J Appl Genat 50(4), 2009, pp. 329–339 Original article Different patterns of genetic structure of relict and isolated populations of endangered peat-bog pine (Pinus uliginosa Neumann) W. Wachowiak1, W. Prus-G³owacki2 1 2 Polish Academy of Sciences, Institute of Dendrology, Kórnik, Poland Faculty of Biology, Institute of Experimental Biology, Department of Genetics, Adam Mickiewicz University, Poznañ, Poland Abstract. Recent changes in environmental conditions in populations of peat-bog pine (Pinus uliginosa Neumann) caused rapid decline or even extinction of the species in several stands in Central Europe. Conservation strategies for P. uliginosa require information about the evolutionary history and genetic structure of its populations. Using isozymes we assessed the genetic structure of P. uliginosa from four isolated stands in Poland and compared the results to genetic structures of other closely related pine species including eight populations of Pinus mugo, ten of Pinus sylvestris and one of Pinus uncinata. The level of genetic variability of P. uliginosa measured by the mean number of alleles per locus and average heterozygosity was similar to others related to P. uliginosa taxa from the reference group but it differs among populations. High genetic similarity was found between two populations of P. uliginosa from Low Silesian Pinewood. The populations were genetically distinct as compared to other populations including locus classicus of the species from the peat bog at Batorów Reserve. Very low genetic distance (DN = 0.002) and small genetic differentiation (GST = 0.003) were found between P. uliginosa and P. mugo in the sympatric populations of the species from Zieleniec peat bog suggesting the ongoing natural hybridisation and genetic contamination of peat-bog pine from this area. Some evidence for skew in allele frequency distribution potentially due to recent bottleneck was found in population from Low Silesian Pinewood. The analysed open pollinated progeny derived from two P. uliginosa stands from Low Silesian Pinewood showed the excess of homozygotes as compared to the maternal trees indicating high level of inbreeding (F = 0.105, F = 0.081). The results are discussed in the context of evolution of P. uliginosa populations, taxonomic relationships between the analysed species and conservation strategies for active protection of peat-bog pine. Keywords: endangered species, gene flow, genetic structure, isolated populations, izozymes, peat bog pine, Pinus uliginosa. Introduction Peat-bog pine (Pinus uliginosa Neumann) represents one of the four native pine species in Poland and it grows mainly on peat bogs. The species has been described from the stands in the Sto³owe and Bystrzyckie Mountains, Central Sudetes, Poland (Neumann 1837; Wimmer 1837). Taxonomic position of P. uliginosa is still under discussion. On the basis of morphological data, P. uliginosa was considered as a synonym of Pinus rotundata and it was included in Pinus mugo complex (Christensen 1987). Other biometric and biochemical studies could not exclude the hybrid origin of peat-bog pine that may have resulted from ancient cross-pollination between Scots pine (Pinus sylvestris) and a taxa from P. mugo complex (Prus-G³owacki et al. 1998; Lewandowski et al. 2000). Natural stands of P. uliginosa were described in the Sudety Mountains and in Low Silesian Pinewood, Poland (Boratyñski 1994). However, the Received: May 27, 2008. Accepted: October 6, 2009. Correspondence: W. Prus-G³owacki, Department of Genetics, Institute of Experimental Biology, Adam Mickiewicz University, Umultowska 89, 61–614 Poznañ, Poland; e-mail: prusw@amu.edu.pl 330 W. Wachowiak, W. Prus-G³owacki distribution of peat-bog pine has been seriously restricted in recent decades. Extinction of the species was caused mainly by the altered water conditions and transformation of biotopes including drainage and cultivation of peat bogs (Danielewicz and Zieliñski 2000). Due to the rapid decline of P. uliginosa, it is considered a highly endangered and protected species and hence is included in Polish Plants Red Book (Zarzycki and KaŸmierczakowa 1993). Current inventory of P. uliginosa has demonstrated that in Poland more numerous individuals of the species (at least several dozen) are present only in four stands including (1) Batorów Reserve (locus classicus) located in Sto³owe Mountains, (2) Wêgliniec Reserve in Low Silesian Pinewood, (3) Wêglowiec Forestry in Low Silesian Pinewood and (4) Zieleniec Reserve in the Sudety Mountains. The number of trees from Batorów Reserve was estimated to about 400 individuals (Go³¹b 1999). However, recent storms in this area, air pollution, insect gradation and fungal diseases have significantly reduced this population (Go³¹b, personal communication). Most of the trees are more than 80 years old and they are in poor health condition. Due to the very rapid extinction of mature trees and no natural regeneration observed there, this population is highly endangered (Go³¹b 1999). A somewhat better condition was reported in two populations from Low Silesian Pinewood where natural regeneration can be observed. Nevertheless, even these populations are affected by significant reduction in the number of trees. In 1956, the Wêgliniec Reserve contained 208 individuals while in 1998 only 99 trees were observed including many with reduced viability. In the newly described stand in Wêglowiec Forestry, 181 trees were reported, of which