). The average genetic distance between the reference P. mugo (population 6, 8) and P. sylvestris (population 15–16) is about DN = 0.12. Discussion The presented study documents for the first time, that natural hybridisation in the sympatric population of P. sylvestris, P. mugo and P. uliginosa can lead to propagation of numerous hybrid trees. We identified two groups of hybrids. First one is formed by morphologically highly variable oligoand polycormic individuals which show close genetic identity to both P. mugo and P. uliginosa. Their genetic distance to P. sylvestris from Zieleniec reserve and to P. sylvestris from the two reference populations are even higher that between pure P. sylvestris and P. mugo. This observation and the fact that all individuals in this group display the haplotypes of plastid DNA diagnostic for P. mugo complex indicate that they are hybrids between P. mugo and P. uliginosa with no evidence on contribution of P. sylvestris. No evidence on ongoing hybridisation with P. sylvestris as a pollen donor was also found in the analyses of seeds from P. mugo and P. uliginosa. The close genetic distance of oligoand polycormic hybrids to both P. mugo and P. uliginosa suggests that some of them may represent backcrosses. It is likely that such hybrids are presented in our selected groups of putatively pure P. mugo and P. uliginosa as both groups show much lover genetic distance as found between allopatric populations of the species in other studies (Prus-Głowacki et al. 1998, Lewandowski et al. 2000). The second group of hybrids was identified within monocormic individuals, tentatively classified on the basis of morphological traits as P. sylvestris. In the group of 85 P. sylvestris-like individuals, 35 had cpDNA from P. mugo complex. As chloroplast genome is inherited in paternal line in these species (Wachowiak 2005a), the result indicates that they are hybrids between P. sylvestris as a mother and P. mugo or *** .006 *** .007 .011 *** .005 .007 .008 *** .005 .008 .009 .008 *** .070 .048 .074 .057 .069 *** .006 .084 .057 .083 .074 .087 *** .005 .004 .065 .041 .066 .055 .065 *** .021 .026 .020 .071 .054 .076 .069 .082 *** .050 .022 .022 .025 .158 .113 .148 .129 .145 *** .020 .031 .028 .023 .026 .151 .118 .152 .136 .157 *** .025 .033 .064 .039 .031 .045 .147 .110 .129 .126 .147 *** .051 .068 .068 .049 .024 .028 .025 .038 .025 .033 .023 .036 *** .039 .005 .025 .022 .055 .025 .018 .031 .125 .088 .108 .105 .122 *** .002 .043 .003 .026 .030 .058 .029 .023 .036 .128 .093 .113 .111 .129 *** .134 .125 .030 .153 .169 .154 .097 .072 .089 .073 .014 .014 .012 .004 .007 P. sylvestris PSZ P. mugo PMZ P. uliginosa PUZ P. sylvestris-like H1 Polycormic H2 P. mugo DB P. mugo ZT P. mugo ST P. mugo BP5 P. mugo BP7 P. mugo BP9 P. sylvestris BM1 P. sylvestris BM2 P. sylvestris BM3 P. sylvestris ZF P. sylvestris PN20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Nei’s (1978) genetic distances between studied populations (Table 1) based on allelic frequencies of ten isozyme loci Table 5. *** W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex 16 36 P. uliginosa as a pollen donor. The hybrids reflect the pattern of contemporary hybridisation at this area as revealed by the analyses of an open pollinated progeny. More than 50% of P. sylvestris progeny was of hybrid origin carried the cpDNA from the P. mugo complex. As there is no difference in cpDNA discovered so far between P. mugo and P. uliginosa, the applied markers do not allow to indicate which of the two taxa from P. mugo complex participated in fertilisation. However, our previous studies showed that P. sylvestris can be fertilized in nature by P. uliginosa (Wachowiak et al. 2005b) and P. mugo (Wachowiak et al. 2006b). Therefore, it seems that both P. mugo and P. uliginosa can be a pollen donor to produce monocormic P. sylvestris-like hybrids. The hybrids resembling one of the parental types were found in other pine species including