Contrasting genetic diversity between Planchonella obovata sensu lato (Sapotaceae) on old continental and young oceanic island populations in Japan.

Genetic diversity of plant populations on islands is likely to be influenced by characteristics such as island origin (oceanic or continental) and their age, size, and distance to continental landmasses. In Japan, Planchonella obovata sensu lato which is found on both continental and oceanic islands of varying age, size, and distance to East Asian continental areas-is an ideal system in which to investigate the factors influencing genetic diversity of island plant species. In this study, we examined the genetic diversity of P. obovata s.l. populations, in the context of the species population genetic structure, demography, and between island migration, from 668 individuals, 28 populations and 14 islands including both continental (the Yaeyama Islands) and oceanic islands (the Daito, Bonin, and Volcano Islands) using 11 microsatellite markers. The Yaeyama and Volcano Islands respectively had the highest and lowest genetic diversity, and island origin and age significantly affected genetic diversity. Clustering analysis revealed that populations were grouped into Bonin, Volcano, and Yaeyama + Daito groups. However, Bonin and Volcano groups were distinct despite the relatively short geographical distance between them. Approximate Bayesian Computation analysis suggested that the population size was stable in Bonin and Yaeyama + Daito groups, whereas population reduction occurred in Volcano group, and migration between groups were very limited. Younger oceanic islands showed lower genetic diversity, probably due to limited gene flow and a lack of time to accumulate unique alleles. Genetic structure was generally consistent with the geographic pattern of the islands, but in Volcano, a limited number of founders and limited gene flow among islands are likely to have caused the large genetic divergence observed.

Introduction

Islands have long been important systems in ecology and evolutionary biology [1,2]. Due to the small size of their landmasses, isolation from source areas, simple biotas with relatively small number of species, and high levels of plant endemism, they provide excellent opportunities to investigate the evolutionary processes of plants. Islands are also suitable for studies of population genetics, to examine phenomena such as migration rates, degree of genetic isolation, and the extent of founder effects, since they are surrounded by water restricting the movement of terrestrial organisms [3]. The genetic diversity of plant populations on islands is likely to be influenced by island characteristics such as the geological origin of the island, its age, size, and distance to the nearest continental landmass. In the past decade, ecologists have applied island biogeography theory to investigations of genetic diversity, arguing that genetic and species diversity might be influenced by similar ecological processes [4]. For example, random extinctions of species in island communities are similar to the loss of alleles due to genetic drift [5], and thus factors influencing the species diversity of islands could influence the genetic diversity of the flora and fauna on islands. The genetic diversity of plant populations is generally lower on islands than in continental areas [6-8], and is lower on young islands than on old islands [9-11], although there are some exception [12,13]. Significant relationships have been found between genetic diversity and island area [14-17], and distance to the mainland [15,17-19]. Knowledge about genetic divergence within species, and its relationship to the island origin, age, size, and distance to the continent is essential to understanding the way in which plants colonize new islands and maintain genetic diversity.

Oceanic islands are defined as those which have never been connected to a continental landmass. They are the products of volcanism or tectonic uplift, or the results of organic reef growth upon foundations formed by the first two processes. Most continental islands were joined to other continental landmasses in the past, having since become separated due to tectonics or sea level rise [20]. In Japan there are oceanic and continental islands which have common subtropical climates, facilitating the comparison of genetic diversity between the islands. The Ogasawara Islands, including the Bonin and Volcano Islands, and the Daito Islands, are oceanic islands. However, the geneses of these islands differ, with the Ogasawara Islands being of volcanic origin while the Daito Islands are a result of uplifted atolls. The Yaeyama Islands are continental islands, which have been repeatedly connected to the Asiatic continent through Taiwan. The Bonin Islands are located 1,000 km south of mainland Japan in the Northwest Pacific Ocean, and include the Mukojima, Chichijima, and Hahajima Islands (Fig 1). They developed 44–34 mya [21,22], and were gradually uplifted before the middle Pleistocene [23]. The Volcano Islands are situated 150 km south of the Bonin Islands, and appeared 0.75–0.01 mya [24]. The Daito Islands, located about 360 km east of the mainland of Okinawa and about 1,000 km west of the Bonin Islands, are comprised of three small islands, Kitadaitojima, Minamidaitojima, and Okidaitojima. They developed on the sea bed 48 mya, sank under the sea 42 mya, and were uplifted again around 6 mya [25]. The Yaeyama Islands are located 400 km west of the mainland of Okinawa and 100 km east of the Taiwan. The South Ryukyu, where the Yaeyama Islands are located, developed between the late Paleozoic and the Mesozoic [26], and emerged around the Miocene (23 mya) [27,28]. Due to clear geological differences between these island groups, the genetic diversity, divergence and demography of plants are expected to differ between them.

Fig 1

Location of the Bonin, Volcano, Yaeyama and Daito Islands.

Bold italic letters show the Population ID of the sampled populations, shown in Table 1. Numbers in the parentheses are approximate age of islands used in the lmer analyses. Maps were drawn by ArcGIS using the coast line data downloaded from the Geographical Survey Institute of Japan (https://nlftp.mlit.go.jp/ksj/gml/datalist/KsjTmplt-C23.html), under a CC BY license, with permission from the Geographical Survey Institute of Japan, original copyright 2006.

Table 1

Population genetic parameters estimated from 11 SSR for the 27 Planchonella obovata var. obovata and one P. obovata var. dubia population (CC4_dubia).

Island groups		Islands	Island area(ha)	Population ID	No. of samples	A R	H O	H E	F IS
The Bonin Islands
	The Mukojima Islands	Mukojima	256	MM	32	4.37	0.57	0.58	0.03
	The Chichijima Islands	Otoutojima	520	CO1	22	4.87	0.59	0.57	0.00
				CO2	39	5.01	0.58	0.60	0.05
		Anijima	788	CA1	44	4.68	0.57	0.59	0.05
				CA2	29	4.84	0.55	0.59	0.09
		Nishiijima	48	CN	12	4.15	0.54	0.52	0.02
		Chichijima	2344	CC1	20	4.79	0.56	0.57	0.05
				CC2	21	4.36	0.52	0.55	0.09
				CC3	17	4.50	0.50	0.53	0.09
				CC4	35	4.88	0.54	0.58	0.09
				CC4_dubia	17	4.71	0.55	0.56	0.05
				CC5	24	4.93	0.52	0.57	0.12
				CC6	20	4.74	0.56	0.57	0.04
				CC7	24	4.53	0.55	0.58	0.07
				CC8	21	4.50	0.57	0.55	0.00
	The Hahajima Islands	Hahajima	1988	HHa1	23	4.54	0.56	0.59	0.07
				HHa2	23	4.16	0.57	0.58	0.05
				HHa3	23	4.35	0.59	0.59	0.03
				HHa4	20	4.85	0.66	0.62	-0.04
				HHa5	25	4.38	0.62	0.62	0.03
		Mukoujima	138	HMu	20	4.71	0.56	0.62	0.12
		Imoutojima	123	HI	10	4.36	0.48	0.57	0.20
		Meijima	87	HMe	23	4.49	0.60	0.59	0.00
The Volcano Islands
		Minamiiwoto	354	VM	35	3.17	0.47	0.51	0.09
The Daito Islands
		Kitadaitojima	1194	DK	26	4.24	0.58	0.59	0.04
		Minamidaitojima	3057	DM	23	4.03	0.48	0.50	0.07
The Yaeyama Islands
		Ishigakijima	22250	YIs	14	5.28	0.58	0.65	0.13
		Iriomotejima	28930	YIr	21	5.69	0.72	0.72	0.03

