ORIGINAL PAPER Genetic structure and gene flow of eelgrass Zostera marina populations in Tokyo Bay, Japan: implications for their restoration Norio Tanaka • Teruko Demise • Mitsuhiro Ishii • Yasumasa Shoji • Masahiro Nakaoka Received: 24 January 2010 / Accepted: 22 December 2010 / Published online: 8 January 2011 Ó Springer-Verlag 2011 Abstract Massive losses of eelgrass Zostera marina beds in Japan have occurred over the past 100 years. Toward their restoration, transplantation of eelgrass has been attempted in some areas, including Tokyo Bay. This study examined population genetic structures and gene flow in eelgrass in Tokyo Bay to establish guidelines for con- ducting restoration. Genotypes of a total of 360 individuals from 12 beds were determined using five microsatellite markers. The eelgrass beds in inner bay had above-average genetic diversity. A neighbor-joining tree based on F ST values among beds revealed that a strong gene flow had occurred among six beds in the inner bay. Genetic assignment testing of drifting shoots indicated that those with seeds migrate in both directions between the inner and outer bay. We suggested that the restoration of eelgrass in the innermost part of Tokyo Bay, where natural habitats have been lost, should be conducted using the inner bay beds. Introduction Eelgrass Zostera marina L. is a very widespread species of seagrass occurring in temperate to subarctic coastal areas of the Northern Hemisphere (Den Hartog 1970). Due to their high primary productivity and provision of habitats for a wide variety of associated fauna, eelgrass beds are regarded as one of the most important components of coastal ecosystems (Kikuchi and Pe ´re `s 1980; Jernakoff et al. 1996; Duarte and Chiscano 1999; Hemminga and Duarte 2000; Williams and Heck 2001), although they are becoming more scarce worldwide due to both direct and indirect effects of human activity (Fortes 1988; Shorts and Wyllie-Echeverria 1996; Duarte 2002; Moore and Jarvis 2008; Rueda et al. 2009; Waycott et al. 2009; Martin et al. 2010). Among various efforts for conservation and resto- ration of degraded eelgrass beds, artificial transplantation of eelgrass has been attempted in various regions (e.g., Zimmerman et al. 1995; van Katwijk 2000; Orth et al. 2006). However, some problems associated with artificial transplantation have also become apparent. One of these concerns is that transplantation from remote areas may violate the genetic composition of regional populations. To overcome this, assessment of the genetic composition and gene flow of wild eelgrass populations are required before the initiation of artificial transplantation projects. Recent developments in various genetic markers for seagrasses have advanced progress in molecular studies of phylogeny, biogeography, and population genetics (Reusch 2000a, c; Olsen et al. 2004; Waycott et al. 2006). In the case of eelgrass, development of microsatellite markers has led to improved understanding of genetic structures at both global and local spatial scales (Reusch 2001; Tanaka et al. 2002; Olsen et al. 2004). Intensive research in both Europe and North America has revealed that the genetic clonal Communicated by T. Reusch. N. Tanaka (&) Tsukuba Botanical Garden, National Museum of Nature and Science, 4-1-1, Amakubo, Tsukuba, Ibaraki 305-0005, Japan e-mail: ntanaka@kahaku.go.jp T. Demise Graduate School of Science, Chiba University, 1-33, Yayoicho, Inage, Chiba 263-8522, Japan M. Ishii Á Y. Shoji Chiba Prefectural Fisheries Research Center, 2492, Kotsubo, Futtsu, Chiba 293-0042, Japan M. Nakaoka Akkeshi Marine Station, Field Science Center for Northern Biosphere, Hokkaido University, Aikappu, Akkeshi, Hokkaido 088-1113, Japan 123 Mar Biol (2011) 158:871–882 DOI 10.1007/s00227-010-1614-2