P. taeda and P. echinata (Chen et al. 2004). These studies suggested that they were not F1 individuals but most likely the early generation backcrosses. As shown in our study, the haplotypes diagnostic to both P. sylvestris and P. mugo were found among the seeds derived from P. sylvestris-like hybrids, which indicates their fertility and potential for backcrosses. Further than F1 generation of hybrids were also detected in controlled crosses between P. sylvestris and P. montana (P. mugo complex) (Wachowiak et al. 2006a). In our study, the P. sylvestris-like hybrids showed closer genetic distance to pure P. sylvestris that P. mugo. Therefore it seems that some of the monocormic hybrids can represent backcrosses with P. sylvestris. However, it is unknown if their closer similarity to P. sylvestris can result from selection among the hybrids. In the previous studies, the hybrid seeds and one hybrid tree of P. sylvestris as a mother and P. mugo as a pollen donor were detected in the sympatric population of both species from Bór na Czerwonem reserve in Poland (Wachowiak et al. 2006b). No evidence on reciprocal hybridisation in this population was found either in the group of open pollinated progeny or among the trees. The evidence on reciprocal but very rare ongoing hybridisation was found in the analyses of seeds in the sympatric population of P. sylvestris and W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex 37 P. mugo PMZ P. uliginosa PUZ Hybrid H2 P. mugo DB P. mugo ZT P. mugo ST Policormic BP7 Policormic BP5 Policormic BP9 Hybrid H1 P. sylvestris PSZ P. sylvestris ZF P. sylvestris PN20 Monocormic BM1 Monocormic BM2 Monocormic BM3 0.04 0.03 0.02 0.01 0.00 Fig. 1. Dendrogram constructed on the basis of genetic distances between populations presented in Table 5. In bold - populations from Zieleniec reserve including group of P. sylvestris-like (H1) and Polycormic (H2) pines, considered on the base on isozymes and cpDNA as a hybrids P. uliginosa from We˛gliniec reserve in Poland. However, no hybrid trees were detected there (Wachowiak et al. 2005b). As shown in the presented study, if the three taxa occur sympatrically, two groups of mature hybrids are produced. The results from the three populations of different species composition suggest that P. sylvestris participate in hybridisation as a mother tree. The results show also free natural hybridisation within P. mugo complex (between P. mugo and P. uliginosa). Hybridisation in pines was reported to lead to formation of hybrid zones of different genetic structure (Watano et al. 2004), formation of hybrids resulted from bi-directional introgression and backcrosses (Chen et al. 2004) or to rare hybridisation events limited to individuals of F1 generation (Bucci et al. 1998). The so far estimates of hybridisation intensity and the species composition in the sympatric populations of P. sylvestris and P. mugo varied from rare formation of F1 hybrids to the formation of putative hybrid swarms (Staszkiewicz and Tyszkiewicz 1972, Christensen 1987b, Bobowicz 1990, Neet-Sarqueda 1994). It seems that similarly to the Zieleniec reserve propagation of F1 and next generations of hybrids can freely proceed in the sympatric populations of the taxa from the P. mugo complex. Consequently, the past and contemporary hybridisation within the P. mugo complex could account for the variety of morphological forms reported in other populations of P. mugo (Staszkiewicz and Tyszkiewicz 1969, Filppula et al. 1992, NeetSarqueda 1994). However, the so far data do not support the hypothesis that hybridisation with only P. sylvestris could transform the population of P. mugo or P. uliginosa into a hybrid swarm. The existence of viable hybrids, well adapted to the specific microhabitats of complex environments may lead to their dissemination followed by mutual competition with the parental types (Harrison 1990). The hybrids within P. mugo complex show such signs of expansion