J For Res (2011) 16:292–301 DOI 10.1007/s10310-010-0227-4

ORIGINAL ARTICLE

Process to extinction and genetic structure of a threatened Japanese conifer species, Picea koyamae Toshio Katsuki • Ken’ichi Shimada Hiroshi Yoshimaru

•

Received: 18 October 2009 / Accepted: 7 August 2010 / Published online: 23 September 2010 Ó The Japanese Forest Society and Springer 2010

Abstract Since the end of the glacial age, Picea koyamae has been sparsely distributed in Japan as a relict species and is presently threatened with extinction. We investigated the population structure and genetic structure of nine populations of P. koyamae. Population size was assessed at 9–135 individuals in habitats ranging from 0.5 to 11.5 ha, and seedlings and saplings were observed in all but one particular population, which had a Sasa-type (bamboo grass) forest floor. The effective number of alleles per locus (Ne) within peripheral populations in the Yatsugatake Mountains was 1.8–2.7, much lower than that of core populations in the Akaishi Mountains (2.8–4.3) using five nuclear simple sequence repeat loci. This finding suggests that genetic variation in these populations has been reduced by isolation from other populations. The standardized genetic differentiation among populations (G0 ST) was 0.410 and higher than that found in other Japanese conifers, suggesting that isolation and inbreeding have progressed in this species. In two isolated populations at the Yatsugatake Mountains, the fixation index (FIS) was 0.315–0.354, much higher than the values determined for the other populations (-0.188 to 0.263). This suggests that these two populations have survived several generations while increasing the degree of inbreeding. However, the highest seedling density was in a population with low genetic variation and high FIS. The most serious problems at present appear to be the T. Katsuki (&)  K. Shimada  H. Yoshimaru Forestry and Forest Products Research Institute, 1 Matsunosato, Tsukuba, Ibaraki 305-8687, Japan e-mail: [email protected] K. Shimada e-mail: [email protected] H. Yoshimaru e-mail: [email protected]

123

declining number of mature trees and the deterioration of suitable environments for seedling establishment. Keywords Extinction  Genetic structure  Inbreeding  nSSR  Picea koyamae

Introduction Conifers are one of the most important timber resources in the world. However, 355 conifer taxa are of special concern, and 200 taxa are threatened with extinction (Farjon and Page 1999). For the conservation of these species, it is necessary to determine what processes may lead to their extinction. One extinction factor of particular concern is global warming, which may affect many species in the future. It is thus important to investigate the species that declined rapidly after the glacial age to consider the process of extinction by global warming. Picea is the most important constituent of the forests in the northern conifer biome and is widely distributed in Eurasia and North America (Farjon 1990). However, some species at the southern limit of distribution of the genus Picea, such as P. farreri, P. morrisonicola, P. chihuahuana and P. engelmanii ssp. mexicana, are listed as threatened species due to their small and relict distribution after the glacial age (Farjon and Page 1999). Picea koyamae Shiras. (yatsugatake-toˆhi in Japanese) is also one of these relict species at the southern fringe of Picea distribution. Picea koyamae is a tall conifer similar in morphology to the Norway spruce (P. abies (L.) Karst.) and classified into the section Picea subsection Picea series Picea (Schmidt 1989; Farjon 1990). A recent study using chloroplast DNA analysis reported that P. koyamae was allied to P. abies, P. asperata, P. aurantiaca, P. crassifolia,

J For Res (2011) 16:292–301

293

small populations suffer from three effects that further diminish their ability to evolve in response to environmental change: genetic variation is lost at a higher rate than that of larger populations; inbreeding is more rapid and thus reduction in fitness occurs; and there is a higher risk of accumulating deleterious mutations by chance effects (Frankham and Kingsolver 2004). Later consequences of fragmentation involve genetic effects associated with smaller population size and lower migration, which can lead to a higher rate of inbreeding (Colas et al. 2004). The deleterious effects of inbreeding can significantly impair population survival. Conifers are generally anemophilous and outcrossing species, but inbreeding depression has been reported in some species (Sorensen 1999). It is very likely that there is inbreeding depression in P. koyamae. Using allozymes, Katsuki et al. (2004) found that the genetic differentiation among four populations of P. koyamae (GST, 0.067) was higher than that of continuously distributed conifers in Japan (GST, 0.030–0.034; Tomaru et al. 1994; Uchida et al. 1997), but lower than that of discontinuously distributed conifers (GST, 0.144–0.170; Tani et al. 1996; Suyama et al. 1997). They showed that the genetic variation within three populations (Ne, effective number of alleles per locus, 1.23–1.25; He, average expected heterozygosity, 0.185–0.196) was in the range of

P. glehnii, P. koraiensis, P. obovata, P. meyeri and P. retroflexa (Sigurgeirsson and Szmidt 1993; Ran et al. 2006). Picea koyamae was first discovered in the Yatsugatake Mountains (Shirasawa and Koyama 1913), Honshu Island, Japan, and was believed to exist only there (Wilson 1916; Farjon 1990; Yamazaki 1995). Later, its presence was also confirmed in the Akaishi Mountains (Hayashi 1960; Shimizu 1992; Katsuki et al. 2004, 2008) (Fig. 1). However, it is listed as a vulnerable species in the Red Data Plant Report because of its sparse distribution (Farjon and Page 1999; Environmental Agency of Japan 2000), which covers only about 70 km (Katsuki et al. 2008). Picea koyamae is believed to have had a much wider distribution on Honshu Island during the last glacial age, but decreased during the postglacial age (Minaki 1987; Kobayashi et al. 2000). A few small and isolated populations of P. koyamae have been identified, and it is estimated that there are \1,000 living mature trees in the Yatsugatake and Akaishi Mountains combined (Katsuki et al. 2004, 2008). In view of the strong impact that genetic effective population size can have on the probability of population persistence (Newman and Pilson 1997), P. koyamae is threatened with extinction. Populations that cannot adapt to environmental change rapidly dwindle in size. Once this happens, the resultant

E138°17’ 15” 1800m

E138°21’ 00” N35°58’ 00”

1400m

N 36°00’

1500m

Nagano Pref.

