Genome-wide markers reveal geographical isolation by distance and barriers as well as local heterogeneity in the genetic structure of a seagrass

doi:10.21203/rs.3.rs-4714480/v1

Download PDF

Research Article

Genome-wide markers reveal geographical isolation by distance and barriers as well as local heterogeneity in the genetic structure of a seagrass

https://doi.org/10.21203/rs.3.rs-4714480/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Gene flow is a crucial concept in the delineation of conservation units for natural populations of a species. Seagrasses are marine species targeted for conservation because their abundance has declined worldwide during the last century. However, we cannot determine how to delineate conservation units with inadequate knowledge of the genetic structure of seagrasses. This study explored the genetic structure of Zostera marina L. (eelgrass) populations in three semi-enclosed areas using single nucleotide polymorphisms within abundant, genome-wide loci. Genome-wide markers revealed that the genetic structure was isolated by geographical distance and barriers through a narrow strait in an area with linear dimensions less than 200 km. The genetic distance created by the barrier was 6.7 times the genetic distance due to 100 km of geographic distance. The markers revealed the intra-site variability in genetic structure and the heterogeneity among sites on scales less than ~10 km that had not been recognized previously. Our results imply that the use of genomic tools will focus seagrass conservation efforts more locally than before and that assessing relative genetic differences can make delineating conservation units a reality. Identifying the evolutionary and quantitative meaning of genetic differences will be a next challenge for delineating seagrass conservation units.

Conservation unit

Eelgrass

Gene flow

Genetic distance

Biological conservation units are delineated to suppress anthropogenic hybridization and admixture within a natural population of a species (Allendorf et al. 2001). Gene flow, which theoretically homogenizes the genetic structure among populations (Lowe and Allendorf 2010; Ellstrand 2014), is a crucial process that conserves genetic diversity in natural areas (Allendorf et al. 2010; Funk et al. 2012; Hohenlohe et al. 2021). Seagrasses are foundation species that form a habitat for many marine organisms and provide important ecosystem services (Costanza et al. 1997). Because the abundance of seagrasses has been declining worldwide during the last century (Orth et al. 2006; Waycott et al. 2009), there has been a global effort to conserve seagrass beds and recover their ecosystem services (van Katwijk et al. 2016). An understanding of gene flow is necessary to regulate the admixture of previously separated seagrass populations (Procaccini et al. 2007; Kendrick et al. 2017).

The dispersal of pollen and seeds contributes to the flow of genes among seagrass populations but is considered to contribute less to gene flow across long than short distances because pollen and seeds travel only short distances (Hosokawa et al. 2015; Kendrick et al. 2017). However, seeds have the potential to be buoyant if the buoyant tissues in the spathe and shoot buoy the seeds, which have a specific gravity greater than that of seawater (Hosokawa et al. 2015). The seeds can be transported more than 100 km by currents while they are buoyant if the currents are sufficiently swift and the conditions that cause them to remain positively buoyant persist (Orth et al. 1994; Kendrick et al. 2012; Kendrick et al. 2017).

The present genetic structure of seagrasses is the result of the history of gene flow among populations through dispersals and the demographic dynamics of populations (Kendrick et al. 2017). The limitation of dispersal distance can strongly influence the genetic structure of marine species. Examples include isolation by distance (IBD) of seagrass species (Coyer et al. 2004; Olsen et al. 2004) and isolation by resistance (IBR) of marine species due to physical barriers associated with seascape topography (e.g. D’Aloia et al. 2014; Thomas et al. 2015). However, dispersal is not the only process that determines the genetic structure of marine species. Environmental heterogeneity also contributes to genetic structure because organisms adapt to their environment (Oetjen et al. 2010; Nanninga et al. 2014; Sandoval-Castillo et al. 2018; Nguyen et al. 2023). The global-scale IBDs of seagrasses have been explored in the context of their evolutionary history (Coyer et al. 2004; Olsen et al. 2004; Talbot et al. 2016; Yu et al. 2023), but IBD patterns can also be seen at spatial scales of 100–1000 km (Tanaka et al. 2002; Arnaud-Haond et al. 2007; Nakajima et al. 2014; Kim et al. 2017; Stafford-Bell et al. 2019) and less than 100 km (Tanaka et al. 2011; Sinclair et al. 2014; Hori and Sato 2021; Martínez-García et al. 2021). In fact, the genetic structure formed by environmental heterogeneity occurs within a 10-km scale. Examples include the genetic structure found between intertidal and subtidal populations (Oetjen et al. 2010) and between seagrasses within a lagoon with extremely high temperatures and seagrasses outside the same lagoon (Nguyen et al. 2023).

Most studies have revealed genetic differences between seagrasses using microsatellite markers (e.g., Coyer et al. 2004; Olsen et al. 2004). These markers can reveal genetic structure, but the results may be biased and resolution is low when few loci are used (Allendorf et al. 2010). Our perspective on seagrass conservation has recently become local (Harenčár et al. 2018); single nucleotide polymorphisms (SNPs) within loci that are genome-wide and abundant may enable detection of highly resolved genetic structures with less bias and high statistical power on a local scale (Allendorf et al. 2010). The genetic structure of seagrasses has been identified using SNPs on a global scale (Yu et al. 2023) and on a regional scale in South Africa (Phair et al. 2019), but we are aware of only one study that has used SNPs on a local scale (Nguyen et al. 2023). Conservation of seagrasses on a local scale requires use of SNPs to clarify the genetic structure.

This study explored how the genetic structure of seagrasses could be discriminated using SNPs. We targeted Zostera marina L. (eelgrass), which is a seagrass species distributed widely in the northern hemisphere (Short et al. 2007). The genetic structure of eelgrass populations was investigated in three semi-enclosed areas along the coast of Japan facing the Pacific Ocean. The alongshore extent of one area was nearly 200 km, and the corresponding dimension of the other two areas was nearly 20 km. We focused on the spatial scales of IBD and IBR (i.e., seascape topography) in the genetic structure and the genetic heterogeneity on these multiple spatial scales.

Study areas and sampling method

Eelgrass, Zostera marina L., is distributed in the coastal waters of Japan between temperate and subarctic seas (Nakaoka and Aioi 2001). The study areas were the Seto Inland Sea, Tokyo Bay, and the coastal waters of the town of Akkeshi, all of which are located along the Pacific Ocean (Fig. 1a) and are sites of eelgrass beds. All the sampling sites were in sheltered areas and near the offshore ocean.

The Seto Inland Sea (SIS) has mouths at its western and eastern sides that connect it to the Pacific Ocean (Fig. 1a). The calm conditions in the SIS account in part for the widespread eelgrass within it (Yoshida 2012). The area of eelgrass in the SIS began to decline in the 1970s because of the degradation of water quality and the reclamation of sites for coastal development and port construction (Komatsu 1997). Both restoration efforts in the 2000s and natural reproduction have allowed the eelgrass beds to largely recover (Morita et al. 2002; Hosokawa et al. 2019); natural reproduction has provided evidence of connections among eelgrass populations in the SIS. We focused on the populations on the western side (spatial scale < 200 km), which are isolated by a geographical barrier, the Hoyo Strait (Fig. 1b). We sampled at 12 sites within the inner SIS and two sites outside the SIS by the Hoyo Strait.