at the Zieleniec reserve. Their amount in the entire population can be estimated for more than thirty 38 W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex percent and they constitute the majority of individuals in the central, less humid parts of the peat bog. They gradually replace P. mugo and P. uliginosa in this region. The remaining area is inhabited by the mixture of the pure species and hybrids, in some parts of the peat bog occasionally represented by a single individual. The dryer and more solid external parts of the peat bog are mostly occupied by P. uliginosa, P. sylvestris and P. sylvestris-like hybrids. The number of P. sylvestris-like hybrids seems to be similar to P. sylvestris and they are less numerous in this population than the remaining pine taxa. However, P. sylvestris-like hybrids produce seeds and the viability of analysed seedlings seems to be similar to the pure species. Therefore it is likely that these hybrids can strength the competition with parental species and in the future potentially dominate the external parts of the peat bog. The results of presented study support the hypothesis that past hybridisation events in the contact zones between P. sylvestris and other closely related pine species could play a significant role in the evolution of the P. mugo complex. The example from the Zieleniec reserve demonstrates (1) the existence of viable and fertile hybrids, (2) ecological selection which influence their distribution in the complex microhabitats and (3) the limited gene flow among a certain groups of taxa including the parental species, which creates excellent conditions for further diversification of the taxa from this area. Both theoretical models and experimental studies show that ecological selection can promote diversification of hybrids and speciation (Gross and Rieseberg 2005). It seems likely that the ancient hybridisation processes in the contact zones similar to the Zieleniec reserve between P. sylvestris and P. mugo complex, including an isolated population which survived the glacial maxima, could play a role in speciation in Pinus. Homoploid hybrid speciation in pines, which could potentially involve recombination speciation (Lai et al. 2005), is well documented in P. densata from Tibetan Plateau, a hybrid between P. tabulaeformis and P. yunnanensis (Wang et al. 2001). Additionally, different P. densata populations were found to have unique evolutionary histories and most likely independent hybrid origins (Song et al. 2003). However, more studies are needed to evaluate how common could be speciation in pines through hybridisation. Previous studies showed that P. mugo is more resistant than P. sylvestris to some pathogens including needle cast (Lophodermium seditiosum). These observations motivated the attempts of controlled crosses between the species to produce hybrids for breeding purposes (Prus-Głowacki and Stephan 1998). The hybrids identified in presented study, especially the P. sylvestris-like ones, need detailed biometric and biochemical investigations to access their breeding values and their potential use, as in case of other pine hybrids (Dungey 2001). The applied methods proved to be useful for the analyses of microevolutionary processes going on in the sympatric populations of P. sylvestris and P. mugo complex and could be implemented in similar studies in other putatively hybridising populations. Identification of hybrids gives unique opportunity for more complex studies including the genetics of adaptive variation in this group of taxa. Thanks are given to Ireneusz Odrzykoski and Łukasz Myczko for providing isozymes data. The research was supported by the State Committee for Scientific Research, Poland (KBN Grant no. 0306/ P04/2001/21). References Arnold ML (1997) Natural hybridisation and evolution, Oxford University Press, New York, Oxford, pp 3–10 Arnold ML, Bilger MR, Burke JR, Hempel AL, Williams JH (1999) Natural