Senmai-iwa

Mts.Yatsugatake

Nishi-dake

Tennyo-san

Fuki-sawa

Ohdaira Kaminigori-sawa

Todai-gawa Yabu-sawa

Mt. Nishi-dake

Karamatsu-sawa

1000m

Mt. Amigasa-dake

1km

500m

1200m

Mts Akaishi

1600m

2000m

N35°55’ 30”

Yamanashi Pref. Toyoguchi-yama 10km Tenshu-iwa

N 35°20’ E 138°00’

N40°

N Korea Japan E 138°45’

0

400km N30°

E130°

Fig. 1 Mesh area (black squares) in which Picea koyamae was confirmed (Katsuki et al. 2008), geographical position of this species’ natural range within Japan and names of localities and populations

E140°

(white circles) analyzed in this study. Nishi-dake area is shown in the map at the top right; confirmed positions of P. koyamae are shown as gray points on this map

123

294

J For Res (2011) 16:292–301

long-lived tree species (Ne, 1.32; He, 0.173; Hamrick and Godt 1990), but the genetic variation in the fourth population in the Yatsugatake Mountains (Ne, 1.16; He, 0.136) showed a rather low value. These findings suggest that this population may have experienced a bottleneck effect in the past (Katsuki et al. 2004). However, a high inbreeding coefficient was not found in that study due to a problem with the precision of analysis. In this study, therefore, we investigated the genetic variation and inbreeding in several populations of P. koyamae, using analysis of nuclear DNA markers (microsatellites), and the population structure of those populations using the density of seedlings, saplings, young trees and mature trees. From the results, we examined the process of fragmentation and isolation of those populations, paying particular attention to inbreeding depression.

Materials and methods Study site We investigated nine populations of Picea koyamae in six localities in two mountain ranges where material for DNA analysis could be collected (Table 1; Fig. 1). The Karamatsu-sawa and Fuki-sawa populations had long been known as type localities of P. koyamae (Wilson 1916), but the Tennyo-san and Senmai-iwa populations were confirmed only recently (Katsuki et al. 2008). The Fuki-sawa population is about 0.6 km from the Karamatsu-sawa population, and the Senmai-iwa population is about 1.8 km from the Fuki-sawa population (Fig. 1). These four populations in the Yatsugatake Mountains are isolated from each other and surrounded by Japanese larch (Larix kaempferi) plantation forest. These are peripheral populations considered to have been generated after a large-scale disturbance such as forest fire, harvest operation, typhoon or lava flow (Katsuki et al. 2008). It is currently estimated that only a few individual P. koyamae trees exist in the Yatsugatake Mountains area outside these four populations. Table 1 Mountains and localities of populations of Picea koyamae where samples were collected in this study

Mountains Yatsugatake

Akaishi

123

Locality

In the Akaishi Mountains, the Ohdaira population is surrounded by a larch plantation forest followed by pasture, but the other populations occur in natural forests (Katsuki and Seido 1999). The Yabu-sawa and Kaminigori-sawa populations in Todai-gawa regenerated in a rockslide area about 100 years ago. Natural Japanese larch dominates in these stands and P. koyamae and P. maximowiczii, also a threatened spruce in Japan, grow under the larch canopy (Akashi 2006). Although Todai-gawa was believed to be the largest and core locality for the remaining P. koyamae and P. maximowiczii, analysis measurements were made only around the two populations, because other parts of the valley were too steep to access. The Toyoguchi-yama population may be one of the largest populations of P. koyamae, and the Tenshu-iwa population is on the southern limit of the species’ distribution (Katsuki et al. 2005). Picea koyamae reportedly grows on limestone terraces with natural Japanese arborvitae (Thuja standishii), Japanese larch or northern Japanese hemlock (Tsuga diversifolia) (Katsuki et al. 2008). Picea koyamae occurs in areas around these populations. The populations in the Akaishi Mountains are continuous, but those in the Yatsugatake Mountains are not continuous. Population structure Population size is an extremely important factor for population structure and regeneration. Therefore, the number of mature trees and areal size of Picea koyamae for each population were surveyed. Mature tree size was considered to be larger than 20 cm DBH (diameter at breast height), because the minimum size observed for flowering individuals was about 20 cm DBH. Mature trees were confirmed by direct observation, and the position of each tree was measured by GPS. It was settled that the edge of a population is where no other individuals are seen within a range of about 50 m. From this survey, the number of confirmed individuals was considered to be the population number. The areal size of each population was calculated from the position of each individual measured by GPS.