only 87 were fully viable (Danielewicz and Zieliñski 2000). However, the origin and relationship between P. uliginosa populations from that area is unclear. For instance, population from Wêglowiec Forestry is supposedly planted (Danielewicz and Zieliñski 2000). The best P. uliginosa stand seems to be at Zieleniec Reserve in the Sudety Mountains, where numerous viable individuals of the species were found. In this population, P. uliginosa grows together with closely related Dwarf Mountain pine (P. mugo) and Scots pine. Controlled crosses and hybridisation studies in natural populations between P. sylvestris and the taxa from P. mugo complex indicated the possibil- ity of their cross-pollination (Wachowiak et al. 2005a, 2005b; Kormutak et al. 2005; Lewandowski and Dering 2006). Therefore, due to a potential of interspecific gene flow, this population may have a hybrid character (Siedlewska and Prus-G³owacki 1995; Wachowiak and Prus-G³owacki 2008). According to currently available data, the total number of viable P. uliginosa trees in Poland is about several hundreds and continues to decrease. It seems that even active protection of peat-bog pine, including the elimination of competitive trees of other species such as P. sylvestris or Picea abies from the populations and creation of conditions for natural regeneration cannot warrant its survival. Therefore, the idea of setting up a seed orchard and reintroduction of P. uliginosa to its optimum habitats has been recently suggested (Danielewicz and Zieliñski 2000). The success of such protection activities and gene pool conservation requires information about the genetic structure of P. uliginosa populations and its evolutionary history. In this study, isozymes were used as genetic markers to assess the genetic structure of four above-described populations of P. uliginosa. The results were compared to genetic structure of several populations of three closely related pine species including P. mugo, P. sylvestris and P. uncinata. The aim was to address the questions about evolutionary and demographic history of P. uliginosa in the context of species protection and development of conservation strategies. As the present peat-bog pine stands are rather small and isolated, we checked if inbreeding and genetic drift have narrowed the gene pools of the populations compared with other closely related pine species. We evaluated relationships between populations to assess the possibility of potential past plant stock transfers. As P. uliginosa populations are exposed to potential gene flow from other pines including P. mugo and P. sylvestris (Boratyñska et al. 2003; Boratyñski et al. 2003; Lewandowski et al. 2005), we evaluated the possibility of the species decline due to genetic pollution and contamination of P. uliginosa gene pool. We also evaluated the genetic structure of P. uliginosa seeds derived from two stands that can potentially serve as a genetic material for reintroduction. Finally, we integrated the available data to define best conservation units and propose conservation strategies for active protection of the species. Isolated populations P. uliginosa - genetic structure Materials and methods Plant material and sampling Four isolated populations of peat-bog pine (Pinus uliginosa Neumann) were included in isozyme studies (Table 1). Population from “Wielkie Torfowisko Batorowskie” Reserve (henceforth mentioned as Batorów Reserve) in the Sto³owe Mountains (50029’ N, 16019’ E) grows in the vicinity of Picea abies forest and the nearest stand of other pines (P. sylvestris) is at a distance of about 2 km (Go³¹b 1999). Two peat bogs populations from Low Silesian Pinewood from “Torfowisko pod Wêgliñcem” (henceforth mentioned as Wêgliniec Reserve) and from Wêglowiec Forestry (51017’ N, 15014’ E) are surrounded by extensive forest stands of P. sylvestris. In the fourth studied population from “Torfowisko pod Zieleñcem” (henceforth mentioned as Zieleniec Reserve) in the Sudety Mountains (50028’ N, 16023’ E) P. uliginosa grows together with closely related pines P. mugo and P. Sylvestris (Figure 1). From each population winter buds from trees selected at random were collected for isozyme analyses. Additionally, the open-pollinated seedlings derived from seeds collected from 33 trees from Wêgliniec Reserve and 30 trees from Wêglowiec Forestry were analysed. The seeds were isolated from several cones and a mixed pool of seeds was obtained for each tree separately. Ten seeds from each tree were mixed to obtain a random pool for the population. Seeds were placed on a filtration paper in Petri dishes and germinated in darkness for two weeks. Seedlings of about 3 cm in length were used for the analyses. As an out-group for comparative analyses of P. uliginosa genetic structure and 331 taxonomic investigations, three closely related pine species including P. mugo, P. sylvestris and P. uncinata were utilised. Seven P. mugo populations originated from Tatra Mountains and one from Zieleniec Reserve. Eight P. sylvestris populations were derived from provenance trial IUFRO 1982 in Supraœl (Poland), and two from natural stands from Spa³a Forest District and from Sto³owe Mountains close to Zieleniec Reserve, respectively. P. uncinata originated from its natural stands in Pyrenees Mountains, Spain (Table 1). Isozyme analyses Electrophoresis in starch gel was used for isozyme studies. The separation and staining procedures along with genetic interpretation of the results