Tokyo Bay adjoins the Pacific Ocean at the mouth of the bay (Fig. 1a). Although eelgrass beds were widely distributed in the bay in the 1900s (Morita 2013), the present beds are fragmented because much of the eelgrass disappeared after the urbanization of the coastline. The eelgrass was sampled at three sites located ~ 20 km from each other (Fig. 1c). There has been an eelgrass bed at the Futtsu site since before urbanization (Yamakita and Nakaoka 2009; Yamakita et al. 2011). The Yokohama Seaside Park (YSP) is an artificial, shallow flat that lies within 1 km of another seagrass bed that is distributed on the flat (Watanabe et al. 2024). Eelgrass seed was artificially dispersed at the YSP as part of restoration programs (https://www.spf.org/opri/newsletter/120_3.html). Although the Hashirimizu coast has been hypothesized to be the source of the seed (see Fig. 1c for the location of the coast), there is no definitive record with which to test that hypothesis. Eelgrass sampling was performed at subsites YSP1, YSP2, and YSP3, which are 20–40 m from each other within the site. The third sampling site was Kurihama Bay, which is located at the outside of Tokyo Bay (Fig. 1c). Aerial photographs indicate that the eelgrass beds have recovered since the 1990s at this site (Hosokawa et al. 2010). Five microsatellite markers have been used to analyze the genetic differences between the eelgrass at different sites in Tokyo Bay (Tanaka et al. 2011), but the genetic structures of the three populations remain unclear.

The Akkeshi site consisted of the Akkeshi-ko estuary and Akkeshi Bay, which are connected by a narrow channel (Fig. 1d). There are areal gradients of water temperature and salinity between Akkeshi Bay and the mouths of small rivers as well as Bekanbeushi River (Momota and Nakaoka 2018). The biomass of eelgrass, which is distributed in sandy or muddy areas at depths of ~ 5 m in this study area, varies along substrate gradations (Hasegawa et al. 2008; Momota and Nakaoka 2017; Momota and Nakaoka 2018). The tide drives currents, and the outflow of water from the estuary to the bay is governed by the estuarine circulation at the surface (Hasegawa et al. 2018), but the currents are restricted by the vegetation in the estuary (Hasegawa et al. 2008). We chose three sites from the eelgrass beds within the Akkeshi-ko estuary and two sites outside the estuary through a channel in Akkeshi Bay (Fig. 1d). The study area within Akkeshi Bay was the smallest of the three areas that we studied.

We chose 22 sites for eelgrass sampling (Table 1). We collected at least 20 samples at each site from within the bed or 10–20 samples from rafted individuals. The youngest and/or second-youngest eelgrass leaves were cut from each individual to mitigate potential DNA contamination from epiphytes on the eelgrass leaves (Hosokawa et al. 2009). The leaf samples were subsequently rinsed with distilled water, wiped with paper towels, and packed in silica gel for transport to the laboratory. At most sites, the sampling was performed in 2022 and 2023, but at the KRH site in Tokyo Bay in 2019 (Table 1).

Table 1

Samples used in this study. Type of sample: A = sample collected from bed, B = drifted sample, and C = sample obtained in 2019
Region	Site	Type of site	Abbreviation	Sampling		No. samples taken	Type of sample
Region	Site	Type of site	Abbreviation	Year	Month	No. samples taken	Type of sample
SlS	Nakatsu	inner	NKT	2022	Aug	10	B
	Aio	inner	AIO	2022	Dec	20	A
	Aio	inner	AIOd	2022	Aug	20	B
	Murozumi	inner	MZ	2022	Aug	13	B
	Beppu	inner	BPU	2022	Aug	10	B
	Hiji	inner	HIJ	2022	Dec	20	A
	Mitsukue	inner	MTK	2022	Dec	20	A
	Suo-oshima	inner	SUO	2022	Dec	20	A
	Gogo-shima	inner	GOG	2022	Dec	20	A
	Baishinji	inner	BSJ	2022	Aug	16	B
	Mitsukuchi	inner	MIT	2022	Dec	20	A
	Omishima	inner	OMS	2022	Aug	15	B
	Saiki	outer	SIK	2023	Jun	20	B
	Uwajima	outer	UWJ	2023	Jun	20	B
Tokyo Bay	Futtsu	inner	FT	2022	May	40	A
	Yokohama Seaside Park	inner	YSP	2022	May	29	A
	Kurihama	outer	KRH	2019	June	20	A, C
Akkeshi	Akkeshi-ko1	inner	AK1	2022	Jul	39	A
	Akkeshi-ko2	inner	AK2	2022	Jul	41	A
	Central lake	inner	CL	2022	Jul	20	A
	Shinryu	outer	SR	2022	Jul	20	A
	Aininkappu	outer	AI	2022	Jul	20	A

Genetic analysis

In 2019, the samples were freeze-dried and homogenized, and the genomic DNA was extracted using an MPure Bacterial DNA extraction kit (MP Bio Japan K.K., Tokyo, Japan). In 2022 and 2023, the samples were ground in liquid nitrogen, and the genomic DNA was extracted using an Isospin Plant DNA extraction kit (Nippon Gene Co., Ltd., Tokyo, Japan).

Genotyping by Random Amplicon Sequencing-Direct (GRAS-Di®) was used for genome-wide sequencing. The GRAS-Di method involves a two-step polymerase chain reaction (PCR) with random primers to generate a sequence library of amplicons (Enoki and Takeuchi 2018; Enoki 2019); details of the method are described in Patent ID P2018 42548A. The library construction and sequencing were carried out by Bioengineering Lab. Co., Ltd. (Sagamihara, Japan) for the samples in 2019 and by GeneBay, Inc. (Yokohama, Japan) for the samples in 2022 and 2023. The sequencing was performed using a NextSeq 500 (Illumina, San Diego, CA, USA) with a 76-bp paired-end protocol for the samples in 2019 and a DNBSEQ-G400 (MGI Tech Co., Ltd., Shenzhen, China) with a 150-bp paired-end protocol for the samples in 2022 and 2023.

SNP filtering

We used Cutadapt version 2.8 (Martin 2011) to filter the reads generated by next-generation sequencing by trimming low-quality ends in the sequence reads (≤ Q20), filtering of short reads (remaining length > 51 bp) and removal of adapters. The cleaned sequences were mapped to the reference genome (JGI: Zostera marina v3.1; Ma et al. 2021) using the BWA-MEM algorithm in BWA version 0.7.17 (Li and Durbin 2009). Identification of duplicate reads, base quality score recalibration, and SNP calling were performed using GATK version 4.2 (McKenna et al. 2010); base quality score recalibration was performed by hard filtering.