Population

Altitude (m)

Geological type

Tennyo-san

Tennyo-san

1,750

Lava flow

Nishi-dake

Karamatsu-sawa

1,700

Lava flow

Fuki-sawa

1,850

Lava flow Lava flow

Senmai-iwa

1,700

Ohdaira

Ohdaira

1,600

Phyllite

Todai-gawa

Yabu-sawa

1,400

Sandstone

Kaminigori-sawa

1,200

Lime stone

Toyoguchi-yama

Toyoguchi-yama

1,800

Lime stone

Tenshu-iwa

Tenshu-iwa

1,750

Lime stone

J For Res (2011) 16:292–301

Tree succession is also an important factor for regeneration. The density of P. koyamae was surveyed in a quadrate (20 9 20 m) for each population. Each quadrate was set at a place where the density of mature trees was highest within the P. koyamae population. Young trees were considered to be \20 cm in DBH, but [1.3 m in height. Saplings were considered to be \1.3 m in height, but at least 10 years old. Seedlings were considered to be younger than 10 years old, because seedlings in inferior environments may die within several years after germination. The age of seedlings was assessed by measuring the number of branches. Based on these results, the relationship between number or density of mature trees and seedling density was examined using the Pearson product-moment correlation coefficient. Genetic structure For DNA analysis, we collected young branches from about 30 mature individuals from each population. Due to the small number of individuals in each population, the sample size was decided as 30 individuals in each population. However, samples were collected from young trees when there were few mature individuals. Collected samples were stored at -30°C until the DNA was extracted. DNA was extracted from 200 mg of fresh needles using a DNeasy Plant Mini Kit (Qiagen, Tokyo, Japan). Primers for 18 microsatellite markers selected from Pfeiffer et al. (1997), Rajora et al. (2001) and Hodgetts et al. (2001) were examined using samples of P. koyamae. Markers that had clear amplification fragments of DNA were selected and used for DNA analysis for P. koyamae. PCR was performed with 10-lL volumes containing 20 ng genomic DNA, 0.1 mM of each dNTP, 19 PCR buffer (200 mM Tris–HCl, 500 mM KCl), 2.0 mM MgCl2, 0.27 U Taq DNA polymerase, and 0.2 lM of each primer. The PCR thermal profile was as follows: an initial denaturing step for 5 min at 94°C followed by 35 cycles of 45 s each at 94°C (denaturation), 45 s at the annealing temperature, and 45 s at 72°C (extension) before final elongation at 72°C for 10 min in a GeneAmpÒ 9600 PCR System (Applied Biosystems, Perkin Elmer, Foster City, CA, USA). The forward sequence of each primer pair was labeled with a fluorescent dye. The genotypes were determined using an ABI 310 Genetic Analyzer and Genotyper software Ver. 2.0 (Applied Biosystems). Genetic variation within the populations was estimated according to a number of parameters, including: average number of alleles per locus (Na), effective number of alleles per locus (Ne; Kimura and Crow 1964), allelic richness (Rs), number of rare (\5%) alleles per locus (Nr), number of private alleles per locus (Np), average observed

295

heterozygosity (Ho), and average expected unbiased heterozygosity under Hardy–Weinberg equilibrium (He). Fixation index within each population (FIS; Wright 1951), which represent the deviation in genotypic proportions from the Hardy–Weinberg expectation, were calculated for polymorphic loci in each population. The bottleneck effect is also an important factor for genetic variation within a population. The probabilities for the Wilcoxon signed-rank test for recent bottlenecks under the infinite alleles model (IAM), stepwise-mutation model (SMM) and two-phase model of mutation (TPM) were calculated using polymorphic loci in each population (Piry et al. 1999). Genetic differentiation among populations was calculated according to the following genetic diversity statistics: relative extent of genetic differentiation among populations (GST; Nei and Chesser 1983), standardized genetic differentiation among populations (G0 ST; Hedrick 2005), true differentiation (D; Jost 2008), genetic differentiation index among populations (RST; Slatkin 1995), and fixation index among populations (FST). The unbiased estimates of genetic distance (DA) among all populations were determined (Nei and Chesser 1983), and dendrograms generated from neighbor-joining were analyzed using bootstrap probability based on 1,000 replicates. The correlation between genetic distance (DA) and geographical distance was examined using Mantel tests (Mantel 1967). The above calculations were computed by FSTAT 2.9.3.2 (Goudet 2001) and BOTTLENECK (Piry et al. 1999) software programs. In addition, the population substructure of P. koyamae was assessed by the Bayesian method program, STRUCTURE Ver. 2.2 (Falush et al. 2007). STRUCTURE was run 20 times independently for each K (ranging from 1 to 20) for 10,000 Markov chain Monte Carlo (MCMC) iterations after a burn-in period of 10,000 times on the total dataset without any prior information on the original population of each sampled individual. We determined the optimal number of clusters (K) from an ad hoc statistic DK based on the rate of change in the log probability of data between successive K values (Evanno et al. 2005).

Results Population structure The number of mature trees in the populations (except at Toyoguchi-yama and Tenshu-iwa, which could not be confirmed due to the steep slopes at these two population sites) ranged from 9 to 135 individuals, and the areal size of populations was 0.5–11.5 ha (Table 2). Mature tree density was about 10 individuals/ha. These findings

123

296

J For Res (2011) 16:292–301

Table 2 Stand type, population size of Picea koyamae and density of each age stage in nine populations Population

Coverage of canopy (%)

Type of forest floor

Population size of P. koyamaea

Density of P. koyamae (ha-1)b

Mature tree number

Mature tree density (ha-1)

Seedling

Sapling

Young tree

Areal size (ha)

Yatsugatake Mountains area Tennyo-san

45

Sasa

62

4.4

14

-

-

?

Karamatsu-sawa Fuki-sawa

95 95

Moss Moss

135 15

11.5 2.1

12 7

8,200 ?

? 1,075

? 2,550

Senmai-iwa

90

Moss

42

4.4

10

550

-

175

Akaishi Mountains area

a

Ohdaira

90

Moss

20

12.0

2

?

?

?

Yabu-sawa

95

Moss

17

2.5

7

50

?

75

Kaminigori-sawa

85

Moss

9

0.5

18

-

-

175

Toyoguchi-yama

75

Moss

50–80

8.3

6–11

325

?

?