was performed as described by Rudin and Ekberg (1978), Yazdani and Rudin (1982), Szmidt and Yazdani (1984), Muona and Szmidt (1985). All the analysed samples were genotyped at 10 polymorphic enzymatic loci including: glutamate dehydrogenase (GDH) E.C.1.4.1.2, malate dehydrogenase (MDH, 2 loci) E.C.1.1.1.37, shikimate dehydrogenase (ShDH, 2 loci) E.C.1.1.1.14, diaphorase (DIAF) E.C.1.6.99, fluorescent esterase (FEst) E.C.3.1.1.1, glutamate oxaloacetate transaminase (GOT, 2 loci) E.C.2.6.1.1 and 6-phosphogluconate dehydrogenase (6PGD) E.C.1.1.1.44. Calculations of genetic parameters such as the number of alleles per locus (A/L), observed heterozygosity (Ho), expected heterozygosity (He), total heterozygosity (HT), fixation index (F), genotype polymorphism index (genotypic diversity – Pg), genetic diversity within populations (DST) and relative measure of amount of genetic diversity among populations (GST) and gene flow (Nm) were conducted as de- Table 1. Locations and size of samples used for studies Taxa P. uliginosa P. uliginosa (seedlings) P. mugo P. sylvestris P. uncinata Pop. No. 1 2 3 4 5 6 7 8 9 10 11 Location Origin Batorów Reserve (PU1) Wêgliniec Reserve (PU2) Wêglowiec Forestry (PU3) Zieleniec Reserve (PU4) Wêgliniec Reserve (PU2 F1) Wêglowiec Forestry (PU3 F1) Tatry Mts., seven populations (M1-M7)* Zieleniec Reserve (MZ) Supraœl, eight populations (S1-S9)** Zieleniec Reserve (MZ) Pyrenees Mts., Cap de Clots (UN) Natural Natural Natural Natural Natural Natural Natural Natural IUFRO-82 Natural Natural No. of individuals 30 33 30 32 89 61 210 30 240 30 50 * including: M1 – ¯leb pod Czerwienic¹ I, M2 – Prze³êcz miêdzy Kopami, M3 – ¯leb pod Czerwienica II, M4 – Litworowy Staw, M5 – Czarny Staw G¹sienicowy, M6 – Dubrawiska, M7 – Hala G¹sienicowa ** including: S1 – Rashchinskaya Dacha (Russia), S2 – Kondezhskoe (Russia), S3 – Silene (Latvia), S4 – Spa³a (Poland), S5 – Haguenau (France), S6 – Zahorie (Slovakia), S7 – Bystrzyca K³odzka (Poland), S8 – Czersk (Poland), S9 - Szczeliniec (Poland). 332 W. Wachowiak, W. Prus-G³owacki scribed by Jain and Workman (1967), Nei and Roychoudhry (1974), Hedrick (1974), Nei (1987), Slatkin (1985), El-Kassaby (1991). Genetic distances between analysed populations and species were calculated on the basis of allele frequency (Nei, 1972) and the results of cluster analyses were presented on a dendrite based on genetic similarities between studied groups. Genetic differentiation between P. uliginosa populations was studied by FST statistics based on allelic frequency data for each population at all loci combined using Arlequin ver. 3.0. (Excoffier et al. 2005). Statistical significance of the results was evaluated using 1000 permutations, where samples were randomly assigned into different populations. Genetic clustering of P. uliginosa populations was conducted using BAPS 5.2 (Corander et al. 2008). Results of the analyses were also used to detect signs of admixture in each population. For detecting recent effective population size reductions from allele frequencies data we used BOTTLENECK ver. 1.2.02 software (Cornuet and Luikart 1996). A recently bottlenecked population exhibits a correlative reduction in the number of alleles and gene diversity at polymorphic loci (Cornuet and Luikart 1996). To determine whether a population exhibits a significant number of loci with gene diversity excess, we performed sign and Wilcoxon sign-rank tests and used a qualitative descriptor of the allele frequency distribution (mode-shift indicator) to discriminate bottlenecked populations from stable populations (Luikart and Cornuet 1998). Results Genetic structure of populations All isozyme loci were polymorphic in all the studied species. Detailed data on genetic parameters characterising the genetic structure of the analysed populations are listed in Table 2. In general, medium level of genetic variation was observed in P. uliginosa as compared to the three taxa from the reference group. The mean number of alleles per locus (A/L) ranged between 2.09 in P. uliginosa from Wêgliniec Reserve (lowest in whole group) to 2.82 in P. uncinata (the highest one). Similar pattern was observed for mean number of genotypes per locus which showed the highest values in P. uncinata (3.81), the lowest in P. sylvestris from Spa³a Forest District (3.09) and intermediate values in three P. uliginosa and P. mugo populations (3.27) and in P. uliginosa from Wêglowiec For- estry (3.54). The coefficient of genetic polymorphism (Pg) showed high and similar values in two P. uliginosa populations from the Low Silesian Pinewood (Pg = 0.506 and Pg = 0.511) and the lowest values for P. uliginosa from Zieleniec Reserve (Pg = 0.388). Highest observed heterogeneity was noted for P. uliginosa from Wêglowiec Forestry (Ho = 0.376) and the lowest one in the group of trees from Zieleniec Reserve (0.237). Population from Zieleniec showed an excess of homozygotes (about 12%) as compared with P. uliginosa from Wêgliniec Reserve (about 2%) and Wêglowiec peat bog (about 4%). Contrary, P. uliginosa from Batorów Reserve showed some excess of heterozygotes (about 4%), however much lower than that observed in P. mugo population from Tatra Mountains where the mean multilocus F coefficient was F = -0.110. The measures of intra- (DST) and inter-population variability (GST) in pair-wise comparisons of P. uliginosa to the other studied taxa indicated its lowest similarity to P. sylvestris. Very low level of genetic diversity