Repeated SNPs are noise in the analyses of genetic structure and were filtered using bedtools version 2.27.1 (Quinlan and Hall 2010). Repeat sequences in the reference genome were masked. Low-quality SNPs (< Q20 and < 5 in depth) were filtered using VCFtools version 0.1.16 (Danecek et al. 2011), and the insertions and deletions were removed. We defined this filtering to be the first stage of SNP filtering. The rate of genotyping may have been lower for samples collected in 2019 than for newer samples after this stage of filtering (see Figs. S1 and S2 in Supporting Information A).

Analysis of genetic structure

The occurrence of IBD and IBR was tested based on the genetic distances in each study area. The genetic distance between sites was quantified by the value of the pairwise F_ST/(1 − pairwise F_ST). The pairwise F_ST was calculated using SNPs from the first stage of filtering and the individuals that remained after the second stage of filtering (vide infra). The geographic distance of the site pair was calculated using the direct distance of the pair on the map and the distance via the transit point at the pair where the direct distance was not rational on the real current (Fig. 1b, c, d). The possibility of resistance was considered between the outer and inner sites and between outer sites (Table 1). The nonparametric Wilcoxon rank sum test was used to test the difference in the genetic distance between sites with and without the possibility of resistance. The other test was based on a multiple matrix regression with randomization (MMRR) (Wang 2013). We tested for the possibility of a single pattern of IBD or IBR as well as multiple patterns. In that case, we set the paired site with possible resistance to 1 and the paired site without possible resistance to 0. The number of random permutations was set to 9999 in the SIS because of the large number of total, real permutations and to 119 permutations calculated from five sites in Tokyo Bay and Akkeshi. We also demonstrated the statistical significance of IBD at a small number of sites by a simulation using the inner sites in the SIS with the MMRR algorithm. The simulation selected five sites randomly per iteration, and its statistical significance was tested with 119 permutations. The simulation was performed 100,000 times. The number of significant iterations was counted. The pairwise F_ST was calculated using the fs.dosage functions in hierfstat version 0.5–11 of the R package (Goudet et al. 2022). The geographic distance was calculated using the distGeo function in geosphere version 1.5–18 (Hijmans et al. 2022). The Mantel test was performed by using the mantel function in vegan version 2.6-4 (Oksanen et al. 2022).

We analyzed the detailed genetic structure using a principal component analysis (PCA) and ADMIXTURE analysis. These analyses usually required additional filtering to remove noise. This second stage of filtering was set to avoid the drawback of reducing the numbers of samples and SNPs (see Table S1 and Fig. S3 in Supporting Information A). Such a reduction, which would have decreased the statistical power and caused a bias in the calculation of statistics (Weir and Goudet 2017), would have occurred in our dataset because of the increased strength of filtering (Fig. S4) and would have caused a serious problem in comparison of statistics between the samples from 2019 and more recent samples (Fig. S5). In the second stage, we filtered the samples based on their rate of genotyping, the SNPs based on their rate of genotyping, and the SNPs based on their minor allele frequency (see Supplementary Information B). We used strict filtering in the SIS and Akkeshi (Table S2 in Supplementary Information C) and moderate filtering in Tokyo Bay to retain the KRH samples obtained in 2019 (Table S3).

The PCA used a count datum of 0 when the alleles did not match the reference genome at marker j and individual i of the diploid sample, 1 when the alleles matched the reference in heterozygous individuals, and 2 when the alleles matched the reference in homozygous individuals (Patterson et al. 2006). The count data were normalized, and the analysis was performed on a matrix consisting of the count data. A scree plot was used to determine the number of significant principal components in the PCA.

ADMIXTURE analysis is a statistical technique for estimating individual ancestries in K genetic clusters based on the likelihood of an allele’s occurring at locus j in individual i (Alexander et al. 2009). ADMIXTURE analysis was conducted with genetic clusters ranging from K = 1 to K = 10. A fivefold cross-validation was carried out for each number of genetic clusters. We considered the number of genetic clusters with the minimum cross-validation to be the optimized model (Alexander and Lange 2011). We used the model with minimum cross-validation and models with cross-validation close to the minimum cross-validation to find the genetic structure in the three study areas.

Nucleotide diversity, the proportion of variants, and the population-specific F_IS and F_ST were calculated based on the SNPs after the first stage of filtering. Because nucleotide diversity can be biased when calculated using a marker for which the rate of genotyping is biased, we calculated it for the loci for which the genotyping rate was 100% (i) at the site or (ii) in the study area. The first method made it possible to obtain a relatively large number of SNPs, but there was a risk of assessing the nucleotide diversity under different markers across sites. The second method was more likely to assess the nucleotide diversity under same marker across sites, but it may have been less powerful because the number of SNPs was relatively low. We refer to the former as “nucleotide diversity” and the latter as “restricted nucleotide diversity”. The population-specific F_IS is a relative inbreeding coefficient (Weir and Goudet 2017), which has a low value when the degree of heterozygosity is high and a value of 1 when the homozygosity is a maximum. The population-specific F_ST is a measure of deviation from the ancestral population (Kitada et al. 2021) and has a maximum value of 1. The population-specific F_IS and F_ST can take negative values. The two nucleotide diversities and the population-specific F_IS and F_ST were estimated with a bootstrapping method, randomly resampling one-tenth of the SNPs after the first stage of filtering in a bootstrap sample and with 200 bootstrap samplings; their 95% confidence intervals (CI) were also calculated. Because the restricted nucleotide diversity and the population-specific F_IS and F_ST depend on individuals, they were estimated in the individuals retained after strict and moderate filtering in the analysis of Tokyo Bay.

The PCA and ADMIXTURE analysis were performed using the R package pcadapt version 4.3.3 (Luu et al. 2017) and the ADMIXTURE software version 1.3.0 (Alexander and Lange 2011), respectively. Missing data were inputted using the pairwise covariance approach as the default process in the R package and ignored in ADMIXTURE. The nucleotide diversity was calculated by the pi.dosage functions in hierfstat version 0.5–11 of the R package. The population-specific F_IS and population-specific F_ST were calculated using the fs.dosage functions in the hierfstat package.

The first stage of filtering resulted in 176,388 SNPs in Tokyo Bay and even more in the SIS and Akkeshi datasets (Table 2). The second stage of filtering rejected no more than two individuals at a site. The numbers of individuals that remained were 239 in the SIS, 87 in Tokyo Bay, and 139 in Akkeshi. The statistics were calculated on the individuals that remained after the second stage of filtering and the SNPs of the individuals.