A; allelic richness, HO; observed heterozygosity, HE; gene diversity, FIS; fixation index.

All FIS values were not significantly deviated from Hardy-Weinberg equilibrium.

Planchonella obovata sensu lato (Sapotaceae) is distributed in many parts of South-east Asia, south China, Taiwan, Micronesia, northeast Australia, and on islands of the Indian Ocean. In Japan, it occurs on the Ryukyu, Daito, Bonin, and Volcano Islands. Planchonella obovata s.l. is distributed widely from dry scrub to mesic forests on almost all of the Bonin Islands and from coastal areas to limestone areas in the mountains of the Ryukyu Islands, and is a major component of the natural vegetation. Thus, this species is ideal for the investigation of the genetic diversity, genetic structure, and population demography of plants in island areas of Japan.

The genus Planchonella includes around 110 species [29], and two species are found in Japan: P. obovata s.l. and P. boninensis (Nakai) Masam. et Yanagih. Planchonella obovata s.l. has two varieties, P. obovata (R.Br.) Pierre var. obovata and P. obovata var. dubia (Koidz. ex H.Hara) Hatus. ex T.Yamaz. Planchonella obovata var. dubia is distributed only on the Bonin and Daito Islands [30,31] where P. obovata var. obovata is also found. On these islands, Planchonella obovata var. dubia, inhabits dry rocky areas and is characterized by a lower stature and smaller leaves and fruits than P. obovata var. obovata, which grows in wetter habitats. However, the size of the leaves and fruits vary continuously between the two varieties, and are difficult to identify, and Ohashi and Kato [32] recognized P. obovata var. dubia as one of the ecotypes in dry areas.

In the Bonin Islands, adaptive radiation, probably caused by ecological divergence together with strong environmental gradients associated with the transition from mesic forests to dry scrub, has been observed in many genera such as Callicarpa [33], Crepidiastrum, Pittosporum, and Symplocos [34]. Elaeocarpus photiniifolia, endemic to the Bonin Islands, include genetically distinct groups associated with dry and mesic environments within the island [35]. Thus, the two varieties of P. obovata may be genetically distinct, although their phenotypic differences are slight.

The aim of this study was to answer following questions; (a) Are P. obovata var. obovata and P. obovata var. dubia genetically distinct? (b) What are the levels of population genetic diversity of P. obovata s.l., and which factors such as islands origin, age, area, and distance from the continent have the most impact on this diversity? (c) What is the P. obovata s.l. population structure in island areas in Japan? (d) Have the populations in each island group experienced past population size changes, such as expansion or reduction? (e) Is there migration among the island groups? The results of our study will contribute our understanding about colonization of continental plants on islands and how they maintain their genetic diversity.

Materials and methods

Study species and sample collection

Planchonella obovata s.l. is evergreen tree species, which is morphologically gynodioecious, although Kato et al. [36] have pointed out that it is functionally dioecious. Pollen is dispersed by insects, including flies, Oedemeridae beetle, and moths, which have been observed to visit flowers in the Bonin Islands [37,38]. Its berries are black, 1.2–1.5 cm long, and bear several 8–12 mm long seeds [30]. Primates, bats, lizards, and birds are thought to be important seed dispersers [39]. In the Bonin Islands, intact seeds are found in the feces of birds such as Japanese white-eye, Bonin Islands white-eye, and Brown-eared bulbul [40]. Planchonella obovata s.l. is distributed in both oceanic and continental islands, which vary in age, size, and distance to the continent, and is therefore appropriate for the investigation of the way in which genetic diversity is maintained in the islands.

We collected leaf samples of 663 individuals of P. obovata s.l. from 28 populations, 14 islands of the Bonin, Volcano, Daito, and Yaeyama Islands (Fig 1, Table 1). To ascertain whether P. obovata var. obovata and P. obovata var. dubia can be discriminated genetically, we sampled both where they occur in close proximity from the same location on Chichijima island (population CC4). These consisted of 17 individuals from dry rocky areas of typical P. obovata var. dubia, which is short in height with small leaves (CC4_dubia), and 35 individuals from wetter forest areas of typical P. obovata var. obovata, which is tall in height with large leaves (CC4). In the Daito Islands, there should be P. obovata var. dubia according to some previous reports, but we could not find typical P. obovata var. dubia there. After collection, the leaf samples were dried with silica gel. We recorded the locations of the sampled individuals using a GPS (GPSmap60CSx; Garmin, Olathe, Kansas, USA). Voucher specimens of each population were deposited in the herbarium of the Forestry and Forest Products Research Institute, Japan (nos. TF-FDA001379–TF-FDA001431) and Makino Herbarium, Tokyo Metropolitan University, Japan (nos. MAK378903, 378904, 391087).

Microsatellite analysis

Total genomic DNAs of all sampled leaves were extracted, using DNeasy Plant Mini Kits (QIAGEN, Hilden, Germany). Eighteen nuclear microsatellite markers were developed by one P. obovata var. obovata plant from Hahajima in the Bonin Islands, and details of marker development was described in S1 Appendix and S1 Table in S1 File. We genotyped all 663 samples using 18 primer pairs, with the experimental conditions described in S1 Appendix in S1 File. We tested the existence of null alleles using Microchecker [41], and the linkage disequilibrium between loci in each population using FSTAT 2.9.3.2 [42].