was observed among P. uliginosa populations from Low Silesian Pinewood (DST = 0.005) and between P. uliginosa from Zieleniec Reserve and P. mugo (0.001) (Table 3). Genetic distances and further relationship between populations on the basis of allelic frequency are presented on a dendrite based on the closest neighbourhood similarities between the analysed groups (Figure 2). Two clearly distinguished clusters are formed by the group of P. sylvestris populations versus the taxa from P. mugo complex. Mean genetic distance within P. mugo and P. sylvestris populations was much smaller than distance between P. uliginosa populations. Two P. uliginosa populations from Low Silesian Pinewood are clustered together and are genetically distinct from P. uliginosa from Batorów and population from Zieleniec Reserve. The latter population shows striking similarity to P. mugo from Zieleniec as they both cluster together with P. mugo populations from Tatra Mountains. The high gene flow (Nm) also indicates a very close genetic relationship between these two groups (Table 3). Based on FST statistics, high genetic differentiation was found between P. uliginosa populations from different geographical locations. The most genetically different was the population from Zieleniec Reserve as compared with the other groups. No genetic differentiation was found between two populations from Low Silesian Pinewood (Table 4). Similar pattern of population structure was observed in BAPS spatial clustering 333 Table 2. Genetic variation among studied populations. A/L – number of alleles per locus, G/L – number of genotypes per locus, Pg – genotypes polymorphism index, Ho – observed heterozygosity, He – expected heterozygosity, F – fixation index. Description of populations as in Table 1. Taxa/ Population Pinus uliginosa 1. PU1 2. PU2 3. PU3 4. PU4 5. PU2 F1 6. PU3 F1 Pinus mugo 7. M1-7 Pinus sylvestris 8. S1-10 Pinus uncinata 9. UN A/L G/L Pg Ho He F 2.45 2.09 2.36 2.64 2.45 2.82 3.27 3.27 3.54 3.27 3.46 3.91 0.428 0.506 0.511 0.388 0.490 0.525 0.328 0.372 0.376 0.237 0.329 0.342 0.315 0.379 0.391 0.272 0.358 0.382 –0.040 0.018 0.038 0.127 0.081 0.105 2.72 3.27 0.398 0.303 0.273 –0.110 2.27 3.09 0.415 0.299 0.297 –0.008 2.82 3.81 0.442 0.315 0.320 0.016 Table 3. Genetic differentiation between the studied populations. HT – total heterozygosity, HS – heterozygosity in the samples, DST – genetic diversity among populations, GST – proportions of total genetic diversity among populations, Nm – gene flow. PM – average data for 8 populations, PS – average data for 10 populations. Description of populations as in Table 1. Populations (Taxa) P. uliginosa vs. P. uliginosa PU1 / PU2 PU1 / PU3 PU1 / PU4 PU2 / PU3 PU2 / PU4 PU3 / PU4 P. uliginosa vs. P. uncinata PU1 / UN PU2 / UN PU3 / UN PU4 / UN P. uliginosa vs. P. mugo PU1 / PM PU2 / PM PU3 / PM PU4 / M7 PU4 / MZ P. uliginosa vs. P. sylvestris PU1 / PS PU2 / PS PU3 / PS PU4 / PS P. uliginosa vs. P. uliginosa PU2 / PU2 F1 PU3 / PU3 F1 PU2 / PU3 F1 PU3 / PU2 F1 PU2 F1 / PU3 F1 P. sylvestris vs. P. mugo PS/PM HT HS DST GST Nm 0.365 0.370 0.331 0.389 0.353 0.350 0.347 0.351 0.315 0.384 0.324 0.326 0.018 0.019 0.017 0.005 0.029 0.023 0.049 0.052 0.051 0.012 0.082 0.067 4.9 4.6 4.7 19.8 2.8 3.5 0.372 0.370 0.363 0.338 0.350 0.343 0.346 0.323 0.022 0.027 0.017 0.016 0.059 0.073 0.046 0.046 4.0 3.2 5.2 5.2 0.319 0.360 0.352 0.281 0.293 0.295 0.328 0.332 0.269 0.292 0.024 0.032 0.020 0.012 0.001 0.075 0.088 0.058 0.044 0.003 3.1 2.6 4.1 5.4 83.1 0.344 0.380 0.384 0.318 0.306 0.339 0.343 0.273 0.037 0.041 0.041 0.044 0.082 0.108 0.107 0.139 6.9 2.1 2.1 1.6 0.369 0.388 0.388 0.369 0.370 0.364 0.385 0.380 0.366 0.368 0.005 0.003 0.008 0.003 0.002 0.015 0.008 0.020 0.007 0.006 16.9 32.0 12.0 34.0 41.0 0.339 0.285 0.054 0.159 1.3 334 W. Wachowiak, W. Prus-G³owacki to population from Batorów Reserve (data not shown). Genetic structure of seeds of P. Uliginosa from Low Silesian Pinewood Genetic parameters of seedlings derived from the seeds collected at Wêgliniec Reserve and Wêglowiec peat bog are presented in Table 2. In general, higher number of alleles and genotypes was found in the group of seedlings as compared to maternal populations. Higher genetic variation was observed in the pool of seeds from Wêglowiec Forestry than in Wêgliniec Reserve. Low genetic differentiation (about 0.8% and 1.5 %, respectively) and low genetic diversity (DST = 0.005 and DST = 0.003, respectively) was found among the analysed groups of seedlings and maternal populations. The multilocus F coefficient has positive value in both groups from Wêgliniec Reserve and Figure 1. Approximate location of sampled Pinus uliginosa populations from Batorów Reserve (D), Wêgliniec Reserve (), Wêg³owiec Forestry (~) and Zieleniec Reserve () in Poland UN 0.019 MZ 0.002 M7 S1 PU4 0.006 PU2 M6 PU3 0.052 0.018 0.010 0.005 0.006 0.006 0.006 M1 M3 M2 0.004 0.003 0.002 M4 S10 S2 0.011 S4 S3 0.035 0.006 0.007 S6 S5 0.004 0.007 S7 0.006 M5 S8 0.012 0.032 S9 PU1 Figure 2. OTU dendrite according to Jain and Workman (1967) based on genetic similarities of P. uliginosa (PU), P. mugo (M), P. sylvestris (S) and P. uncinata (UN) populations. Description of populations as in Table1. Table 4. Genetic differentiation (FST) between P. uliginosa populations and F1 progeny. Description of populations as in Table 1. PU1 PU2 PU3 PU4 PU2 F1 PU3 F1 0.070*** 0.079*** 0.242*** 0.099*** 0.086*** PU2 0.016 0.262*** 0.035*** 0.040*** analysis with the highest probabilities