Tenshu-iwa

60

Moss

30–50

1.5

20–33

325

450

150

Population size was estimated from direct observation; mature trees, DBH is over 20 cm

b

Density was estimated from a plot count (400 m2); seedling, age under about 10 years old; sapling, height under 1.3 m; young trees, DBH under 20 cm; ?, absent inside observed quadrate, but present in the population; -, absent inside and outside observed quadrate

indicate that Picea koyamae forms small populations of a few dozen individuals within a few hectares. In a survey of regeneration, seedlings or saplings were observed in all populations except Tennyo-san and Kaminigori-sawa. The density of seedlings was not significantly correlated to the density of mature trees, but was significantly correlated to the number of mature trees (r = 0.85; P = 0.003). The forest floor of all stands except for the Tennyo-san population was a moss-type, while the forest floor of the Tennyo-san population was a Sasa-type. The low density of mature trees and high density of young trees in Fuki-sawa, Yabu-sawa, Kaminigori-sawa and Tenshu-iwa populations was a different characteristic compared to the other populations, because the forest age was young. Only in the Fuki-sawa and Tenshu-iwa populations was the density of saplings higher than that of seedlings. Genetic structure Six of the 18 tested markers of nuclear simple sequence repeat (nSSR) (SpAGC1, SpAGD1, SpAGG3, UAPgAC/ AT6, UAPgAG150 and UAPgCA91) showed good amplification of DNA in P. koyamae (Table 3). A total of 281 individuals in nine populations were genotyped at these six markers, and more than 98% of the loci in all individuals were genotyped. UAPgAC/AT6 was monomorphic in all nine populations; the other five markers were polymorphic in at least one population. Indices of genetic variation in each population are shown in Table 4. The number of alleles per locus, Na, within the four populations in the Yatsugatake Mountains was 4.2–4.6, somewhat lower than the number (5.6–9.0)

123

Table 3 Microsatellite DNA loci used and the number and size of alleles detected at each microsatellite DNA locus Locus namea

Repeat unit

SpAGC1

(TC)5TT(TC)10

7

66–96

SpAGD1

(AG)25

33

125–211

SpAGG3 UAPgAC/AT6

(GA)24 (AC)10(AT)7

17 1

101–148 117

UAPgAG150

(AG)19

3

132–140

UAPgCA91

(CA)20

10

114–142

Total number of alleles

Allele size range (bp)

a SpAGC1, SpAGD1 and SpAGG3 are from Pfeiffer et al. (1997); UAPgAC/AT6, UAPgAG150 and UAPgCA91 are from Hodgetts et al. (2001)

within the five populations in the Akaishi Mountains. Except for heterozygosity, the other indices within the four populations in the Yatsugatake Mountains (Ne, 1.8–2.7; RS, 4.1–4.4; Nr, 1.4–2.8; Np, 0.0–0.4) were also lower than those in the Akaishi Mountains (Ne, 2.7–4.3; RS, 5.5–8.7; Nr, 2.8–5.0; Np, 0.2–1.0). Among the Akaishi populations, the allelic indices of Tenshu-iwa (Na, 5.6; Ne, 2.7; RS, 5.5; Nr, 2.8) were the lowest. Ho (0.269–0.585) and He (0.341–0.502) of the Yatsugatake populations did not show a clear difference with the respective values of the Akaishi populations (Ho, 0.439–0.588; He, 0.524–0.596). But Ho and He of the populations in Tennyo-san (Ho, 0.275; He, 0.393) and Karamatsu-sawa (Ho, 0.269; He, 0.341) were clearly lower than those of the other populations (Ho, 0.431–0.585; He, 0.484–0.502). Ne of the populations in Tennyo-san (1.8) and Karamatsu-sawa (1.8) were also relatively low.

J For Res (2011) 16:292–301

297

Table 4 Number of alleles, heterozygosity, fixation indices and probability of recent bottleneck effects within nine populations of Picea koyamae Population

Sample number

Na

Ne

RS

Nr

Np

Ho

He

FIS

IAM

TPM

SMM

Yatsugatake Mountains area Tennyo-san

32

4.6

1.8

4.4

2.8

0.0

0.275

0.393

0.354*

0.563

0.906

0.938

Karamatsu-sawa

32

4.4

1.8

4.2

2.6

0.4

0.269

0.341

0.315*

1.000

1.000

1.000

Fuki-sawa

31

4.2

2.7

4.1

1.4

0.0

0.585

0.484

-0.188*

0.063

0.438

0.563

Senmai-iwa

32

4.4

2.4

4.3

1.4

0.2

0.431

0.502

0.117*

0.063

0.438

0.906

Akaishi Mountains area Ohdaira Yabusawa

32 32

9.0 6.4

4.3 2.8

8.7 6.2

5.0 3.4

1.0 0.4

0.588 0.550

0.596 0.551

0.038 0.048

0.438 0.563

0.844 0.906

0.969 0.969

Kaminigori-sawa

28

7.8

4.1

7.8

2.8

0.2

0.443

0.573

0.263*

0.438

0.906

0.969

Toyoguchi-yama

31

6.6

3.7

6.4

3.0

0.8

0.439

0.524

0.181*

0.031

0.156

0.969

Tenshu-iwa

31

5.6

2.7

5.5

2.8

0.8

0.560

0.561

0.156

0.438

0.906

-0.002

Na number of alleles per locus, Ne effective number of alleles per locus, RS allelic richness (based on 28 samples), Nr number of rare alleles (\5%) per locus, Np number of private alleles per locus, Ho average observed heterozygosity, He average expected heterozygosity, FIS fixation indices, IAM provability for Wilcoxon signed-rank test for recent bottlenecks under the infinite allele model, SMM provability for Wilcoxon signed-rank test for recent bottlenecks under the stepwise mutation model, TPM provability for Wilcoxon signed-rank test for recent bottlenecks under the two-phase model of mutation * Significance level (P \ 0.05) and refers to a deviation from zero