Five (Po124, Po200, Po281, Po579, and Po583) out of 18 markers might have null alleles since estimated null allele frequency was significant in more than five populations. Two other markers, Po290 and Po623, were characterized by large amounts of missing data. No significant linkage disequilibrium was observed between loci in any population for the 11 markers, excluding the above mentioned seven markers. Thus, we used 11 markers for further population genetic analyses. Microsatellite genotype data for 11 markers are available in S1 Data.

Data analysis

Statistical analyses of genetic diversity

To evaluate the genetic diversity of each population, allelic richness (AR) [43], observed heterozygosity (HO) and gene diversity (HE) [44] were calculated for each locus and each population using GenAlEx ver. 6.501 [45]. The fixation index (FIS) was calculated and tested by randomization using FSTAT ver. 2.9.3 [46]. We used linear mixed-effect models (lmer) to examine the associations between genetic diversity (AR, HE) within each population and island characteristics such as its island age, area, origin (oceanic or continental), and distance to mainland China using the R package lme4 [47]. The island age (1: young; Volcano, 2: middle; Bonin and Daito, 3: old; Yaeyama), area (log-transformed), origin (1: oceanic, 0: continental), and distance to nearest continent were treated as fixed effects, and the differences between loci as random effects. Island area was log-transformed to reduce skewness and increase the normality of its distribution. Variables were checked for collinearity using the variance inflation factor (VIF) value, which should be less than 5 [48], using the R package car [49]. The VIF for island origin was 5.11. We eliminated the island age from the lmer analysis, since island age and origin were highly correlated. We also conducted lmer analysis omitting the data of continental island populations, to eliminate the effect of island origin. The island age (0: young, 1: middle), area (log-transformed), and distance to nearest continent were treated as fixed effects, and the differences between loci as random effects. Data for HE were arcsine-transformed to obtain closer approximations to normality. Eight candidate models with 0–3 fixed explanatory variables were constructed for each response variable. We calculated Akaike’s information criterion to evaluate the candidate models (AIC) [50] values for each of them. The differences between the AIC values and the minimum AIC value (ΔAIC) were calculated for each of the models, and models with ΔAIC value ≤ 2 were selected as the best models [51]. The analyses were conducted using R software 4.0.2 [52]. For the lmer analysis, data from individuals sampled as P. obovata var. dubia in population CC4_dubia were merged with population CC4, since there was no genetic difference between the two populations (see details in Results). Data used in the lmer analysis are available in S2 Data.

Statistical analyses of genetic structure

For population genetic structure analysis, we firstly conducted Bayesian clustering analysis using 35 typical P. obovata var. obovata and 17 typical P. obovata var. dubia individuals from the Bonin Islands, and 49 samples from the Daito Islands, to check whether P. obovata var. dubia was genetically different from P. obovata var. obovata, using the program STRUCTURE 2.3.4 [53,54]. In this analysis, we chose allele frequency correlated model and admixture model to detect the admixture of lineages, and each run involved 100,000 Markov chain Monte Carlo (MCMC) iterations after a burn-in period of 50,000 iterations. The analysis was run 30 times with each K, ranging from 1 to 10. Then, the population genetic structure of all P. obovata s.l. populations were investigated using STRUCTURE, reducing the sample size of the Chichijima and Hahajima Islands populations to 45 randomly selected individuals each, since the program may show a bias when the sampling design is unbalanced [55]. In this analysis, each run involved 100,000 MCMC iterations after a burn-in period of 50,000 iterations. The analysis was run 30 times with each K, ranging from 1 to 15. The optimal value for K was evaluated using the ΔK [56] and the mean log likelihood at each K [54]. STRUCTURE tends to detect the highest level of a population hierarchy [56], and there can be lower hierarchy of structuring within the highest clusters. Thus, we also explored the substructure within each detected cluster. Results were summarized using the CLUMPAK [57].

The genetic relationships among populations were assessed by a neighbor-joining (NJ) tree based on the DA genetic distance [46] between them, using the program Populations 1.2.32 [58]. The significance of each node in the tree was evaluated by 1,000 bootstraps. The isolation by distance (IBD) [59] pattern was evaluated by the Mantel test [60] on population pairwise natural logarithms of geographical distance (ln (1 + geographical distance)) and FST/(1 –FST) [61].

Statistical analyses of population demography

STRUCTURE analysis detected clear genetic structure among three island groups: the Bonin Islands; the Volcano Islands; and the Yaeyama and Daito Islands (see details in Results). In order to estimate the population demography of these three island groups, we conducted approximate Bayesian computation (ABC). As there are various patterns in combinations of population demography—population size change, population divergence, and migration patterns—we sequentially executed ABC analyses [11,62]. In the first step, we applied single population size change models for each island group (Fig 2A). In the second step, we applied three-population divergence models without migration for the three island groups (Fig 2B). Finally, in the third step, we examined divergence models with and without migration (Fig 2C). The detail of ABC analysis was described in S2 Appendix in S1 File.

Fig 2

Population demographic models.

Three single population size change models (a; SNM, standard neutral model; PGM, population growth model; SRM, size reduction model). Five three-population divergence models (b; B, Bonin group; V, Volcano group; Y+D, Yaeyama and Daito group). Three migration patterns (c).

Results

Genetic difference between P. obovata var. obovata and P. obovata var. dubia

According to a NJ tree based on the DA genetic distance, the P. obovata var. obovata (CC4) and P. obovata var. dubia (CC4_dubia) sampled from the same site on Chichijima island in the Bonin Islands, formed a cluster with a high bootstrap value (Fig 3). The pairwise FST between CC4 and CC4_dubia was low as 0.012 (Fig 4). STRUCTURE analysis showed no genetic differentiation between CC4 and CC4_dubia with increasing K (S1 Fig in S1 File). In the Daito Islands, there was genetic differentiation between the two sampled populations, Kitadaitojima and Minamidaitojima, however no clear genetic sub-structuring which would imply the existence of P. obovata var. dubia was found within populations.

Fig 3

A neighbor-joining dendrogram for 27 Planchonella obovata var. obovata and one P. obovata var. dubia populations based on 11 SSR markers based on genetic distance, DA.

Numbers in the internal nodes indicate bootstrap value larger than 50; gray oval, populations in the Bonin Islands.

Fig 4

Relationship between pairwise genetic differentiation and natural logarithms of geographical distance for all the 27 P. obovata var. obovata and one P. obovata var. dubia populations.