hybridisation: how low can you go and still be important? Ecology 80(2): 371–381 Bobowicz MA (1990) Pinus mugo Turra · Pinus sylvestris L. hybrids from Bór na Czerwonem reservation in Kotlina Nowotarska (in Polish). Wyd. Nauk. UAM, Poznań Boratyńska K, Boratyński A, Lewandowski A (2003) Morphology of Pinus uliginosa (Pinaceae) needles W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex from populations exposed to and isolated from direct influence of Pinus sylvestris. Bot J Linn Soc 142: 83–91 Boratyński A, Boratyńska K, Lewandowski A, Goła˛b Z, Kiciński P (2003) Evidence of the possibility of natural reciprocal crosses between Pinus sylvestris and P. uliginosa based on the phenology of reproductive organs. Flora 198: 1227–1239 Bucci G, Anzidei M, Madaghiele A, Vendramin GG (1998) Detection of haplotypic variation and natural hybridisation in halepensis-complex pine species using chloroplast simple sequence repeat (SSR) markers. Molec Ecol 7: 1633–1643 Cheddadi R, Vendramin GG, Litt T, Francois L, Kageyama M, Lorentz S, Laurent JM, de Beaulieu JL, Sadori L, Jost A, Lunt D (2006) Imprints of glacial refugia in the modern genetic diversity of Pinus sylvestris. Glob Ecol Biog 15: 271–282 Chen JW, Tauer CG, Bai GH, Huang YH, Payton ME, Holley AG (2004) Bidirectional introgression between Pinus taeda and Pinus echinata: evidence from morphological and molecular data. Canad J For Res 34(12): 2508–2516 Christensen K I (1987a) Taxonomic revision of the Pinus mugo complex and P. · rhaetica (P. mugo · P. sylvestris) (Pinaceae). Nord J Bot 7(4): 383–408 Christensen KI (1987b) A morphometric study of the Pinus mugo Turra complex and its natural hybridisation with P. sylvestris L. (Pinaceae). Feddes Repert 98: 623–635 Christensen K, Dar GH (1997) A morphometric analysis of spontaneous and artificial hybrids of Pinus mugo · P. sylvestris (Pinaceae). Nord J Bot 17: 77–86 Critchfield WB, Little EL (1971) Geographic distribution of the pines of the world. US Department of Agriculture Forest Service, Miscellaneous Publication 991, Washington Dungey HS (2001) Pine hybrids – a review of their use, performance and genetics. For Ecol Manage 148: 243–258 Filppula S, Szmidt AE, Savolainen O (1992) Genetic comparison between Pinus sylvestris and P. mugo using isozymes and chloroplast DNA. Nord J Bot 12: 381–386 Gross BL, Rieseberg LH (2005) The ecological genetics of homoploid hybrid speciation. J Heredity 96(3): 241–252 Harrison RG (1990) Hybrid zones: windows on evolutionary process. Oxford Surv Evol Biol 7: 69–128 39 Kormutak A, Ostrolucka M, Vookova B, Pretova A, Feckova M (2005) Artificial hybridisation of Pinus sylvestris L. and Pinus mugo Turra. Acta Biol Crac Ser Bot 47(1): 129–134 Lai Z, Nakazato T, Salmaso M, Burke JM, Tang S, Knapp SJ, Rieseberg LH (2005) Extensive chromosomal repatterning and the evolution of sterility barriers in hybrid sunflower species. Genetics 171: 291–303 Lewandowski A, Boratyński A, Mejnartowicz L (2000) Allozyme investigations on the genetic differentiation between closely related pines – Pinus sylvestris, P. mugo, P. uncinata, and P. uliginosa (Pinaceae). Pl Syst Evol 221: 15–24 Lewontin RC, Birch LC (1966) Hybridisation as a source of variation for adaptation to new environment. Evolution 20: 315–336 Myczko Ł (2001) Porównanie polimorfizmu genetycznego plantacji sosny zwyczajnej (Pinus sylvestris L.) przed i po cie˛ciu piele˛gnacyjnym w stadium dra˛gowiny. MS thesis, the Adam Mickiewicz University, Poznań, Poland Neet-Sarqueda C (1994) Genetic differentiation of Pinus sylvestris L. and Pinus mugo aggr. populations in Switzerland. Silvae Genet 43: 207–215 Nei M (1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: 583–590 Nei M (1987) Molecular evolutionary genetics, Columbia University Press, New York Neumann C (1837) Über eine auf den Seefeldern bei Reinerz und einigen ähnlichen Gebirgsmooren der königl. Oberförsterei Karlsberg in der Graftschaft Glatz vorkommende noch unbeschriebene