of grouping the populations into three clusters corresponding to Batorów Reserve, Zieleniec Reserve and two populations from Low Silesian Pinewood clustered together. Some evidence of admixture was found in one individual from Wêliniec Reserve showing the combination of alleles, more similar PU3 0.236*** 0.015* 0.008 PU4 0.254*** 0.235*** PU2 F1 0.003 Wêglowiec Forestry, indicating 8% and 11% excess of homozygotes, respectively. The F1 progenies were not genetically different from each other based on FST statistics. However, both groups of seedlings were differentiated as compared to all other groups of populations (Table 4), except the progeny from Wêg³owiec For- Isolated populations P. uliginosa - genetic structure estry being genetically uniform as compared to its maternal population (FST = 0.008, P > 0.05). Based on Bayesian population structure analysis, the two groups of seedlings clustered together with the two maternal populations from Low Silesian Pinewood and are genetically differentiated as compared to populations from both Batorów and Zieleniec Reserve. A sign of admixture from other groups was found only in one individual of F1 progeny from Wêglowiec Forestry (data not shown). Genetic evidence for population bottlenecks Under infinite allele mutation model applied, some evidence for significant excessive heterozygosity was found at GOT B locus in two populations from Low Silesian Pinewood (P < 0.05 and P < 0.005 in sign and Wilcoxon tests for mutation drift equilibrium across loci, respectively) and among the F1 progeny from the population from Wêgliniec Reserve (P < 0.05 in both tests). Excess of heterozygotes was also found at GDH and MDH C loci atBatorów and Wêgliniec Reserve, respectively. For most populations, the allele frequency distribution was approximately L-shaped as expected under mutation-drift equilibrium. Evidence for a significant mode-shift in allele frequencies was found only for samples from Wêgliniec Reserve (PU2) suggesting a recent bottleneck in that population. Discussion Rapid decline of P. uliginosa was observed in the last decades. This has resulted in the extinction of the species from several previously reported stands and reduction of its range to a few isolated populations of limited number of individuals and weak reproductive ability. Such changes in the species population size may significantly influence its genetic structure. However, our data indicates that despite the low number of individuals in the studied P. uliginosa populations, their genetic variability level (A/L, Pg, Ho) is high and similar to P. mugo and P. uncinata populations and even higher than that in P. sylvestris. Therefore, it seems that genetic diversity has been shaped in the past when the populations were much more numerous and their current decline did not affect much of their genetic variability. The only signatures of effective recent reduction of population size could be detected at one population from Low Silesian Pinewood, which still preserve much of 335 the genetic variation. Isozyme studies of Lewandowski et al. (2002) using partially different loci have also shown that despite some deviations from allelic frequency distribution at Wêgliniec and Batorów Reserve, the populations approach the Hardy-Weinberg equilibrium in a manner similar to the large panmictic populations of P. sylvestris and P. mugo from our reference group (Hamrick et al. 1989; Frankham 1997; Shea and Furnier 2002). High genetic variability found in the taxa studied is consistent with previous studies, reporting high genetic variation in pines as compared to other out-crossing species (Goncharenko et al. 1995). The genetic structure indicates different evolutionary histories of the studied P. uliginosa populations. Analysis of population structure shows very close relationship of two P. uliginosa stands from Low Silesian Pinewood. This result suggests a common origin of the two populations, which are separated at present by a distance of a few kilometers. It cannot be excluded that the population from Wêglowiec Forestry has been planted from the seeds originating from Wêgliniec Reserve (Mr. Józefczyk – personal communication). This is possible due to its demographic structure and a similar age of the individual trees. Both populations are genetically differentiated from the locus classicus population of P. uliginosa from Batorów Reserve. Distinct character of the latter population has also been demonstrated in previous isozyme (Lewandowski et al. 2002) and biometric studies (Boratyñska et al. 2003; Marcysiak et al. 2003). These results suggest that the specificity of the population could result from its long-term isolation, different colonisation events after the last glacial period or influence of hybridisation processes. The population from Zieleniec Reserve is genetically distinct from both Batorów Reserve and populations from Low Silesian Pinewood and shows lower level of heterozygosity and genotype polymorphism as compared to the remaining populations. Inbreeding coefficient indicated significant excess of homozygotes in this population, which may reflect a small effective population size of the crossing individuals. It can also result from directional selection induced by specific microhabitat of the peat-bog environment. The excess of homozygotes in P. uliginosa from this area was noted by Siedlewska (1994) and by Siedlewska and Prus-G³owacki (1995) ascribing the phenomenon to Wahlund’s