Solid circle, population pair between P. obovata var. obovata (CC4) and P. obovata var. dubia (CC4_dubia); open circles, population pairs within the Bonin Islands; squares, population pairs between the Bonin and Volcano Islands; cross signs, population pairs between the Bonin and Daito Islands; plus signs, population pairs between the Bonin and Yaeyama Islands; triangles, population pairs between other islands; solid gray line, regression line for all 28 populations; black broken line, regression line for population pairs within the Bonin Islands; gray dotted line, regression line for population pairs whose geographical distance over 270 km (ln (1 + geographical distance > 5.6)).

Genetic diversity

At the 11 SSR loci examined, allelic richness (AR) ranged from 3.17–5.69 (mean 4.58), and gene diversity (HE) ranged from 0.50–0.72 (mean 0.58, Table 1). The AR values of each island group were significantly different (Kruskal–Wallis test, p < 0.01). The AR values of the Yaeyama Islands were significantly higher than that of the Volcano Islands (pairwise t-test, p < 0.01, Fig 5A), and those of the Yaeyama and Daito, and Bonin and Volcano Islands were marginally significant (pairwise t-test, p = 0.08, 0.07, respectively). The HE values of each island group were significantly different (Kruskal–Wallis test, p < 0.05), and the HE of the Yaeyama and Volcano islands was marginally significant (pairwise t-test, p = 0.08, Fig 5B).

Fig 5

Mean allelic richness (AR) and gene diversity (HE) in the four islands of Planchonella ovobata (± SE).

Different letters indicate significant differences among islands (p < 0.05, pairwise t-test with Bonferroni correction). Island groups are ordered from youngest (left) to oldest (right).

In the lmer analysis investigating the associations between genetic diversity and island characteristics for all populations, we eliminated island age from the analysis, since island origin and age were highly correlated. For the lmer analysis explaining AR and HE, the best model both consisted only of island origin (S2 Table in S1 File), and the coefficients of island origin were negative for both AR and HE (S3 Table in S1 File). To eliminate the effect of island origin, we also conducted lmer analysis using only oceanic islands data. For the analysis explaining AR, the best model included only island age. For the analysis explaining HE, the null model had the smallest AIC, while the Δtheilet AICsecond-best model was ≤ 2, and included island age. Coefficients of island age were positive for both AR and HE.

Genetic structure

In an NJ tree based on genetic distance, DA, P. obovata s.l. populations clustered into four distinct groups: the Bonin, Volcano, Daito, and Yaeyama Islands with high bootstrap values (Fig 3). In the Bonin Islands, populations were grouped into two: the Mukojima and Chichijima Islands, and the Hahajima Islands. The Daito Islands were closer to the Yaeyama Islands than to the Bonin Islands. The Volcano Islands population was located between the Bonin Islands and the Yaeyama and Daito Islands, with a long branch.

STRUCTURE analysis for all P. obovata s.l. populations showed that the ΔK was highest when K = 3: the Bonin Islands; the Volcano Islands; and the Yaeyama and Daito Islands were separated (Fig 6). At K = 4, the Daito and Yaeyama islands were separated. At K = 5, the Mukojima (MM) and Chihijima Islands (CO1-CC8), and the Hahajima Islands (HHa1-HMe). At K = 6, the Mukojima and Chihijima Islands were separated. At K = 7, the log likelihood reached highest value, Kitadaito (DK) and Minamidaito Islands (DM) were separated.

Fig 6

Results of STRUCTURE analysis for all Planchonella obovata s.l. populations based on 11 SSR markers.

Forty-five samples from the Chichijima and Hahajima Islands were selected at random. Changes in loglikelihood and ΔK as the number of clusters (K ranging from 1 to 15), barplots of 241 genotypes at K = 2 to 9. Vertical columns represent individuals; heights of bars are proportional to the posterior means of the estimated admixture proportions.

There were significant IBD among all populations (R2 = 0.692, p < 0.05, Fig 4) and among populations in the Bonin Islands (R2 = 0.480, p < 0.01), and regression coefficients were higher among all populations (all populations = 0.020; the Bonin Islands = 0.009). On the other hand, significant negative correlation was detected between geographical distance and pairwise FST among populations whose geographical distance over 270 km (ln (1 + geographical distance) > 5.6, r = -0.577, p < 0.001). Higher pairwise FST values were observed between the populations of the Bonin and Volcano Islands.

Population demography and divergence

The best of the three single population size change models, as selected by ABC-RF, were the standard neutral model (SNM) in the Bonin and Yaeyama + Daito groups, and the size reduction model (SRM) in the Volcano group, with relatively low error rates (0.255–0.279) and high posterior probabilities (0.757–0.791; Fig 2A, S4 Table in S1 File). Using the best models, we estimated the posterior distribution of parameters. All parameters, except for the relative ancestral effective population size (RNANC) in log10 scale, and the mean geometric parameter in the generalized stepwise mutation model (PGSM) in the SRM of the Volcano Island group, showed a clear single peak (S2 Fig in S1 File). The Bonin and Yaeyama + Daito groups showed similar level of current effective population size (NCUR) and its posterior modes (95% HPD) were 10,436 (6,517–18,411) and 12,468 (7,283–23,472), respectively, while Volcano had a smaller value of NCUR than the others, at 1,829 (401–4,561; S4 Table in S1 File). Most of the summary statistics predicted by the posterior distribution fell near the observed values, and we concluded that the goodness-of-fit values of the single population size change models to the observed data were high (S3 Fig in S1 File).

In a comparison of the five three-population divergence models, Model3 was the best model, with a posterior probability of 0.431 (Fig 2B, Table 2). However, votes of Model1 (0.265) was not low compared with those of Model3 (0.369), and the error rate was also relatively high (0.388). This high error rate was due to the high classification error rate of Model5, as shown in the confusion matrix (0.646; S5 Table in S1 File). Thus, we removed Model5 and compared the remaining four models. In this comparison, although Model3 was again selected as the best model, and the error rate was reduced (0.295), the posterior probability was still not high (0.516), and the votes of Model1 were also still not low (0.352) compared to those of Model3 (0.429) (Table 2). As the votes for the other two models were low (0.117 for Model2 and 0.102 for Model4), we removed these two models and compared only Model1 and Model3. In this comparison, Model3 was again selected as the best model, and had a high posterior probability (0.752) and a low error rate (0.184) (Table 2). Therefore, we confirmed that Model3 was the best model among the five three-population divergence models. Using the best divergence model, Model3, we compared models with and without migration using ABC-RF, and the without migration model was strongly supported, having a posterior probability of 0.954 (Fig 2C, Table 2). We thus estimated the posterior distribution of the parameters in Model3, and all parameters showed a clear single peak (S4 Fig in S1 File). The posterior mode (95% HPD) of the current effective population sizes in the Bonin and Yaeyama + Daito groups (NBYD), and the Volcano group (NV) were 15,213 (8,743–29,273) and 1,710 (690–4,537), respectively (S6 Table in S1 File). The posterior mode (95% HPD) of the time of divergence of the Volcano group from the Bonin group (T1) was 1,623 (295–5,169) generations ago. That between the Bonin and the Yaeyama + Daito groups (T2) was 8,652 (2,888–21,509) generations ago. Most of the summary statistics predicted by the posterior distribution fell near the observed values, and we concluded that the goodness-of-fit of Model3 to the observed data was high (S5 Fig in S1 File).