Table 2

Numbers of individuals and single nucleotide polymorphisms (SNPs) by filtering
Study area	SNPs after the 1st stage of filtering	2nd stage of filtering			SNPs after the 1st stage of filtering on the individuals that remained
Study area	SNPs after the 1st stage of filtering	# of individuals rejected at site	# of individuals remained/analyzed	SNPs remained
SIS	323,645	2 at MZ, 1 at BSJ, 1 at OMS, 1 at UWJ	239/244	4893	100,203
Tokyo Bay	176,388	1 at FT, 1 at KRH	87/89	1827	49,451
Akkeshi	215,394	1 at AK1	139/140	2988	65,776

IBD and IBR

The genetic distance was lower in pairs without possible resistance than in pairs with resistance in the SIS (Fig. 2). The difference was statistically significant by a non-parametric test (Wilcoxon rank sum test: W = 0, p = 2.29×10^− 13). The single matrix regressions in the SIS showed that both models with IBD (MMRR: r² = 0.16, p = 0.0001) and IBR (r² = 0.87, p = 0.0001) were statistically significant (Fig. 2a) and the coefficient of determination, r², was greater in the model with IBR. A multiple matrix regression showed that the genetic distance of the SIS could be explained by integrating the IBD and IBR models (MMRR: r² = 0.89, p = 0.0001). The strength of the IBR was 6.7 times that of the IBD along 100 km of geographic distance. The simulation of the IBD test at small sites in the SIS resulted in a 4.7% significant iteration. The slopes of significant IBDs were variable.

In Tokyo Bay, the genetic difference was statistically significant between pairs with and without possible resistance based on the nonparametric Wilcoxon rank sum test (W = 0, p = 0.0095). The single matrix regression revealed a statistically significant IBD (Fig. 2b, MMRR: r² = 0.47, p = 0.033), whereas the IBR regression was not significant but had greater explanatory power (r² = 0.83, p = 0.117). A multiple matrix regression was not statistically significant but had greater explanatory power (MMRR: r² = 0.86, p = 0.1).

In Akkeshi (Fig. 2c), the difference in genetic distance between pairs with and without possible resistance was not significant by a non-parametric test (Wilcoxon rank sum test: W = 4, p = 0.183). The matrix regression revealed no significant IBD or IBR by the single matrix regression (IBD: r² = 0.78, p = 0.09 and IBR: r² = 0.28, p = 0.2) and by multiple matrix regression (r² = 0.78, p = 0.225).

Genetic structure in the SIS

The PCA and ADMIXTURE analysis were performed using 4893 SNPs in the SIS (Table 2). The scree plot showed no apparent breaking PC in the proportion of eigenvalues, but their proportion at PC4 was approximately twice the proportion between PC10 and PC20 (Fig. 3a). Two outer sites, SIK and UWJ, and the inner sites were clustered in the first two dimensions of principal component space (PC1 and PC2) (Fig. 3b), and they were also isolated by resistance (Fig. 2a). The ADMIXTURE analysis at K = 3 also separated these three clusters in the PCA (Fig. 3d; see Table S4 for the fivefold cross-validation). The inner portion of the SIS formed a cluster on the PC1 and PC2 axes (Fig. 3b). However, the genetic variability among sites appeared in the inner portion on the PC3 and PC4 axes (Fig. 3c) and was also apparent in the IBD (Fig. 2a). The structure on the PC3 and PC4 axes formed a triangle with vertices at MTK (close to the Hoyo Strait); MIT and OMS (eastern side of the SIS); and NKT, AIO, and AIOd (western side). The ADMIXTURE analysis at K = 5 also showed a structure in the inner SIS portion (Fig. 3d).

The nucleotide diversity at SIK and UWJ was 0.085–0.098 (95% CI) and lowest among all sites (Fig. 3e). The restricted nucleotide diversity had a large CI at all sites, and its mean value was lowest at the SIK and UWJ among all sites. Although the nucleotide diversity was lowest at MZ among the inner-portion sites of the SIS, it might have been biased because of the low number of SNPs at MZ. The restricted nucleotide diversity was not lower at MZ than at the other inner-portion sites. The population-specific F_ST values were 0.189–0.280 at SIK and 0.189–0.311 (95% CIs) at UWJ. These values exceeded that at MZ (95% CI = 0.046–0.122), where the population-specific F_ST was the greatest at the inner sites. The population-specific F_IS was lower at SIK (95% CI = − 0.291 to − 0.119) than at NKT, AIOd, BSJ, MIT, and OMS, where the 95% CI of the population-specific F_IS varied between − 0.185 and − 0.088.

Genetic structure in Tokyo Bay

We used 1827 SNPs to perform the PCA and ADMIXTURE analysis in Tokyo Bay (Table 2). The scree plot of eigenvalues stepped between PC1 and PC2, and the eigenvalues were almost the same for PC2 and PC3 (Fig. 4a). On PC1, variabilities of individuals were apparent within the YSP and within the subsites of YSP (Fig. 4b). The intra-site variability was also apparent on PC3 (Fig. 4c). The individual at the KRH was distinguished from other inner-site individuals on PC2 (Fig. 4b). The ADMIXTURE analysis revealed a difference between the FT individuals and those from two other sites and the variability within the YSP and within the YSP subsites (Fig. 4d). The ADMIXTURE analysis at K = 3 resulted in the dominance of KRH individuals. The genetic cluster that dominated at KRH was also apparent among YSP individuals. The KRH individuals that remained after moderate filtering were placed between the PC1 and PC2 of the PCA based on strict filtering (Fig. S9). The addition of KRH individuals did not affect the probability of genetic clusters in other individuals in the ADMIXTURE analysis using strict filtering.

Nucleotide diversity was estimated using 49,451 SNPs of the 87 individuals (Table 2). The nucleotide diversity at KRH was 0.137–0.169 (95% CI) and lower than at FT (95% CI = 0.182–0.197) (Fig. 4e). The mean of the restricted nucleotide diversity was also lower at KRH than at FT, although the 95% CI ranges were large. The population-specific F_ST was greater at KRH than at the other sites. The population-specific F_ST was variable across the subsites at the YSP. The values of the statistics under moderate filtering differed from those under strict filtering, but the relative results were unaffected (Fig. S9).

Genetic structure at Akkeshi

The PCA and ADMIXTURE analysis were performed using 2988 SNPs at Akkeshi (Table 2). The PCA scree plot showed that the genetic structure was two-dimensional (Fig. 5a). An outer site AI was distinguished from the other four sites on PC1 (Fig. 5b). The three inner sites and another outer site, SR, were distributed on PC2 and formed an admixed structure. The ADMIXTURE analysis at K = 3 also showed this genetic structure (Fig. 5d; see Table S4 for the fivefold cross-validation). The PCA and ADMIXTURE analysis revealed that the individuals at AK2 showed admixture between the AK1 individuals and the SR and CL individuals, although the geographical location of AK2 was not intermediate between them (Fig. 1d). The variance associated with the PC3 and PC4 axes was governed by the individual noise at AK1 and AK2 (Fig. 5c). The nucleotide diversity and restricted nucleotide diversity did not appear to differ but tended to be lower at outer sites than at AK1 (Fig. 5e). The population-specific F_ST also did not appear to differ but tended to increase from AK1 to AI.

Genome-wide SNPs distinguished the genetic structure in the eelgrass population at three different spatial scales between linear dimensions of 10 km and 200 km. Isolations by geographic distance and resistance were revealed over a distance of less than 200 km. At a distance of 20 km, an IBD was found, and there was variability across individuals within the site. At a distance of 10 km, the genetic structure was governed by local heterogeneity, and each IBD or IBR was not statistically significant. These results indicated that recently developed genetic tools could reveal the genetic structure of seagrasses on local scales. This study also showed that recently developed genomic tools could discriminate the local scale of genetic structure and could contribute to the advancement of scientific knowledge.