effect — hybridisation of related forms in the closest neigh- 336 W. Wachowiak, W. Prus-G³owacki bourhood. Our data support the hybridisation hypothesis as they reflect a very high genetic similarity between P. uliginosa and P. mugo trees from Zieleniec Reserve, despite their distinct phenotype. This result suggests that the trees classified morphologically as P. uliginosa represents mostly the hybrids or introgressants between both taxa. High gene flow coefficient (Nm = 83) between these two groups of trees support this suggestion. This conclusion agrees with recent controlled crosses data and molecular studies in natural populations which indicated the possibility of cross-pollination between P. uliginosa and P. mugo (Lewandowski and Dering 2006; Wachowiak and Prus-G³owacki 2008). Natural hybridisation in sympatric populations of the taxa from P. mugo complex and P. sylvestris was suggested in biometric (B¹czkiewicz 1995; Bobowicz 1990) and biochemical studies (Prus-G³owacki and Szweykowski 1979, 1983, Siedlewska 1994; Siedlewska and Prus-G³owacki 1995). The phenological observations supported the possibility of interspecific crosspollinations indicating the overlapping periods of pollen perception by P. uliginosa and pollen release by P. sylvestris (Boratyñski et al. 2003). Our two populations from Low Silesian Pinewood are surrounded by extensive forest stand of P. sylvestris and the population from Batorów Reserve is also not totally isolated from the influence of foreign pollen from the nearest Scots pine populations. However, our data indicate that P. uliginosa populations from Batorów Reserve and Low Silesian Pinewood preserved their own genetic character despite the close vicinity of P. sylvestris. This result can be explained by recent molecular studies, which demonstrated rare events of natural reciprocal cross-pollination between P. uliginosa and P. sylvestris (Wachowiak et al. 2005b; Lewandowski et al. 2005). However, the present study clearly indicates that genetic contamination through hybridisation can proceed in the sympatric populations of P. uliginosa and the taxa from P. mugo complex, in a manner similar to the studied area from Zieleniec Reserve. This finding is supported by changes in allele frequency, as no diagnostic isozyme alleles discriminating the species have been detected so far. Interestingly, this P. uliginosa stand has the best sanitary status from all the species populations included in the study. It seems likely that hybridisation may be a specific life strategy of peat-bog pine to escape extinction in changing environments, no longer suitable for the species. As concluded by Wachowiak and Prus-G³owacki (2008), the past and contemporary hybridisation within P. mugo complex could account for the variety of morphological forms observed in sympatric stand of the taxa and potentially lead to the transformation of such populations into a hybrid swarm. If this is the case, hybridisation could be the manner of moderate expansion of P. uliginosa and other taxa of P. mugo complex. A similar example is the expansion of Quercus taxa (Petit et al. 2004). In our study, P. uliginosa clusters into P. mugo complex together with P. mugo and P. uncinata. However, similar values of genetic distance to both P. mugo and P. sylvestris (DN approximately equal to 0.05), which reflects the distance between taxa, do not allow the rejection of hybridisation hypothesis. Previous biometric and biochemical study suggested that P. uliginosa may be the result of P. mugo, P. uncinata and P. sylvestris hybridisations (Lauranson-Broyer et al. 1997) or it is a stabilised hybrid taxon originating from an ancient cross pollination between P. mugo and P. sylvestris (Prus-G³owacki and Szweykowski 1983). However, lack of detected diagnostic alleles for the species so far, limits the use of isozymes in taxonomic revisions of the species. It seems that comparative analyses of orthologous regions of cpDNA and mtDNA (inherited in pines without sexual recombination in paternal and maternal line, respectively) could help to clarify the taxonomic status of P. uliginosa. So far, cpDNA studies indicated closer similarity of P. uliginosa to the P. mugo complex than P. sylvestris (Wachowiak et al. 2005a). Additionally, the study of nucleotide sequence polymorphisms of nuclear regions is a promising approach to test alternative models of speciation with or without gene flow between diverging taxa (Kliman et al. 2000). Due to gradual decline of P. uliginosa, several protective activities for the species were formulated. These include active protection of the remaining stands and creating conditions for natural regeneration, archiving the genotypes and creating the secondary stands of P. uliginosa using the plant material derived from seeds from natural populations (Danielewicz and Zieliñski 2000). Our study shows that due to hybridisation, the population from Zieleniec Reserve is a very interesting object for studies on microevolution and the speciation processes, but the seed material from this population should not be used for reintroduction of peat-bog pine in spite of very good vegetative conditions of the trees. Therefore, active protection should be concentrated on other popu- Isolated populations P. uliginosa - genetic structure lations represented by pure individuals of high genetic variability. Based on population structure analysis, remaining geographical locations of the species can be recognised