Table 2

Proportion of votes by random forest, posterior probability of the best model, and classification error rate in population divergence and migration analyses.

	Proportion of votes a					Posterior	Classification
Compared model set	Model1	Model2	Model3	Model4	Model5	probability	error rate
Five divergence models	0.265	0.096	0.369	0.068	0.202	0.431	0.388
Four divergence models	0.352	0.117	0.429	0.102	—	0.516	0.295
Two divergence models	0.366	—	0.634	—	—	0.752	0.184
	Model3M1	Model3M2	Model3M3	Model3
With/without migration models	0	0.004	0.018	0.978		0.954	0.216

a Best model is shown in bold.

Discussion

P. obovata var. obovata and P. obovata var. dubia

We could not differentiate P. obovata var. obovata and P. obovata var. dubia samples collected in the Bonin Islands genetically using either a NJ tree or STRUCTURE analysis. In the Daito Islands, no genetic sub-structuring which would imply the existence of P. obovata var. dubia were found within populations using the STRUCTURE analysis. Thus, we concluded that P. obovata var. dubia is one of the phenotypic variations of P. obovata var. obovata in this study. However, considering that the leaves of the individuals we sampled from the Daito Islands were larger than those of specimens identified as P. obovata var. dubia in the Daito Islands (nos. RYU4937 and RYU4955), there is a possibility we did not sample genuine P. obovata var. dubia there. More intensive search for typical P. obovata var. dubia and research combining genetic data and phenotypic data such as the sizes of leaves and fruits should be conducted in the Daito Islands in the future.

Genetic diversity of Planchonella obovata sensu lato

Allelic richness (AR) was highest in the Yaeyama Islands (mean 5.49), moderate in the Bonin and Daito Islands (means 4.60 and 4.14, respectively), and lowest in the Volcano Islands (3.17). Gene diversity (HE) showed the same pattern as AR, being highest in the Yaeyama Islands (mean 0.68), moderate in the Bonin and Daito Islands (means 0.58 and 0.55, respectively), and lowest in the Volcano Islands (0.51). This is consistent with previous studies with Maki [63] finding that allozyme diversity of plants endemic to the continental Ryukyu Islands (HT 0.134–0.321) are about five times higher than plants endemic to the oceanic Bonin Islands (HT 0.000–0.0083). In addition, using microsatellites, lower genetic diversity in the Volcano Islands compared with the Bonin Islands were also observed in Pandanus boninensis, an endemic species in the Bonin and Volcano Islands [11].

The lmer analyses suggested that island origin and age contributed most to the AR and HE levels. Island origin affected the genetic diversity, with populations in the continental islands having higher genetic diversity than oceanic ones. This is probably due to difference in number of founders with populations on oceanic-islands being established from limited seeds arriving via rare long-distance dispersal. In contrast, populations on continental-islands should have experienced greater seed influx during and post- establishment through seed dispersal by birds and other animals in the past when the continental islands were connected to continental landmasses. Island age also affected the genetic diversity, with populations on older islands having higher genetic diversity. This observation is consistent with studies in other island groups, such as the Canary Islands [9] and New Caledonia [10]. This phenomenon probably arises because unique alleles have not had enough time to accumulate via mutation and recombination on younger islands [64,65]. However, island area and distance to nearest continent did not contribute to the observed differences in genetic diversity. Although, higher genetic diversity in large populations is usually observed [66], and in the Bonin Islands, a positive correlation between genetic diversity and island area has been found in Pa. boninensis [11], in P. obovata s.l., island size may not be a good indicator of population size since its distribution is restricted to the coastal area especially in the Yaeyama Islands. Interestingly, distance to nearest continent did not impact genetic diversity probably because geographical distance used in lmer were too large to test the effect on the genetic diversity of P. obovata s.l. This is likely due to the fact that migration among island groups is extremely low, as evidenced by Model3 without migration being selected in the population divergence analysis, and, although overall isolation by distance was significant, a significant negative correlation for the population pairs whose distance over 270 km was observed, suggesting no isolation by distance pattern over this geographical distance. Given that the shortest distance to nearest continent in our data is 450 km (the Yaeyama Islands) recent migration into investigated islands from continental areas would also very limited and might have no effects on the genetic diversity.

Genetic structure of Planchonella obovata sensu lato

NJ tree and STRUCTURE analysis of all P. obovata s.l. populations identified three genetic groups: the Bonin Islands; the Volcano Islands; and the Yaeyama and Daito Islands. This finding was broadly consistent with the geographic arrangement of the islands. The Yaeyama and Daito Islands populations were adjacent in the NJ tree, and they were in the same cluster in the STRUCTURE analysis when K = 4, suggesting that the younger Daito Islands populations may have diverged from the older Ryukyu Islands where the Yaeyama Islands are located. This result is plausible, because most of the plants in the Daito Islands are shared with the Ryukyu Islands [67]. The Volcano Islands population was equally distant from the Bonin and Yaeyama + Daito Islands populations in the NJ tree, and STRUCTURE analysis could not determine which genetic groups were close to the Volcano Islands (Fig 6C, K = 2, 3). Population divergence analysis was able to unravel this question. Model3, in which the Bonin and Yaeyama + Daito groups diverged from the ancestral population, and then the Volcano group diverged from the Bonin group, was selected as the best model. The younger Volcano Islands population was inferred as having diverged from the older Bonin Islands populations. This pattern was also found in Pa. boninensis in the same area [11], and other islands, Plantago in the Hawai’ian Islands [68] and Drimys in the Juan Fernandez archipelago [69]. We only sampled P. obovata s.l. in Japan, and other candidates, such as Northern Mariana Islands, 540 km south of the Volcano Islands, were not included in our analyses. However, if the Mariana Islands is a source population of the Volcano Islands, Model1 or 2 should have been selected by ABC-RF. We should therefore include these islands to correct the estimation of the origins of the populations in the future.