Fixation indices (FIS) within the nine populations were estimated using four polymorphic loci, but not UAPgAG150, which was monomorphic in seven populations. FIS values of five populations were significantly (P [ 0.05) positive, especially for Tennyo-san (0.354) and Karamatsusawa (0.315) in the Yatsugatake Mountains and the populations in Kaminigori-sawa (0.263) and Toyoguchi-yama (0.181) in the Akaishi Mountains. This indicates that inbreeding or self-fertilization have occurred in those populations. Bottleneck analysis under the IAM, the SMM and the TPM indicated significant heterozygosity excess for only the Toyoguchi-yama population in the Akaishi Mountains under IAM (Table 4). For the other populations, the Wilcoxon signed-rank test under the three models indicated no greater heterozygosity than that expected for populations at mutation-drift equilibrium. Levels of genetic differentiation among the nine populations of P. koyamae were calculated from five polymorphic loci. The relative extent of genetic differentiation among populations (GST) was 0.188, standardized genetic differentiation among populations (G0 ST) was 0.410, true differentiation (D) was 0.274, genetic differentiation index among populations (RST) was 0.173, and fixation index among populations (FST) was 0.209. Genetic distance among the populations was between 0.058 (Ohdaira vs Kaminigori-sawa) and 0.510 (Tennyosan vs Senmai-iwa) (Table 5). The relationship between geographical distance and genetic distance among populations derived using the Mantel test did not show a large positive correlation (r = 0.234; P = 0.172).

Tennyo-san Fuki-sawa Toyoguchi-yama 908

Tenshu-iwa 880

Ohdaira Kaminigori-sawa

564

Senmai-iwa

Yabu-sawa 0.1

Karamatsu-sawa

Fig. 2 Dendrogram for nine populations of Picea koyamae from neighbor-joining analysis based on pairwise DA distance. The figures in the phylogenetic tree represent bootstrap probability based on 1,000 replicates, and only the values [500 are shown

The dendrogram from the neighbor-joining analysis based on pairwise DA distances is shown in Fig. 2. The bootstrap probabilities suggest combining the Tennyo-san and Fuki-sawa populations, and the Ohdaira and Kaminigori-sawa populations. We determined K = 2 as the optimal number of clusters for use with STRUCTURE. The value of DK at K = 2 was the highest, and the one at K = 8 and 9 were also high. As the result, it was shown that all individuals were firstly sorted by three populations at the Yatsugatake Mountains and six other populations including another population at the Yatsugatake Mountains (Fig. 3). Furthermore, those two groups were calculated within each populations, and we determined K = 3 for three populations at Yatsugatake, and K = 6 for the other six populations. Three populations

123

298

J For Res (2011) 16:292–301

Table 5 DA distance (Nei and Chesser 1983) for all pairs from each of the nine populations of Picea koyamae Population

TN

KR

FK

SM

OD

YB

KM

TY

Tennyo-san (TN)

–

Karamatsu-sawa (KR)

0.502

–

Fuki-sawa (FK)

0.296

0.474

–

Senmai-iwa (SM)

0.510

0.333

0.295

–

Ohdaira (OD)

0.337

0.346

0.258

0.177

Yabusawa (YB)

0.413

0.303

0.396

0.270

0.146

–

Kaminigori-sawa (KM)

0.381

0.373

0.303

0.218

0.058

0.142

–

Toyoguchi-yama (TY)

0.362

0.470

0.410

0.266

0.219

0.274

0.238

–

Tenshu-iwa (TS)

0.383

0.469

0.411

0.286

0.222

0.312

0.227

0.301

–

(Tennyo-san, Karamatsu-sawa, Fuki-sawa) in the Yatsugatake Mountains consist of completely different clusters from the other populations. The Senmani-iwa population in the Yatsugatake Mountains might have some relationship with those in the Akaishi Mountains and contain two kinds of clusters. The Tenshu-iwa population and part of the Toyoguchi-yama population had nearly different clusters from the others in the Akaishi Mountains. Individuals from Ohdaira to part of Toyoguchi-yama consisted of multiple clusters.

Discussion The processes of fragmentation and isolation The number (9–135) and areal size (0.5–12.0 ha) of the populations of Picea koyamae confirmed in this study (Table 2) are considered to be much smaller compared to other spruces occurring in boreal forests. Based on observations of large contiguous stands of P. glauca, O’Connell et al. (2006) estimated that a threshold population size of 180 trees is needed to reduce the negative effects of pollen limitation and inbreeding and maintain seed yields. Picea jezoensis var. hondoensis occurring in the same region as P. koyamae (Central Honshu, Japan), also often forms large populations of over 1,000 individuals. However, the small population size appears to be unexceptional for P. koyamae, because there were small populations in natural habitats. The population size of P. koyamae likely depends on the scale of disturbances that occurred in the past such as forest fire, rockslide, slope failure, typhoon, forest harvest, etc. Such disturbances in relatively flat regions may occur on a scale as large as over a dozen hectares. However, it is believed that disturbances in the present habitat of P. koyamae occurred on a small scale (under a dozen hectares) due to the region being mountainous.