Isolation by distance was apparent in the SIS (Fig. 2a). In the inner portion of the SIS, the variable slopes of the IBD should result in a variable strength of genetic connectivity between sites through transportation of seeds by currents. However, the triangular form of the genetic structure (Fig. 3c) implicated the factors that governed the variability. The current driven by the complex topography (Fig. S10) would be a factor governing the formation in the inner SIS. The admixture among the vertices of the triangle suggested that the genetic structure of the eelgrass within the inner portion of the SIS was formed through multigenerational connectivity (Legrand et al. 2022). The coexistence of IBR with IBD and the greater strength of IBR than IBD (Fig. 2a) may be a new finding in seagrasses. The greater IBR may have resulted from the regulation of current connectivity between the inner and outer waters at the Hoyo Strait (Fig. S10).

In Tokyo Bay, IBD was statistically significant (Fig. 2b). However, genetic structure was apparent across sites as well as within sites in the YSP (Fig. 3). The intra-site variability was an exceptional feature of that site compared to the SIS and Akkeshi sites. Intra-site variability may not be caused by environmental selection, because there is less of a gradient in water quality within the subsites around the YSP (Watanabe et al. 2024). The heterogeneity of the eelgrass population in the YSP has been impacted the most by transplantation during the ten years since the construction of the shallow flat (https://www.spf.org/opri/newsletter/120_3.html). Individuals transplanted from different donor sites could coexist within an area with linear dimensions of 20–40 m. Although connectivity can result from pollen dispersal (Kendrick et al. 2012; Kendrick et al. 2017), such genetic heterogeneity would mean that mating was limited after immigration for ten years. The present bed in the YSP has been maintained by the clonal growth of immigrated individuals, and it is likely that mating among individuals from different donor sites has been limited.

Although neither IBD nor IBR was supported statistically (Fig. 2c), the genetic structure of the eelgrass was distinguished in Akkeshi (Fig. 5). However, a mismatch between the geographic location and the genetic structure at the three sites was found in the Akkeshi-ko estuary. The complexity of the environmental gradients is known at the three sites (Momota and Nakaoka 2018). Such environmental conditions can cause a local adaptation and heterogeneous genetic structure on a local scale (Oetjen et al. 2010). In addition, because the canopy of eelgrass leaves regulates the horizontal transport of water (Abdelrhman 2003), a complex current pattern may be formed by the eelgrass meadow, which is widely distributed in the estuary. As a result, environmental selection and/or pollen and seed dispersal via complex currents may create heterogenous genetic structure within the estuary.

The population-specific F_ST tended to be higher at the outer sites of the SIS (Fig. 3e), Tokyo Bay (Fig. 4e), and Akkeshi (Fig. 5e). These results indicated that the populations at the inner sites could be ancestral. As discussed above, the limited connectivity between the ocean currents outside the Hoyo Strait and the regional currents in the inner portion may form IBR in the SIS. On the other hand, estuarine circulation includes a residual current toward the ocean in the surface layer of the inner bay (MacCready and Geyer 2010; Geyer and MacCready 2014), and there is such a current within Tokyo Bay (Hosokawa and Okura 2022) and Akkeshi Bay (Hasegawa et al. 2008). It is likely that immigration from the outer to inner bay by the rafting of seed is regulated by the strong estuarine circulation, especially around the outer sides. The ancestral structure and the patterns of transport by the estuarine circulation suggest that the origins of eelgrass populations are inland seas, semi-enclosed bays, and estuaries. In addition, we noted the usual gradients of water temperature, salinity, and nutrients between the inner and outer sites. The suggested gene flow from the inner to outer populations may be formed by genetic drift and environmental selection in these areas. However, our results were insufficient to verify this scenario. Future studies are needed to test this hypothesis.

The high statistical discriminability in genetic structure was achieved by the large number of loci analyzed by a tool with genome-wide markers, which facilitated detection of the statistical significance of the genetic differences. However, we fear that such high discriminability will cause extreme regulation of seagrass transplantations from other populations. Extreme regulation would hinder the recovery of seagrass beds and their ecosystem services. To avoid extreme regulation, we should know that statistically significant differences and the strength of genetic differences facilitate distinguishing genetic structure (Palsbøll et al. 2007). The case of SIS showed that assessing relative genetic differences can make delineating conservation units a reality. The genetic structure of SIS implies that we should consider the transplantations between populations across the Hoyo Strait carefully by comparing those within the inner portion because of their clear genetic difference across the strait. Although in the cases of Tokyo Bay and Akkeshi it was possible to discriminate among sites genetically, it is difficult to discern how we could delineate conservation units in these areas. Because the patterns we can refer to for relative assessments may occur on a scale larger than local areas, identifying only the genetic structure of seagrasses at these local scales may be insufficient. How do they differ evolutionally in ancestral structure, and how much do they differ in evolutional meaning? The answers to these questions are directly related to genetic differences and will be required for assessing conservation units where the assessing scale is local.

In conclusion, genome-wide SNPs revealed that the isolations by geographical distance and barrier governed the genetic structure of eelgrass populations on scales of 200 km and that the heterogeneity with the intra-site variability and spatial complexity occurred in two areas with linear dimensions of 10 and 20 km where the number of populations was limited within the areas. This study has been the first to reveal the scale of transition where the relative importance of local heterogeneity increases from the genetic structures of seagrasses governed by geographical isolation by distance and resistance. Also, the results of this study implied that our perspective of seagrass conservation should be more local than before genomic tools were developed. The next challenge for delineating seagrass conservation units would be identifying the evolutional meaning of genetic distance and developing methods to assess the genetic difference.

Acknowledgements We are grateful to the Hiroshima Research and Engineering Office for Port and Airport of the Chugoku Regional Development Bureau, Ministry of Land, Infrastructure, Transport, and Tourism for providing opportunities to conduct the field surveys. We thank T. Izumi at the Port and Airport Research Institute, Japan, and Y. Kagami of the Laboratory of Aquatic Science Consultant Co., Ltd., Japan, for assistance in field surveys. We also thank A. Kajita at the Suiken Research Co., Ltd., and K. Sudo of the National Research Institute of Fisheries Technology, Japan Fisheries Research and Education Agency for giving us the information about eelgrass sites in the SIS. We also thank I. Fujita of the Port and Airport Research Institute, Japan, for offering us the opportunity to use a powerful computer for filtering and analysis of the genetic data.

Author Contributions SH and KM conceived the ideas and designed the methodology; SH, KM, MS, KW, YW, SH, and SO performed sample collection; SH analyzed the data; SH, KM, MS, SU, TK, and YU guided their interpretation; SH wrote the first draft of the manuscript. All authors contributed critically to the drafts.

Funding The present research was financially supported by the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research 22H01605 and 23K22875 to SH, KM, TK and YU and 22K12470 to KM, MS, and KW.