The results of NJ tree, STRUCTURE analysis, and IBD analysis indicated populations in the Bonin and Volcano Islands were more distinct than may be expected given the relatively short geographical distance between them. This is probably caused by the following two factors. One is the founder effect would have occurred in the Volcano Islands since the size reduction model (SRM) was selected from population size change analysis in the Volcano Island group. This finding suggests that the number of founders which migrated from the Bonin Islands to the Volcano Islands was very limited. The other is gene flow from the Bonin Islands to the Volcano Islands was very limited as described above, and thus increased the genetic divergence between them.

In the Bonin Islands, P. obovata s.l. populations were clustered into two genetic groups: the Mukojima and Chichijima Islands, and the Hahajima Islands using NJ tree and STRUCTURE analysis with K = 5. This genetic pattern is also found in other flora and fauna dispersed by ocean currents in the Bonin Islands, such as Terminalia catappa [70], Pa. boninensis [11], Hibiscus [71], lizards [72] and land snails [73]. Fruits of P. obovata s.l. consist of a black berry and are thought to be adapted to dispersal by birds. Intact seeds of P. obovata s.l. were found in feces of Japanese white-eyes, Bonin Islands white-eyes, and Brown-eared bulbuls in the Bonin Islands [40]. Brown-eared bulbuls have high mobility, however, they are unlikely to disperse seeds between the Mukojima and Chichijima Islands, which are separated by 32 km, or the Chichijima and Hahajima Islands, which are 35 km apart, according to the retention time in the guts of the Brown-eared Bulbul, which is 30 min at most for 9.3 mm seeds [74] and their cruising speed of 29–36 km/h, calculated from body weight [75]. However, Swenson et al. [76] reported that Planchonella was dispersed as far as 8,900 km between Palau in the Pacific and the Seychelles in the Indian Ocean, based on a maximum clade credibility tree. Seeds of P. obovata s.l. are resistant to sea water [77]. Similar genetic patterns have been observed in ocean-distributed species, which have the potential to be dispersed over very long distances by water, and the salt tolerance of the seeds suggests that seeds would be dispersed by ocean currents among the Mukojima, Chichijima, and Hahajima Islands. This phenomenon whereby seeds are dispersed by vectors different from those to which they are best suited is called non-standard mechanisms of dispersal [78], and could play a role in the long distance dispersal of this species, resulting in island colonization [79].

Population demography and divergence times

In the population size change analysis, the SNM was selected in the Bonin, Yaeyama + Daito groups, and the SRM was selected in the Volcano group. These results suggest that the population size was stable in the Bonin, Yaeyama and Daito Islands, while population reduction occurred in the Volcano Islands. A similar analysis was conducted for Pa. boninensis in the Bonin and Volcano Islands, and the population growth model (PGM) was selected in the Bonin Islands, while the SRM was selected for the Volcano Islands [11]. The SRM was selected in the Volcano Islands for both species, suggesting that the effect of a founder event in the young islands is still detectable. The divergence time of the Bonin and Yaeyama + Daito groups from the ancestral population (T2) for P. obovata s.l. was 8,652 generations ago, while that of Bonin from the ancestral population for Pa. boninensis was 91,925 (S4 Table in S1 File in Setsuko et al. [11]), and P. obovata s.l. is one digit younger than Pa. boninenisis, assuming that the generation time of P. obovata s.l. and Pa. boninensis is almost the same. This finding is consistent with the fact that the Pa. boninensis is endemic to the Bonin and Volcano Islands, so a sufficient amount of time has passed for the ancestral Pandanus to have speciated into an endemic species. In contrast, the colonization of P. obovata s.l. in the Bonin Islands is likely to be relatively recent, and colonization of the islands already occupied by other plant species, may prevented the species from undergoing an increase in population size as Pa. boninenisis did, and thus the SNM would be selected in the population size change analysis. However, population demography and divergence time estimates of P. obovata s.l. in this study were derived from only limited number of SSR markers using simple ABC approaches, and further investigation using markers with higher resolution such as genome wide SNPs and other statistical approaches should be undertaken in the future.

Conclusion

We examined the genetic diversity, structure, and population demography of P. obovata s.l. on both continental (the Yaeyama Islands) and oceanic islands (the Daito, Bonin, and Volcano Islands) using 11 microsatellite markers. We could not differentiate P. obovata var. obovata and P. obovata var. dubia genetically, and concluded that P. obovata var. dubia is part of the phenotypic variation found in P. obovata var. obovata. Island origin and age had significant effects on the genetic diversity of P. obovata s.l. Genetic diversity was higher in the old continental islands (the Yaeyama Islands) and lower in the young oceanic islands (The Volcano Islands). This difference was probably caused by two reasons. One is difference in number of founders, which is greater on continental islands, and the other is difference of time to accumulate the new alleles by mutation and recombination. Genetic structure was generally consistent with the geographic pattern of the islands, but in the young oceanic islands, a limited number of founders and limited gene flow among islands is likely to have caused the large genetic divergence observed. ABC analysis revealed population size was stable in the old continental and older oceanic islands (the Bonin Islands), while population reduction occurred in the young oceanic islands, migration among the island groups were very limited, and suggested that the young oceanic islands were colonized by geographically close, older oceanic islands. Results of our study provides a good example about colonization of continental plants on islands and how they maintain their genetic diversity.

(DOCX)

Click here for additional data file.

Microsatellite genotype data for 11 markers.

(XLSX)

Click here for additional data file.

Data used in the lmer analysis.

(XLSX)

Click here for additional data file.

18 Apr 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ricardo Alia, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

“HK; 18370038; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

HK; 23310167; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

HK; 26290073; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 15K07203; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 21K05694; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 4-1402; the Environment Research and Technology Development Fund of the Ministry of the Environment, Japan

SS; 95200; the support program of FFPRI for researchers having family responsibilities”

Please state what role the funders took in the study.  If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

“This work was funded by Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science (18370038, 23310167, 26290073, 15K07203, 21K05694), the Environment Research and Technology Development Fund of the Ministry of the Environment, Japan (4-1402), and the support program of FFPRI for researchers having family responsibilities.”