Form der Gattung Pinus. Jahresber Schles Gesellsch Vaterl Cultur 11: 52–57 Odrzykoski IJ (2002) Studies on genetic variability in dwarf pine (Pinus mugo) using biochemical and molecular markers (in Polish). Wyd Nauk UAM, Poznań Prus-Głowacki W, Sadowski J, Szweykowski J, Wiatroszak I (1981) Quantitative and qualitative analysis of needle antigens Pinus sylvestris, Pinus mugo, Pinus uliginosa and Pinus nigra and some individuals from a hybrid swarm population. Genet Polon 22: 447–454 Prus-Głowacki W, Bujas E, Ratyńska H (1998) Taxonomic position of Pinus uliginosa Neumann as related to the other taxa of Pinus mugo complex. Acta Soc Bot Pol 67: 269–274 Prus-Głowacki W, Stephan BR (1998) Immunochemical and isoenzymatic characterization of hybrids 40 W. Wachowiak and W. Prus-Głowacki: Hybridisation of P. sylvestris and P. mugo complex from controlled crosses between Pinus montana var. rostrata and Pinus sylvestris. Forest Genet 5: 155– 163 Siedlewska A (1994) Isoenzymatic differentiation in putative hybrid swarm population (Pinus mugo Turra · P. sylvestris L.) from ‘‘Torfowisko Zieleniec’’ peat-bog. Acta Soc Bot Pol 63: 325–332 Song B H, Wang X Q, Wang X R, Ding K Y, Hong D Y (2003) Cytoplasmic composition in Pinus densata and population establishment of the diploid hybrid pine. Molec Ecol 12(11): 2995–3001 Staszkiewicz J, Tyszkiewicz M (1969) Natural hybrids of Pinus mugo Turra · Pinus silvestris L. in Kotlina Nowotarska (in Polish). Fragm Florist Geobot 15: 187–212 Staszkiewicz J, Tyszkiewicz M (1972) Variability of natural Pinus sylvestris L. · P. mugo Turra (P. · rotundata L.) hybrids in southwestern Poland and in selected stands of Czech and Moravia (in Polish). Fragm Florist Geobot 18(2): 173–191 Staszkiewicz J (1993) Variability of Pinus mugo · P. sylvestris (Pinaceae) hybrid swarm in the Tisovnica nature reserve (Slovakia). Pol Bot Stud 5: 33–41 Vendramin GG, Lelli L, Rossi P, Morgante M (1996) A set of primers for the amplification of 20 chloroplast microsatellites in Pinaceae. Molec Ecol 5: 595–598 Wachowiak W, Leśniewicz K, Odrzykoski I, Augustyniak H, Prus-Głowacki W (2000) Species specific cpDNA markers useful for studies on the hybridization between Pinus mugo · P. sylvestris. Acta Soc Bot Pol 69(4): 273–276 Wachowiak W, Lewandowski A, Prus-Głowacki W (2005a) Reciprocal controlled crosses between Pinus sylvestris and P. mugo verified by a species-specific cpDNA marker. J Appl Genet 46(1): 41–43 Wachowiak W, Celiński K, Prus-Głowacki W (2005b) Evidence of natural reciprocal hybridisation between Pinus uliginosa and P. sylvestris in the sympatric population of the species. Flora 200: 563–568 Wachowiak W, Stephan BR, Schulze I, PrusGłowacki W, Ziegenhagen B (2006a) A critical evaluation of reproductive barriers between closely related species using DNA markers – a case study in Pinus. Pl Syst Evol 257: 1–8 Wachowiak W, Odrzykoski I, Myczko Ł, PrusGłowacki W (2006b) Lack of evidence on hybrid swarm in the sympatric population of Pinus mugo and P. sylvestris. Flora 201: 307–316 Wagner DB, Furnier GR, Saghai-Maroof MA, Williams SM, Dancik BP, Allard RW (1987) Chloroplast DNA polymorphism in lodgepole and jack pines and their hybrids. Proc Natl Acad Sci USA 84: 2097–2100 Wang XR, Szmidt AE, Savolainen O (2001) Genetic composition and diploid hybrid speciation of a high mountain pine, Pinus densata, native to the Tibetan plateau. Genetics 159: 337–346 Watano Y, Imazu M, Shimizu T (1996) Spatial distribution of cpDNA and mtDNA haplotypes in a hybrid zone between Pinus pumila and P. parviflora var. pentaphylla (Pinaceae). J Pl Res 109: 403–408 Watano Y, Kanai A, Tani N (2004) Genetic structure of hybrid zones between Pinus pumila and P. parviflora var. pentaphylla (Pinaceae) revealed by molecular hybrid index analysis. Amer J Bot 91(1): 65–72 Willis K J, Andel T H (2004) Trees or no trees? The environmental of central and eastern Europe during the last glaciation. Q Sci Rev 23: 2369–2387