Disclosure statement The authors declare no competing interests

Data availability GRAS-Di sequence data are available from DDBJ under accession numbers: DRR571563–DRR571582, DRR575036–DRR575439 and DRR575503–DRR575551.

Abdelrhman, M. A. 2003. Effect of eelgrass Zostera marina canopies on flow and transport. Marine Ecology Progress Series 248: 67–83.
Alexander, D. H., and K. Lange. 2011. Enhancements to the ADMIXTURE algorithm for individual ancestry estimation. BMC Bioinformatics 12. https://doi.org/10.1186/1471-2105-12-246.
Alexander, D. H., J. Novembre, and K. Lange. 2009. Fast model-based estimation of ancestry in unrelated individuals. Genome Research 19: 1655–1664. https://doi.org/10.1101/gr.094052.109.
Allendorf, F. W., P. A. Hohenlohe, and G. Luikart. 2010. Genomics and the future of conservation genetics. Nature Reviews Genetics 11: 697–709. https://doi.org/10.1038/nrg2844.
Allendorf, F. W., R. F. Leary, P. Spruell, and J. K. Wenburg. 2001. The problems with hybrids: Setting conservation guidelines. Trends in Ecology and Evolution 16: 613–622. https://doi.org/10.1016/S0169-5347(01)02290-X.
Arnaud-Haond, S., M. Migliaccio, E. Diaz-Almela, S. Teixeira, M. S. Van De Vliet, F. Alberto, G. Procaccini, C. M. Duarte, and E. A. Serrão. 2007. Vicariance patterns in the Mediterranean Sea: East-west cleavage and low dispersal in the endemic seagrass Posidonia oceanica. Journal of Biogeography 34: 963–976. https://doi.org/10.1111/j.1365-2699.2006.01671.x.
Costanza, R., R. D’Arge, R. De Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, et al. 1997. The value of the world’s ecosystem services and natural capital. Nature 387: 253–260.
Coyer, J. A., O. E. Diekmann, E. A. Serrão, G. Procaccini, N. Milchakova, G. A. Pearson, W. T. Stam, and J. L. Olsen. 2004. Population genetics of dwarf eelgrass Zostera noltii throughout its biogeographic range. Marine Ecology Progress Series 281: 51–62. https://doi.org/10.3354/meps281051.
D’Aloia, C. C., S. M. Bogdanowicz, R. G. Harrison, and P. M. Buston. 2014. Seascape continuity plays an important role in determining patterns of spatial genetic structure in a coral reef fish. Molecular Ecology 23: 2902–2913. https://doi.org/10.1111/mec.12782.
Danecek, P., A. Auton, G. Abecasis, C. A. Albers, E. Banks, M. A. DePristo, R. E. Handsaker, et al. 2011. The variant call format and VCFtools. Bioinformatics 27: 2156–2158. https://doi.org/10.1093/bioinformatics/btr330.
Ellstrand, N. C. 2014. Is gene flow the most important evolutionary force in plants? American Journal of Botany 101: 737–753. https://doi.org/10.3732/ajb.1400024.
Enoki, H. 2019. The construction of psedomolecules of a commercial strawberry by denovoMAGIC and new genotyping technology, GRAS-Di. In Proceedings of the Plant and Animal Genome Conference XXVII. San Diego, CA.
Enoki, H., and Y. Takeuchi. 2018. New genotyping technology, GRAS-Di, using next generation sequencer. In Proceedings of the Plant and Animal Genome Conference XXVI. San Diego, CA.
Funk, W. C., J. K. McKay, P. A. Hohenlohe, and F. W. Allendorf. 2012. Harnessing genomics for delineating conservation units. Trends in Ecology and Evolution 27: 489–496. https://doi.org/10.1016/j.tree.2012.05.012.
Geyer, W. R., and P. MacCready. 2014. The estuarine circulation. Annual Review of Fluid Mechanics. https://doi.org/10.1146/annurev-fluid-010313-141302.
Goudet, J., T. Jombart, Z. N. Kamvar, E. Archer, O. Hardy. 2022. Estimation and tests of hierarchical F-statistics. URL https://cran.r-project.org/web/packages/hierfstat/hierfstat.pdf
Harenčár, J. G., G. A. Lutgen, Z. M. Taylor, N. P. Saarman, and J. M. Yost. 2018. How population decline can impact genetic diversity: a case study of eelgrass (Zostera marina) in Morro Bay, California. Estuaries and Coasts 41: 2356–2367. https://doi.org/10.1007/s12237-018-0421-8.
Hasegawa, N., H. Abe, T. Onitsuka, and S. Ito. 2018. Association between the planktonic larval and benthic stages of Manila clam Ruditapes philippinarum in eastern Hokkaido, Japan. Fisheries Science 84: 237–249. https://doi.org/10.1007/s12562-017-1168-z.
Hasegawa, N., M. Hori, and H. Mukai. 2008. Seasonal changes in eelgrass functions: Current velocity reduction, prevention of sediment resuspension, and control of sediment-water column nutrient flux in relation to eelgrass dynamics. Hydrobiologia 596: 387–399. https://doi.org/10.1007/s10750-007-9111-4.
Hijmans, R., Karney, C., Williams, E., Vennes, C. 2022. Spherical trigonometry. URL https://cran.r-project.org/web/packages/geosphere/geosphere.pdf
Hohenlohe, P. A., W. C. Funk, and O. P. Rajora. 2021. Population genomics for wildlife conservation and management. Molecular Ecology 30: 62–82. https://doi.org/10.1111/mec.15720.
Hori, M., and M. Sato. 2021. Genetic effects of eelgrass restoration efforts by fishers’ seeding to recover seagrass beds as an important natural capital for coastal ecosystem services. Population Ecology 63: 92–101. https://doi.org/10.1002/1438-390X.12073.
Hosokawa, S., T. Hara, and M. Nozu. 2019. Advanced engineering techniques and environmental measures for beneficial sediment use in Japan. In Australasian Coasts and Ports 2019 Conference, 611–616.
Hosokawa, S., Y. Nakamura, and T. Kuwae. 2009. Increasing temperature induces shorter leaf life span in an aquatic plant. Oikos 118. https://doi.org/10.1111/j.1600-0706.2009.17288.x.
Hosokawa, S., M. Nakaoka, E. Miyoshi, T. Kuwae. 2010. Seed dispersal supporting areal expansion of meadow in the seagrass, Zostera marina L. Report of the Port and Airport Research Institute 49: 31–49. in Japanese with English abstract
Hosokawa, S., M. Nakaoka, E. Miyoshi, and T. Kuwae. 2015. Seed dispersal in the seagrass Zostera marina is mostly within the parent bed in a protected bay. Marine Ecology Progress Series 523: 41–56. https://doi.org/10.3354/meps11146.
Hosokawa, S., and S. Okura. 2022. Long-term observation of current at the mouth of Tokyo Bay. Coastal Engineering Journal 64: 648–659. https://doi.org/10.1080/21664250.2022.2122300.