We note that you have provided additional information within the Acknowledgements Section that is not currently declared in your Funding Statement. Please note that funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“HK; 18370038; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

HK; 23310167; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

HK; 26290073; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 15K07203; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 21K05694; Grants-in-Aid for Science Research from the Japanese Society for Promotion of Science

SS; 4-1402; the Environment Research and Technology Development Fund of the Ministry of the Environment, Japan

SS; 95200; the support program of FFPRI for researchers having family responsibilities”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

5. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

Additional Editor Comments:

The paper present interesting data based in a limited set of SSR to address different questions related to the diversity of continental vs island populations of Planchonella obovata. The reviewers have included some comments in order to improve the manuscript. In my opinion, one major limitation is the number of SSR used in the study, that can limit the demographic models used in the study. Therefore, I suggest to clearly discuss the limitatons of this study and to provide clear information on the validity of these models. I also suggest to reduce the number of figures and tables in the main text, and to avoid including m&M in the introduction (and figures). Please, carefully check the comments to prepare a revised version of your manuscript. Please, also improve the data availiability statement, by including a reference to a public repository or to the data as supplementary material, or other valid option.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this manuscript, the authors focus on the genetic diversity of plant populations in island environments. Through the study of a tree species, Planchonella obovata, collected on a dozen Japanese islands with contrasting origins (continental vs. oceanic) and ages, they seek to understand the colonisation modalities and to highlight factors contributing to the observed diversity of populations. The present manuscript is well written, with proper analyses and provides interesting results. I only have a reservation about the demographic analyses and estimates of divergence dates. Indeed, the genetic data are limited to 11 SSR markers and the models tested by the ABC approaches are obviously very simple. The authors should be cautious in their interpretation and mention the limitations of these analyses.

Reviewer #2: General comments: This manuscript has scientific merit and significant contribution to population genetics. The objectives are clearly defined. The Materials and Methods section has a strong scientific background. The results are well presented. The discussion shows that the authors have complete understanding on the results, on their implications on genetic diversity, with excellent use of previously published studies. The conclusions are direct related with the objectives. Thus, my recommendation is acceptance with minor revisions. The recommendation of minor revisions is based on some limitations of the current version, emphasized in the following specific comments.

1. Abstract: No materials and methods. Include.

2. Too long introduction (more than three pages including Figure 1). Note that the text in lines 86-87 is objective and the text in lines 118-119 is Materials and Methods. Remove.

3. There is no justification to change the distance to the nearest continent (specify the continent) or the geographical difference between two populations to a natural logarithm scale. Note that the correlation between the geographical difference in km and in Ln(km) is low. I computed 0.14.

4. Is there a justification to Ln-transform the island area?

5. There is no justification to arcsine transformation of He (expected heterozygosity) values. This is unacceptable.

6. It is important to specify the parameter set defined in the Structure software. How you defined the Ancestry model? How you defined the Allele Frequency model?

7. Insert the reference for the DA genetic distance (line 206): Nei et al (1983) J Mol Evol 19:153-170.

8. Figure 2 is referenced in the Materials and Methods section but the content is Result.

9. You can improve the conclusion section simply by providing objective answers to the questions raised in the objectives.

10. There are an excessive number of Figures and Tables. To keep a Table or Figure, it is important that it be informative to the readers. See my opinion below:

Figure 1: Informative; keep.

Table 1: Informative; keep. Please, provide tests for Hardy-Weinberg equilibrium using Fisher’s permutation test or chi-square with Bonferroni correction. In this way, you can assess if there is a significant FIS (inbreeding coefficient) and use this information to discuss population size.

Figure 2: Non-informative; note that you did not reference it in the results section.

Figure 3: Informative; keep.

Figure 4: There is no justification to fit a regression model transforming the geographical distance to a natural logarithm scale. This is unacceptable. Is there a justification to use Fst/(1-Fst)?

Figure 5: I can accept keeping this Figure if the AR and HE were ordered from the lowest value to the highest.

Table 2: Is there a justification to keep?

Figure 6: Informative but only the results for K = 3 should be presented. Remove also the items a) and c). There is no justification to present and, especially, to discuss the results for the other values of the suggested number of clusters. This is also unacceptable. There is no doubt that the Yaeyama and Daito’s plants represent a single population in Hardy-Weinberg equilibrium.

Table 3: Is there a justification to keep?

Table 4: Is there a justification to keep? A Table must be self-informative. None parameters are defined in this table.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Philippe LASHERMES

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

13 Jun 2022

Thank you for your careful consideration to our manuscript. Our answers to reviewers’ comments were followed by the letter “A”.

3. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: No

----------------------------------------

A. We have provided data for the SSR primers (accession numbers LC076449- LC076466) and genotypic data of SSRs (S1 Data) used in this study. However, reviewer2 pointed out that we did not provide all data, thus we also provide data used in linear mixed-effect models, examining the associations between genetic diversity and island characteristics (S2 Data), which can be calculated by SSR genotype data and island characteristics information stated in the text (lines 191-192, 730 in the manuscript, and lines 203-204, 764 in the revised manuscript with track changes).

5. Review Comments to the Author

Reviewer #1: In this manuscript, the authors focus on the genetic diversity of plant populations in island environments. Through the study of a tree species, Planchonella obovata, collected on a dozen Japanese islands with contrasting origins (continental vs. oceanic) and ages, they seek to understand the colonisation modalities and to highlight factors contributing to the observed diversity of populations. The present manuscript is well written, with proper analyses and provides interesting results. I only have a reservation about the demographic analyses and estimates of divergence dates. Indeed, the genetic data are limited to 11 SSR markers and the models tested by the ABC approaches are obviously very simple. The authors should be cautious in their interpretation and mention the limitations of these analyses.

A. Thank you for your comment. We added a sentence explaining the limitation of this analysis in lines 475-478 in the manuscript, and 500-503 in the revised manuscript with track changes, and removed description about divergence time from the conclusion (lines 521-525 in the revised manuscript with track changes).

Reviewer #2: General comments: This manuscript has scientific merit and significant contribution to population genetics. The objectives are clearly defined. The Materials and Methods section has a strong scientific background. The results are well presented. The discussion shows that the authors have complete understanding on the results, on their implications on genetic diversity, with excellent use of previously published studies. The conclusions are direct related with the objectives. Thus, my recommendation is acceptance with minor revisions. The recommendation of minor revisions is based on some limitations of the current version, emphasized in the following specific comments.

1. Abstract: No materials and methods. Include.

A. We added concrete number of samples, populations and genetic markers in the abstract (lines 26-28 in the manuscript and revised manuscript with track changes). More detailed methods would not necessary in the abstract.

2. Too long introduction (more than three pages including Figure 1). Note that the text in lines 86-87 is objective and the text in lines 118-119 is Materials and Methods. Remove.