as distinct conservation units. Low natural regeneration recorded at Batorów Reserve could be overcome by controlled cross pollinations to obtain the seeds for conservation programs at this area. The analyses of seedlings originating from seeds of both Low Silesian Pinewood populations showed high number of alleles and genotypes. At the same time the population of embryos exhibit some level of inbreeding also noted in other seedling populations of coniferous trees. Our result supports the previous reports, which detected low level of out-crossing in one of the P. uliginosa populations from Wêgliniec Reserve (Lewandowski et al. 2005). However, inbreeding is usually reduced in natural populations by selection against homozygotes demonstrating inbreeding depression (Kärkkäinen and Savolainen 1993). Due to the high variation in selfing rates of individual mother trees, the studies of Lewandowski et al. (2005) suggested a strategy for maximising the genetic diversity of the offspring by controlled pollinations of mother trees with high outcrossing rate and pollen donors of high genetic variability. As both Low Silesian Pinewood populations represent the same gene pool, crossing of individuals between populations could minimise genetic consequences of bottleneck detected at Wêgliniec Reserve. It seems that this approach of controlled crosses between genetically differentiated individuals should provide genetically good seed materials suitable for reintroduction. Simultaneously, there is a strong need for archiving the genotypes of individual trees of P. uliginosa from each of the defined conservation units and setting its seed orchards, as the species showed an extremely rapid decline in the last few years. All above-mentioned conservation strategies should be implemented promptly to successfully address the issue of protection of peat-bog pine from its likely extinction. Conclusions Different patterns of genetic structure shaped by divergent history of particular populations were found between isolated P. uliginosa stands. The historical factors involved in this process were: hybridisation with related pine species, bottleneck effect and stabilising selection. 337 Populations from Batorów Reserve and Low Silesian Pinewood seems to be the most suitable source of genetic material for conservation programs due to relatively high level of genetic variation and no evidence of genetic contamination due to hybridisation with other pine species. Controlled pollinations of individuals within each conservation unit are recommended to maximise the genetic diversity of the offsprings and provide suitable seed source for reintroduction. Acknowledgements. We would like to thank Prof. J. Zieliñski and Dr. W. Danielewicz for valuable comments and inspiration. Thanks are also given to Forestry Officer Mr Jan Józefczyk for his help in sampling of plant material and to Dr. Ewa Bujas and Dr. £ukasz Myczko for help in data collection. The study was supported by the State Committee for Scientific Research, Poland (Grant No. 0306/PO4/ 2001/21). REFERENCES B¹czkiewicz A, 1995. Biometrical study of some individuals chosen from Pinus mugo Turra populations in the peat bog “Bór na Czerwonem”. Acta Soc Bot Pol 64: 71–80. Bobowicz MA, 1990. Hybrids of Pinus mugo Turra × Pinus sylvestris L. from “Bór na Czerwonem” Reserve in the Nowotarska Valley (in Polish). Wyd. Naukowe UAM Poznañ. Boratyñska K, Boratyñski A, Lewandowski A, 2003. Morphology of Pinus uliginosa (Pinaceae) needles from populations exposed to and isolated from the direct influences of Pinus sylvestris. Bot J Lin Soc 142: 83–91. Boratyñski A, 1994. Protected and worth protection trees and shrubs of the Polish part of the Sudety Mountains and the Sudety Foothills. 7. Pinus mugo Turra and Pinus uliginosa Neumann. (in Polish). Arboretum Kórnickie 39: 63–85. Boratyñski A, Boratyñska K, Lewandowski A, Go³¹b Z, Kiciñski 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: 377–388. Christensen KI, 1987. Taxonomic revision of the Pinus mugo complex and P. × rhaetica (P. mugo × sylvestris) (Pinaceae). Nord J Bot 7: 383–408. Corander J, Sirén J, Arjas E, 2008. Bayesian Spatial Modelling of Genetic Population Structure. Computation Stat 23: 111–129. Cornuet JM, Luikart G, 1996. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001–2014. Danielewicz W, Zieliñski J, 2000. Protection of the Pinus uliginosa Neumann on the area of the Low 338 W. Wachowiak, W. Prus-G³owacki Silesian Pinewood. (in Polish). Przegl¹d Przyrodniczy XI 2–3: 113–124. Excoffier L, Laval G, Schneider S, 2005. Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol Bioinform Online 1: 47–50. El-Kassaby YA, 1991. Genetic variation within and among conifer populations: review and evaluation of methods. In: Fineschi S., Malvolti M. E., Cannata F., Hattemer H. H, eds. Biochemical markers in the population genetics of forest trees. Acad Pub Hague: 63–74. Frankham R, 1997. Do island populations have less genetic variation than mainland populations? Heredity 78: 311–327. Go³¹b Z, 1999. Pinus uliginosa Neumann in Wielkie Torfowiska Batorowskie (The Great Batorowskie Peatbog) in the Sto³owe Mountains. (in Polish). Szczeliniec 3: 41–48. Goncharenko GG, Silin AE, Padutov VE, 1995. Intraand interspecific genetic differentiation in closely related pines from Pinus subsection Sylvestres (Pinaceae) in the former Soviet Union. Plant Syst Evol 194: 39–54. Hamrick JL, Blanton HM, Hamrick KJ, 1989. Genetic structure of