Kendrick, G. A., R. J. Orth, J. Statton, R. Hovey, L. R. Montoya, R. J. Lowe, S. L. Krauss, and E. A. Sinclair. 2017. Demographic and genetic connectivity: The role and consequences of reproduction, dispersal and recruitment in seagrasses. Biological Reviews 92: 921–938. https://doi.org/10.1111/brv.12261.
Kendrick, G. A., M. Waycott, T. J. B. Carruthers, M. L. Cambridge, R. Hovey, S. L. Krauss, P. S. Lavery, et al. 2012. The central role of dispersal in the maintenance and persistence of seagrass populations. Bioscience 62: 56–65.
Kim, J. H., J. H. Kang, J. E. Jang, S. K. Choi, M. J. Kim, S. R. Park, and H. J. Lee. 2017. Population genetic structure of eelgrass (Zostera marina) on the Korean coast: Current status and conservation implications for future management. PLoS ONE 12. https://doi.org/10.1371/journal.pone.0174105.
Kitada, S., R. Nakamichi, and H. Kishino. 2021. Understanding population structure in an evolutionary context: Population-specific F_ST and pairwise F_ST. G3: Genes, Genomes, Genetics 11. https://doi.org/10.1093/g3journal/jkab316.
Komatsu, T. 1997. Long-term changes in the Zostera bed area in the Seto Inland Sea (Japan), especially along the coast of the Okayama Prefecture. Oceanologica Acta 20: 209–216.
Legrand, T., A. Chenuil, E. Ser-Giacomi, S. Arnaud-Haond, N. Bierne, and V. Rossi. 2022. Spatial coalescent connectivity through multi-generation dispersal modelling predicts gene flow across marine phyla. Nature Communications 13. https://doi.org/10.1038/s41467-022-33499-z.
Li, H., and R. Durbin. 2009. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25: 1754–1760. https://doi.org/10.1093/bioinformatics/btp324.
Lowe, W. H., and F. W. Allendorf. 2010. What can genetics tell us about population connectivity? Molecular Ecology 19: 3038–3051. https://doi.org/10.1111/j.1365-294X.2010.04688.x.
Luu, K., E. Bazin, and M. G. B. Blum. 2017. pcadapt: an R package to perform genome scans for selection based on principal component analysis. In Molecular Ecology Resources, 17:67–77. https://doi.org/10.1111/1755-0998.12592.
Ma, X., J. L. Olsen, T. B. H. Reusch, G. Procaccini, D. Kudrna, M. Williams, J. Grimwood, et al. 2021. Improved chromosome-level genome assembly and annotation of the seagrass, Zostera marina (eelgrass). F1000Research 10. https://doi.org/10.12688/f1000research.38156.1.
MacCready, P., and W. R. Geyer. 2010. Advances in estuarine physics. Annual Review of Marine Science 2: 35–58. https://doi.org/10.1146/annurev-marine-120308-081015.
Martin, M. 2011. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet Journal 17. https://doi.org/https://doi.org/10.14806/ej.17.1.200.
Martínez-García, L., B. Hansson, and J. Hollander. 2021. Assessment of local genetic structure and connectivity of the common eelgrass Zostera marina for seagrass restoration in northern Europe. Marine Ecology Progress Series 664: 103–116.
McKenna, A., M. Hanna, E. Banks, A. Sivachenko, K. Cibulskis, A. Kernytsky, K. Garimella, et al. 2010. The genome analysis toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data. Genome Research 20: 1297–1303. https://doi.org/10.1101/gr.107524.110.
Momota, K., and M. Nakaoka. 2017. Influence of different types of sessile epibionts on the community structure of mobile invertebrates in an eelgrass bed. PeerJ 2017: e2952. https://doi.org/10.7717/peerj.2952.
Momota, K., and M. Nakaoka. 2018. Seasonal change in spatial variability of eelgrass epifaunal community in relation to gradients of abiotic and biotic factors. Marine Ecology 39: e12522. https://doi.org/10.1111/maec.12522.
Morita, K. 2013. Study on suitable conditions for eelgrass bed and its effect on surrounding organisms. Doctoral thesis in The University of Tokyo.
Morita, K., M. Hamabata, H. Masuhara, K. Fukase, and Y. Kawasaki. 2002. Zostera (eelgrass) bed restoration in Japan. Fisheries Science 68: 1771–1774. https://doi.org/10.2331/fishsci.68.sup2_1771.
Nakajima, Y., Y. Matsuki, C. Lian, M. D. Fortes, W. H. Uy, W. L. Campos, M. Nakaoka, and K. Nadaoka. 2014. The Kuroshio Current influences genetic diversity and population genetic structure of a tropical seagrass, Enhalus acoroides. Molecular Ecology 23: 6029–6044. https://doi.org/10.1111/mec.12996.
Nakaoka, M., and K. Aioi. 2001. Ecology of seagrasses Zostera spp. (Zosteraceae) in Japanese waters: a review. Otsuchi Marine Science 26: 7–22.
Nanninga, G. B., P. Saenz-Agudelo, A. Manica, and M. L. Berumen. 2014. Environmental gradients predict the genetic population structure of a coral reef fish in the Red Sea. Molecular Ecology 23: 591–602. https://doi.org/10.1111/mec.12623.
Nguyen, H. M., Ruocco, M., Dattolo, E., Cassetti, F. P., Calvo, S., Tomasello, A., Marín-Guirao, L., Pernice, M., Procaccini, G. 2023. Signs of local adaptation by genetic selection and isolation promoted by extreme temperature and salinity n the Mediterranean seagrass Posidonia oceanica. Molecular Ecology 32: 4313–4328. https://doi.org/10.1111/mec.17032.
Oetjen, K., S. Ferber, I. Dankert, and T. B. H. Reusch. 2010. New evidence for habitat-specific selection in Wadden Sea Zostera marina populations revealed by genome scanning using SNP and microsatellite markers. Marine Biology 157: 81–89. https://doi.org/10.1007/s00227-009-1297-8.
Oksanen, J., G.L. Simpson, F.G. Blanchet, R. Kindt, P. Legendre, P.R. Minchin, R.B. O'Hara, P. Solymos, M.H.H. Stevens, E. Szoecs, H. Wagner, M. Barbour, M. Bedward, B. Bolker, D. Borcard, G. Carvalho, M. Chirico, M. De Caceres, S. Durand, H.B.A. Evangelista, R.F. John, M. Friendly, B. Furneaux, G. Hannigan, M.O. Hill, L. Lahti, D. McGlinn, M.H. Ouellette, E.R. Cunha, T. Smith, A. Stier, C.J.F. Ter Braak, J. Weedon. 2022. Community Ecology Package. URL https://cran.r-project.org/web/packages/vegan/vegan.pdf
Olsen, J. L., W. T. Stam, J. A. Coyer, T. B. H. Reusch, M. Billingham, C. Boström, E. Calvert, et al. 2004. North Atlantic phylogeography and large-scale population differentiation of the seagrass Zostera marina L. Molecular Ecology 13: 1923–1941. https://doi.org/10.1111/j.1365-294X.2004.02205.x.