A. We deleted the part you mentioned (lines 86-87 and 118-119 in the previous manuscript) and ecological characteristics of the species, which were not necessary in the introduction, were moved to Materials & Methods. We also changed the section title into “Study species and sample collection” from “Sample collection” (lines 121-131 in the manuscript, and lines 133-143 in the revised manuscript with track changes). The introduction length has now decreased to within three pages.

3. There is no justification to change the distance to the nearest continent (specify the continent) or the geographical difference between two populations to a natural logarithm scale. Note that the correlation between the geographical difference in km and in Ln(km) is low. I computed 0.14.

A. First of all, we have not log-transformed the distance to the nearest continent in in linear mixed-effect models. The nearest continent had been specified as mainland China in the “Statistical analyses of genetic diversity” (line 173 in the manuscript, and line 186 in the revised manuscript with track changes). Transformation of geographical distances between populations and Fst/(1-Fst) (Fig. 5) are recommended in Rousset (1997), and this method is widely adopted. Transformation was made to reduce the skewness and increase the normality of its distribution. We are not sure how obtained the correlation value of 0.14. According to our calculations the correlation coefficient between geographical distance with the closest continent and its logarithmic value was 0.98, and that between inter-population distance and its logarithmic value was 0.86.

4. Is there a justification to Ln-transform the island area?

A. Transformation of island area is no problem for the same reason as comment 3. We added the explanation in lines 176-177 in the manuscript, and line 189 in the revised manuscript with track changes.

5. There is no justification to arcsine transformation of He (expected heterozygosity) values. This is unacceptable.

A. Transformation of He is no problems for the same reason as comment 3. Especially for He, as its range is from 0 to 1 and like a proportional data, so we selected arcsine transformation. The explanation of transformation had written in lines 183-184 in the manuscript, and lines 195-196 in the revised manuscript with track changes.

6. It is important to specify the parameter set defined in the Structure software. How you defined the Ancestry model? How you defined the Allele Frequency model?

A. We selected "Allele frequency Correlated Model" and "Admixture model" to detect the admixture of lineages. We specified the selected models and reason in lines 199-200 in the manuscript, and lines 211-212 in the revised manuscript with track changes.

7. Insert the reference for the DA genetic distance (line 206): Nei et al (1983) J Mol Evol 19:153-170.

A. We inserted Nei et al. (1983) (ref. no. [46]) in line 212 in the manuscript, and line 224 in the revised manuscript with track changes.

8. Figure 2 is referenced in the Materials and Methods section but the content is Result.

A. We also referenced the Figure 2 in the Results (lines 318, 329, 342, and lines 338, 349, 362 in the revised manuscript with track changes).

9. You can improve the conclusion section simply by providing objective answers to the questions raised in the objectives.

A. We have rewritten the conclusion providing answers to the question raised in the objectives (lines 481-497 in the manuscript, and lines 506-527 in the revised manuscript with track changes).

10. There are an excessive number of Figures and Tables. To keep a Table or Figure, it is important that it be informative to the readers. See my opinion below:

Figure 1: Informative; keep.

Table 1: Informative; keep. Please, provide tests for Hardy-Weinberg equilibrium using Fisher’s permutation test or chi-square with Bonferroni correction. In this way, you can assess if there is a significant FIS (inbreeding coefficient) and use this information to discuss population size.

A. All Fis values were not significantly deviated from HWE, and describe the methods and results in lines 149, 170-171 in the manuscript, and lines 161, 181-183 in the revised manuscript with track changes.

Figure 3: Informative; keep.

Figure 4: There is no justification to fit a regression model transforming the geographical distance to a natural logarithm scale. This is unacceptable. Is there a justification to use Fst/(1-Fst)?

A. Please see answer to comment 3 above.

Figure 5: I can accept keeping this Figure if the AR and HE were ordered from the lowest value to the highest.

A. Four island groups have been ordered from west to east, the same order as Table 1, Figs 2 and 6. We are not sure whether ordering from the lowest value to the highest is really informative. As island age and origin affected the genetic diversity, we reordered then according to the island age, from youngest to oldest, and added the explanation they are ordered according to island age in line 274 in the manuscript, and line 286 in the revised manuscript with track changes.

Table 2: Is there a justification to keep?

A. Table 2 was moved to the supporting information.

Figure 6: Informative but only the results for K = 3 should be presented. Remove also the items a) and c). There is no justification to present and, especially, to discuss the results for the other values of the suggested number of clusters. This is also unacceptable. There is no doubt that the Yaeyama and Daito’s plants represent a single population in Hardy-Weinberg equilibrium.

A. We are afraid to say that we cannot agree with this comment. The �  K is not always a good measure of the best K as suggested by Wang (2017) and Funk et al (2020), and we would like to search genetic structure within the well sampled Bonin island group. Thus, we also calculated the best K based on the original method by log-likelihood (Pritchard et al. 2000) and discussed the result from K=3 to K=7.

Figure 2: Non-informative; note that you did not reference it in the results section.

Table 3: Is there a justification to keep?

A. In order to avoid the confusion of readers about which divergence pattern was the best model, Fig 2 and Table 3 were left in the main text. Moreover, we cited Fig 2 into the Result section.

Table 4: Is there a justification to keep? A Table must be self-informative. None parameters are defined in this table.

A. Table 4 was moved to the supporting information. We moved a total two tables from the main text.

References

Funk, S. M., Guedaoura, S., Juras, R., Raziq, A., Landolsi, F., Luís, C., Martínez, A. M., Musa Mayaki, A., Mujica, F., Oom, M. d. M., Ouragh, L., Stranger, Y.-M., Vega-Pla, J. L. & Cothran, E. G. (2020) Major inconsistencies of inferred population genetic structure estimated in a large set of domestic horse breeds using microsatellites. Ecology and Evolution, 10(10), 4261-4279.

Rousset (1997) Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance, DOI: 10.1093/genetics/145.4.1219

Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945–959

Wang J (2017) The computer program STRUCTURE for assigning individuals to populations: easy to use but easier to misuse. Molecular Ecology Resources 17:981–990

Submitted filename: Response_to_Reviewers_e.docx

Click here for additional data file.

17 Aug 2022

Contrasting genetic diversity between Planchonella obovata sensu lato (Sapotaceae) on old continental and young oceanic island populations in Japan

PONE-D-22-03882R1

Dear Dr. Setsuko,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Judi Hewitt

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

23 Aug 2022

PONE-D-22-03882R1

Contrasting genetic diversity between Planchonella obovata sensu lato (Sapotaceae) on old continental and young oceanic island populations in Japan

Dear Dr. Setsuko:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Judi Hewitt

Academic Editor

PLOS ONE