123

The nSSR-derived value (FST, 0.209; RST, 0.173; G0 ST, 0.410; D, 0.274) of genetic differentiation among populations in P. koyamae was high compared with nSSR-derived values of other Japanese conifers, such as Cryptomeria japonica (FST, 0.028; RST, 0.032; G0 ST, 0.125; Takahashi et al. 2005), Chamaecyparis obtusa (FST, 0.039; GST, 0.040; G0 ST, 0.188; Matsumoto et al. 2010), Picea alcoquiana (FST, 0.071; G0 ST, 0.164; Aizawa et al. 2008) and Picea jezoensis (FST, 0.101; Aizawa et al. 2009). Here, the G0 ST values could be calculated from the results of the above reference papers, but D could not be calculated because of the lack of HT value in their papers. The comparison of G0 ST values indicates that differentiation among populations of P. koyamae is substantial even though its geographic distribution area is small. It was suggested that the genetic variation of P. koyamae is maintained by small but differentiated populations. The analyses of genetic distance (Table 5; Fig. 2) and STRUCTURE (Fig. 3) indicated that genetic differentiation among the three populations in the core area (Ohdaira, Yabu-sawa, Kaminigori-sawa) was smaller than that of the six populations in the peripheral area. In addition, it was found that the genetic variation within populations was higher in the core area than in the peripheral area (Table 4). Picea koyamae is a wind pollination species, and a sufficient quantity of pollen must be scattered for pollination. These results suggest that if there is sufficient gene flow by pollen at the high-density core area, genetic variation within a small population is maintained, which is probably the case for Kaminigori-sawa. On the other hand, there was little gene flow at the lowdensity peripheral area, where the population hardly exchanges genes with other populations. It was suggested that the genetic distance between Karamatsu-sawa and Fuki-sawa was relatively long compared to that between the other populations, even though they are only 0.6 km from each other, in view of the small gene flow. It appears that there was little gene flow among the Yatsugatake populations. These results suggest that fragmentation and isolation progress in the peripheral populations of P. koyamae. The FIS value (0.101) of P. koyamae indicated that this species was basically an outcrossing species like other Picea spp., but there was also an influence of inbreeding. Low FIS values have been reported for other Picea spp., such as P. alcoquiana (0.018; Aizawa et al. 2008), P. asperata (0.005; Luo et al. 2005), P. breweriana (0.003; Ledig et al. 2005), P. mariana (-0.025; Gamache et al. 2003; 0.02; Perry and Bousquet 2001) and P. rubens (0.021; Rajora et al. 2000). In addition, significantly positive FIS values have been reported for P. chihuahuana (0.185; Ledig et al. 1997), P. glauca (0.226; Rajora et al. 2005) and P. sitchensis in peripheral areas (0.17; Gapare et al. 2005). A clear relationship is not shown between the

J For Res (2011) 16:292–301

299

Fig. 3 Summary plot of estimates of Q for Picea koyamae. Each individual is represented by a single vertical line broken into K segments, with lengths proportional to each of the K inferred clusters. a Calculated for all populations; b calculated for three populations; c calculated for six populations

(a)

K=2

1.0 0.8 0.6 0.4 0.2 0.0

1.0

(b) K = 3

(c) K = 6

0.8 0.6 0.4 0.2 0.0 Tennyo-san

Fuki-sawa

Karamatsu-sawa

FIS value and the distribution characteristic of the species. For example, the FIS value of P. breweriana, an endangered species, was low. On the other hand, the FIS value of P. glauca was significantly positive. Inbreeding may occur in small and large populations. Both high and low FIS values in populations in the Akaishi Mountains (Table 4) indicate that any population of P. koyamae that was derived from a small number of related individuals in core or peripheral areas might increase the degree of inbreeding. It was found that small population size and isolation led to the low values of genetic variation and high values of FIS in the Tennyo-san and Karamatsu-sawa populations. Although the analysis of BOTTLENECK indicated that these populations have not experienced a recent reduction in effective population size, it does not deny that they experienced bottlenecks in the far past. Some populations of Picea have shown high fixation indices (Ledig et al. 1997; Gapare et al. 2005; Rajora et al. 2005) and the existence of selfing (Innes and Ringius 1990; Ledig et al. 2000; Flores-Lope´z et al. 2005). These results suggest that over several generations, small and isolated populations of P. koyamae have survived with little gene flow, and have sustained themselves while increasing the degree of inbreeding. Disincentives for regeneration The present populations of P. koyamae are likely to have become isolated and small through exploitation and the

Ohdaira Senmai-iwa

Kaminigori-sawa

Yabusawa

Tenshu-iwa

Toyoguchi-yama

plantation of larch during the past few decades. The high density of seedlings at Karamatsu-sawa (Table 2) indicates that the number and density of mature trees are more important than inbreeding depression for regeneration. Some possible reasons for the absence of seedlings at Tennyo-san are the effect of Sasa bamboo grass covering the forest floor and damage by seed chalcids (T. Katsuki et al., unpublished data), while at Kaminigori-sawa it might be due to the density of mature trees. However, it is also important to note that the decline of genetic variation may cause degeneration through inbreeding. Mosseler et al. (2003) found a strong relationship between height growth and measure of genetic diversity based on allozyme analyses of P. rubens. O’Connell et al. (2006) indicated that the proportion of empty seeds was high in small populations of P. glauca. A population with low genetic variation and high inbreeding degree is considered to be facing a crisis situation. Conserving genetic diversity in peripheral populations may require larger reserves for in situ conservation than what is required in core populations (Gapare and Aitken 2005). It appears that some special measures must be taken, such as promoting natural regeneration, or planting saplings. Acknowledgments We gratefully acknowledge Prof. C. Campbell (University of Maine) for valuable input; K. Kitamura, Y. Tsumura, K. Yoshimura (Forestry and Forest Products Research Institute) and T. Sugaya (Tokyo University of Agriculture) for their help in collecting and examining samples; H. Nishikawa, T. Nagaike (Yamanashi Forest Research Institute), S. Tanaka (The Botanical Society of Yamanashi), K. Akashi (Iida City Museum), and K. Bessho and

123

300 M. Ohnaka (Tama Forest Science Garden, Forestry and Forest Products Research Institute) for their help in collecting and field surveying; and the Chubu Regional Forest Office, Yamanashi Prefecture and Ohshika Village for their permission to conduct the study. This study was supported by funds from the Japanese Ministry of the Environment.