Orth, R. J., T. J. B. Carruthers, W. C. Dennison, C. M. Duarte, J. W. Fourqurean, K. L. Heck JR, A. R. Hughes, et al. 2006. A global crisis for seagrass ecosystems. Bioscience 56: 987–996.
Orth, R. J., M. Luckenbach, and K. A. Moore. 1994. Seed dispersal in a marine macrophyte: Implications for colonization and restoration. Ecology 75: 1927–1939.
Palsbøll, P. J., M. Bérubé, and F. W. Allendorf. 2007. Identification of management units using population genetic data. Trends in Ecology and Evolution 22: 11–16. https://doi.org/10.1016/j.tree.2006.09.003.
Patterson, N., A. L. Price, and D. Reich. 2006. Population structure and eigenanalysis. PLoS Genetics 2: 2074–2093. https://doi.org/10.1371/journal.pgen.0020190.
Phair, N. L., R. J. Toonen, I. Knapp, and S. Von Der Heyden. 2019. Shared genomic outliers across two divergent population clusters of a highly threatened seagrass. PeerJ 2019. https://doi.org/10.7717/peerj.6806.
Procaccini, G., J. L. Olsen, and T. B. H. Reusch. 2007. Contribution of genetics and genomics to seagrass biology and conservation. Journal of Experimental Marine Biology and Ecology 350: 234–259. https://doi.org/10.1016/j.jembe.2007.05.035.
Quinlan, A. R., and I. M. Hall. 2010. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 26: 841–842. https://doi.org/10.1093/bioinformatics/btq033.
Sandoval-Castillo, J., N. A. Robinson, A. M. Hart, L. W. S. Strain, and L. B. Beheregaray. 2018. Seascape genomics reveals adaptive divergence in a connected and commercially important mollusc, the greenlip abalone (Haliotis laevigata), along a longitudinal environmental gradient. Molecular Ecology 27: 1603–1620. https://doi.org/10.1111/mec.14526.
Short, F., T. Carruthers, W. Dennison, and M. Waycott. 2007. Global seagrass distribution and diversity: A bioregional model. Journal of Experimental Marine Biology and Ecology 350: 3–20.
Sinclair, E. A., S. L. Krauss, J. Anthony, R. Hovey, and G. A. Kendrick. 2014. The interaction of environment and genetic diversity within meadows of the seagrass Posidonia australis (Posidoniaceae). Marine Ecology Progress Series 506: 87–98. https://doi.org/10.3354/meps10812.
Stafford-Bell, R. E., W. F. D. Van Dongen, R. W. Robinson, and A. A. Chariton. 2019. Connectivity of the seagrass Zostera muelleri within south-eastern Australia. Marine and Freshwater Research 70: 1056–1064. https://doi.org/10.1071/MF18333.
Talbot, S. L., G. K. Sage, J. R. Rearick, M. C. Fowler, R. Muñiz-Salazar, B. Baibak, S. Wyllie-Echeverria, A. Cabello-Pasini, and D. H. Ward. 2016. The structure of genetic diversity in eelgrass (Zostera marina L.) along the north pacific and bering sea coasts of Alaska. PLoS ONE 11. https://doi.org/10.1371/journal.pone.0152701.
Tanaka, N., T. Demise, M. Ishii, Y. Shoji, and M. Nakaoka. 2011. Genetic structure and gene flow of eelgrass Zostera marina populations in Tokyo Bay, Japan: Implications for their restoration. Marine Biology 158: 871–882.
Tanaka, N., Y. Omori, M. Nakaoka, and K. Aioi. 2002. Gene flow among populations of Zostera caespitosa Miki (Zosteraceae) in Sanriku Coast, Japan. Otsuchi Marine Science 27: 17–22.
Thomas, L., W. Jason Kennington, M. Stat, S. P. Wilkinson, J. T. Kool, and G. A. Kendrick. 2015. Isolation by resistance across a complex coral reef seascape. Proceedings of the Royal Society B: Biological Sciences 282. https://doi.org/10.1098/rspb.2015.1217.
van Katwijk, M. M., A. Thorhaug, N. Marbà, R. J. Orth, C. M. Duarte, G. A. Kendrick, I. H. J. Althuizen, et al. 2016. Global analysis of seagrass restoration: The importance of large-scale planting. Journal of Applied Ecology 53: 567–578. https://doi.org/10.1111/1365-2664.12562.
Wang, I. J. 2013. Examining the full effects of landscape heterogeneity on spatial genetic variation: A multiple matrix regression approach for quantifying geographic and ecological isolation. Evolution 67: 3403–3411. https://doi.org/10.1111/evo.12134.
Watanabe, K., T. Tokoro, H. Moki, T. Kuwae. 2024. Contribution of marine macrophytes to pCO₂ and DOC variations in human-impacted coastal waters. Biogeochemistry. https://doi.org/10.1007/s10533-024-01140-4
Waycott, M., C. M. Duarte, T. J. B. Carruthers, R. J. Orth, W. C. Dennison, S. Olyarnik, A. Calladine, et al. 2009. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proceedings of the National Academy of Sciences of the United States of America 106: 12377–12381.
Weir, B. S., and J. Goudet. 2017. A unified characterization of population structure and relatedness. Genetics 206: 2085–2103. https://doi.org/10.1534/genetics.116.198424.
Yamakita, T., and M. Nakaoka. 2009. Scale dependency in seagrass dynamics: How does the neighboring effect vary with grain of observation? Population Ecology 51: 33–40.
Yamakita, T., K. Watanabe, and M. Nakaoka. 2011. Asynchronous local dynamics contributes to stability of a seagrass bed in Tokyo Bay. Ecography 34: 519–528.
Yoshida, G. 2012. Relationship of seaweed/seagrass bed and fisheries production of the Seto Inland Sea. Fisheries Engineering 49: 77–83. in Japanese with English abstract
Yu, L., M. Khachaturyan, M. Matschiner, A. Healey, D. Bauer, B. Cameron, M. Cusson, et al. 2023. Ocean current patterns drive the worldwide colonization of eelgrass (Zostera marina). Nature Plants 9: 1207–1220. https://doi.org/10.1038/s41477-023-01464-3.

02supplementalinfo05r.docx

Download PDF

Reviewers agreed at journal
15 Jul, 2024
Reviewers invited by journal
15 Jul, 2024
Editor invited by journal
10 Jul, 2024
Editor assigned by journal
10 Jul, 2024
First submitted to journal
09 Jul, 2024

You are reading this latest preprint version

Genome-wide markers reveal geographical isolation by distance and barriers as well as local heterogeneity in the genetic structure of a seagrass

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Study areas and sampling method

Genetic analysis

SNP filtering

Analysis of genetic structure

Results

IBD and IBR

Genetic structure in the SIS

Genetic structure in Tokyo Bay

Genetic structure at Akkeshi

Discussion

Declarations

References

Supplementary Files

Status:

Version 1