geographically marginal populations of ponderosa pine. Am J Bot 76: 1559–1568. Hedrick PW, 1974. Genetic similarity and distance: comments and comparisons. Evolution 29: 362–366. Jain SK, Workman PL, 1967. The use of generalized F-statistics in the theory of inbreeding and selection. Nature 214: 674–678. Kärkkäinen K, Savolainen O, 1993. The degree of early inbreeding depression determines the selfing rate at the seed stage: model and results from Pinus sylvestris (Scots pine). Heredity 71: 160–166. Kliman RM, Andolfatto P, Coyne JA, Depaulis F, Kreitman M, 2000 The population genetics of the origin and divergence of the Drosophila simulans complex species. Genetics 156: 1913–1931. Kormutak A, Ostrolucka M, Vookova B, Petrova A, Feckova M, 2005. Artificial hybridization of Pinus sylvestris L. and Pinus mugo Turra. Acta Biol Cracov Bot 47/1: 129–134. Lauranson-Broyer J, Krzakowa M, Lebreton PH, 1997. Reconnaissance chimio-systematique et biometrique du pin de tourbiere Pinus × uliginosa (Neumann). C R Acad Sci Paris, Sci de la vie. 320: 557–565. Lewandowski A, Boratyñski A, Mejnartowicz L, 2000. Allozyme investigations on the genetic differentiation between closely related pines – Pinus sylvestris, P. mugo, P.uncinata, and P. uliginosa (Pinaceae). Plant Syst Evol 221: 15–24. Lewandowski A, Samoæko J, Boratyñska K, Boratyñski A, 2002. Genetic differences between two Polish populations of Pinus uliginosa, compared to P. sylvestris and P. mugo. Dendrobiology 48: 51–57. Lewandowski A, Burczyk J, Wachowiak W, Boratyñski A, Prus-G³owacki W, 2005. Genetic evaluation of seeds of highly endangered Pinus uliginosa Neumann from Wêgliniec Reserve for ex-situ conservation program. Acta Soc Bot Pol 74: 237–242. Lewandowski A, Dering M, 2006. Crossability between Pinus uliginosa and its putative parental species Pinus sylvestris and Pinus mugo. Silvae Genet 55: 52–54. Luikart G, Cornuet JM, 1998. Empirical evaluation of a test for identifying recently bottlenecked populations from allele frequency data. Conserv Biol 12: 228–237. Marcysiak K, Boratyñska K, Mazur M, 2003. Variability of Pinus uliginosa cones from the peat-bog in Wêgliniec. Dendrobiology 49: 43–47. Muona O, Szmidt AE, 1985. A multilocus study of natural populations of Pinus sylvestris. Lect Notes Biomath 60: 226–240. Nei M, 1972. Genetic distance between populations. Am Nat 106: 283–292. Nei M, Roychoudhry AK, 1974. Sampling variances of heterozygosity and genetic distance. Genetics 76: 379–390. Nei M, 1987. Molecular Evolutionary Genetics. Columbia University Press, New York. Neumann C, 1837. Über eine auf den Seefeldern bei Reinerz u. einigen ähnlichen Gebirgsmooren der königl. Oberförsterei Karlsberg in der Graftschaft Glatz vorkommende noch unbeschrieben Form der Gattung Pinus. – Jahresber. Schlesische Gesellschaft für Vaterländische Kultur 11: 52–57. Petit RJ, Bodénès C, Ducousso A, Roussel G, Kremer A, 2004. Hybridization as a mechanism of invasion in oaks. New Phytol 161: 151–164. Prus-G³owacki W, Szweykowski J, 1979. Studies on antigenic differences in needle proteins of Pinus sylvestris L., P. mugo Turra, P. uliginosa Neumann and P. nigra Arnold. Acta Soc Bot Pol 48: 217–238. Prus-G³owacki W, Szweykowski J, 1983. Studies on isoenzyme variability in populations of Pinus sylvestris L., Pinus mugo Turra, Pinus uliginosa Neumann and individuals from a hybrid swarm population. B Sci Amis Poznañ 22: 107–122. Prus-G³owacki W, Bujas E, Ratyñska H, 1998. Taxonomic position of Pinus uliginosa Neumann as related to other taxa of Pinus mugo complex. Acta Soc Bot Pol 67: 269–274. Rudin D, Ekberg I, 1978. Linkage studies in Pinus sylvestris using macro-gametophyte allozymes. Silvae Genet 27: 1–2. Shea KL, Furnier GR, 2002. Genetic variation and population structure in central and isolated populations of balsam fir, Abies balsamea (Pinaceae). Am J Bot 89: 783–791. 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. Siedlewska A, Prus-G³owacki W, 1995. Genetic structure and taxonomic position of Pinus uliginosa Neumann population from Wielkie Torfowisko Batorowskie in Sto³owe Mts. (locus classicus). Acta Soc Bot Pol 64: 51–58. Isolated populations P. uliginosa - genetic structure Slatkin M, 1985. Rare alleles as indicators of gene flow. Evolution 39: 53–65. Szmidt AE, Yazdani R, 1984. Electrophoretic studies of genetic polymorphism of shikimate and 6-phosphogluconate dehydrogenases in Scots pine (Pinus sylvestris). Arboretum Kórnickie 29: 63–72. Wachowiak W, Lewandowski A, Prus-G³owacki W, 2005a. Reciprocal controlled crosses between Pinus sylvestris and P. mugo verified by a species-specific cpDNA marker. J Appl Genet 46: 41–43. Wachowiak W, Celiñski K, Prus-G³owacki W, 2005b. Evidence of natural reciprocal hybridisation between Pinus uliginosa and P. sylvestris in the sympatric population of the species. Flora 200: 563–568. 339 Wachowiak W, Prus-G³owacki W, 2008. Hybridization processes in sympatric populations of pines Pinus sylvestris L., P. mugo Turra and P. uliginosa Neumann. Plant Syst Evol 271: 21–40. Wimmer F, 1837. Über die Zapfen von Pinus silvestris, pumilio und uliginosa. Verh Bot Sect, Übersicht d. Arb. U. Veränd. der Schlesischen Ges. f. Vaterländ. Kultur, Breslau 1837–1840: 93–98. Yazdani R, Rudin D, 1982. Inheritance of fluorescence esterase and B-galatosidase, in haploid and diploid tissues of Pinus sylvestris L. Hereditas 96: 191–194. Zarzycki K, KaŸmierczakowa R, 1993. The Polish Red Data Book of Plants. (in Polish). Kraków 1993.