References Aizawa M, Yoshimaru H, Katsuki T, Kaji M (2008) Imprint of postglacial history of a narrow endemic spruce Picea alcoquiana observed in nuclear microsatellite and organelle DNA markers in central Japan. J Biogeogr 35:1295–1307 Aizawa M, Yoshimaru H, Sato H, Katsuki T, Kawahara T, Kitamura K, Shi F, Sabirov R, Kaji M (2009) Range-wide genetic structure in a north-east Asian spruce (Picea jezoensis) determined using nuclear microsatellite markers. J Biogeogr 36:996–1007 Akashi K (2006) Microtopography, debris flow deposits and forest vegetation on the floodplains in the upper stream of the Todai River, northwestern Akaishi Mountains. Nat Hist Rep Inadani 7:33–78 (in Japanese) Colas B, Chris D, Thomasm D, Hanski I (2004) Adaptive responses to landscape disturbances: empirical evidence. In: Ferrie`re R, Dieckmann U, Couvet D (eds) Evolutionary conservations biology. Cambridge University Press, Cambridge, pp 284–304 Environmental Agency of Japan (2000) Threatened wildlife of Japan. Red data book 2nd edn. vol 8, Vascular plants, Japan Wildlife Research Center, Tokyo (in Japanese) Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software structure: a simulation study. Mol Ecol 14:2611–2620 Falush D, Stephens M, Pritchard JK (2007) Inference of population structure using multilocus genotype data: dominant markers and null alleles. Mol Ecol Notes 7:574–578 Farjon A (1990) Pinaceae. Koeltz, Konigstein Farjon A, Page CN (eds) (1999) Conifers. Status survey and conservation action plan. IUCN, Gland Flores-Lope´z C, Lope´z-Upton J, Vargas-Herna´ndez JJ (2005) Reproductive indicators in natural populations of Picea mexicana Martı´nez. Agrociencia 39:117–126 Frankham R, Kingsolver J (2004) Responses to environmental change: adaptation or extinction. In: Ferrie`re R, Dieckmann U, Couvet D (eds) Evolutionary conservation biology. Cambridge University Press, Cambridge, pp 85–100 Gamache I, Jaramillo-Correa JP, Payette S, Bousquet J (2003) Diverging patterns of mitochondrial and nuclear DNA diversity in subarctic black spruce: imprint of a founder effect associated with postglacial colonization. Mol Ecol 12:891–901 Gapare WJ, Aitken SN (2005) Strong spatial genetic structure in peripheral but not core populations of Sitka spruce [Picea sitchensis (Bong.) Carr.]. Mol Ecol 14:2659–2667 Gapare WJ, Aitken SN, Ritland CE (2005) Genetic diversity of core and peripheral Sitka spruce (Picea sitchensis (Bong.) Carr) populations: implications for conservation of widespread species. Biol Conserv 123:113–123 Goudet J (2001) FSTAT, a program to estimate and test gene diversities and fixation indices, version 2.9.3. http://www2. unil.ch/popgen/softwares/fstat.htm Hamrick JL, Godt MJW (1990) Allozyme diversity in plant species. In: Brown AHD, Clegg MT, Kahler AL, Weir BS (eds) Plant population genetics, breeding and genetic resources. Sinauer, Sunderland, pp 43–63

123

J For Res (2011) 16:292–301 Hayashi Y (1960) Notes on Japanese trees and shrubs (4). Bull Gov For Exp Stn 125:71–74 Hedrick PW (2005) A standardized genetic differentiation measure. Evolution 59:1633–1638 Hodgetts RB, Aleksiuk MA, Brown A, Clarke C, Macdonald E, Nadeem S, Khasa D (2001) Development of microsatellite markers for white spruce (Picea glauca) and related species. Theor Appl Genet 102:1252–1258 Innes DJ, Ringius GG (1990) Mating system and genetic structure of two populations of white spruce (Picea glauca) in eastern Newfoundland. Can J Bot 68:1661–1666 Jost L (2008) GST and its relatives do not measure differentiation. Mol Ecol 17:4015–4026 Katsuki T, Seido S (1999) Population of Picea koyamae and P. maximowiczii in Hakushu-cho, Yamanashi Prefecture. Trans Kanto Jpn For Soc 50:69–70 (in Japanese) Katsuki T, Sugaya T, Kitamura K, Takeuchi T, Katsuta M, Yoshimaru H (2004) Geographic distribution and genetic variation of a vulnerable conifer species, Picea koyamae (Pinaceae). Acta Phytotaxon Geobot 55:19–28 Katsuki T, Aizawa M, Akashi K, Shimada Ke, Shimada Ka (2005) Present conditions of Picea koyamae and P. maximowiczii in Ohshika Village, Nagano Prefecture. Two threatened plant species at the southern limit of distribution. Nat Hist Rep Inadani 6:71–76 (in Japanese) Katsuki T, Akashi K, Tanaka S, Iwamoto K, Tanaka N (2008) An estimation of the present distribution of Picea koyamae and P. maximowiczii using climatic and geological factors. Jpn J For Environ 50:25–34 (in Japanese) Kimura M, Crow JF (1964) The number of alleles that can be maintained in a finite population. Genetics 49:725–738 Kobayashi K, Yoshikawa J, Suzuki M (2000) DNA identification of Picea species of the last glacial age in northern Japan. Jpn J Hist Bot 8:67–80 Ledig FT, Jacob-Cervantes V, Hodgskiss PD, Eguiluz-Piedra T (1997) Recent evolution and divergence among populations of a rare Mexican endemic, Chihuahua spruce, following Holocene climatic warming. Evolution 51:1815–1827 Ledig FT, Bermejo-Vela´zquez B, Hodgskiss PD, Johnson DR, FloresLo´pez C, Jacob-Cervantes V (2000) The mating system and genic diversity in Martı´nez spruce, an extremely rare endemic of Me´xico’s Sierra Madre Oriental: an example of facultative selfing and survival in interglacial refugia. Can J For Res 30:1156–1164 Ledig FT, Hodgskiss PD, Johnson DR (2005) Genic diversity, genetic structure, and mating system of Brewer spruce (Pinaceae), a relict of the Arcto-Tertiary forest. Am J Bot 92:1975–1986 Luo J, Wang Y, Korpelainen H, Li C (2005) Allozyme variation in natural populations of Picea asperata. Silva Fenn 39:167–176 Mantel N (1967) The detection of disease clustering and a generalized regression approach. Cancer Res 27:209–220 Matsumoto A, Uchida K, Taguchi Y, Tani N, Tsumura Y (2010) Genetic diversity and structure of natural fragmented Chamaecyparis obtusa populations as revealed by microsatellite markers. J Plant Res. doi:10.1007/s10265-009-0299-4 Minaki M (1987) Fossil plants of the last glacial age and their evolutional significance. Iden 41(12):30–35 (in Japanese) Mosseler A, Major JE, Rajora OP (2003) Old-growth red spruce forests as reservoirs of genetic diversity and reproductive fitness. Theor Appl Genet 106:931–937 Nei M, Chesser RK (1983) Estimation of fixation indices and gene diversities. Ann Hum Genet 47:253–259 Newman D, Pilson D (1997) Increased probability of extinction due to decreased genetic effective population size: experimental populations of Clarkia pulchella. Evolution 51:354–362

J For Res (2011) 16:292–301 O’Connell LM, Mosseler A, Rajora OP (2006) Impacts of forest fragmentation on the reproductive success of white spruce (Picea glauca). Can J Bot 84:956–965 Perry DJ, Bousquet J (2001) Genetic diversity and mating system of post-fire and post-harvest black spruce: an investigation using codominant sequence-tagged-site (STS) markers. Can J For Res 31:32–40 Pfeiffer A, Olivieri AM, Morgante M (1997) Identification and characterization of microsatellites in Norway spruce (Picea abies). Genome 40:411–419 Piry S, Luikart G, Cornuet J-M (1999) BOTTLENECK: a computer program for detecting recent reductions in the effective population size using allele frequency data. J Hered 90:502–503 Rajora OP, Mosseler A, Major J (2000) Indicators of population viability in red spruce, Picea rubens. II. Genetic diversity, population structure, and mating behavior. Can J Bot 78:941– 956 Rajora OP, Rahman MH, Dayanandan S, Mosseler A (2001) Isolation, characterization, inheritance and linkage of microsatellite DNA markers in white spruce (Picea glauca) and their usefulness in other spruce species. Mol Gen Genet 264:871–882 Rajora OP, Mann IK, Shi Y-Z (2005) Genetic diversity and population structure of boreal white spruce (Picea glauca) in pristine conifer-dominated and mixedwood forest stands. Can J Bot 83:1096–1105 Ran JH, Wei XX, Wang XQ (2006) Molecular phylogeny and biogeography of Picea (Pinaceae): implications for phylogeographical studies using cytoplasmic haplotypes. Mol Phylogenet Evol 41:405–419 Schmidt PA (1989) Beitrag zur Systematik und Evolution der Gattung Picea A, Dietr. Flora 182:435–461 Shimizu T (1992) Taxonomy and phytogeography of conifers, with special reference to some subalpine genera (2). The genus Picea. Jpn J Hist Bot 9:3–11 (in Japanese)

301 Shirasawa Y, Koyama M (1913) Some new species of Picea and Abies in Japan. Bot Mag Tokyo 27:127–131 (in Japanese) Sigurgeirsson A, Szmidt AE (1993) Phylogenetic and biogeographic implications of chloroplast DNA variation in Picea. Nord J Bot 13:233–246 Slatkin M (1995) A measure of population subdivision based on microsatellite allele frequency. Genetics 139:457–462 Sorensen FC (1999) Relationship between self-fertility, allocation of growth, and inbreeding depression in three coniferous species. Evolution 53:417–425 Suyama Y, Tsumura Y, Ohba K (1997) A cline of allozyme variation in Abies mariesii. J Plant Res 110:219–226 Takahashi T, Tani N, Taira H, Tsumura Y (2005) Microsatellite markers reveal high variation in natural populations of Cryptomeria japonica near refugial areas of the last glacial period. J Plant Res 118:83–90 Tani N, Tomaru N, Araki M, Ohba K (1996) Genetic diversity and differentiation in populations of Japanese stone pine (Pinus pumila) in Japan. Can J For Res 26:1454–1462 Tomaru N, Tsumura Y, Ohba K (1994) Genetic variation and population differentiation in natural populations of Cryptomeria japonica. Plant Species Biol 9:191–199 Uchida K, Tomaru N, Tomaru C, Yamamoto C, Ohba K (1997) Allozyme variation in natural population of hinoki, Chamaecyparis obtusa (Sieb. et Zucc.) Endl, and its comparison with the plus-tree selected from artificial stands. Breed Sci 47:7–14 Wilson EH (1916) The conifers and taxads of Japan. Cambridge University Press, Cambridge Wright S (1951) The genetical structure of population. Ann Eugen 15:323–354 Yamazaki T (1995) Pinaceae. In: Iwatsuki K, Yamazaki T, Boufford DE, Ohba H (eds) Flora of Japan. Pteridophyta and Gymnospermae, vol 1. Kodansha, Tokyo, pp 266–277

123