The genetic variability of four species of the genus Zoysia collected from South Korea was analyzed using an inter-simple sequence repeat (ISSR) marker system. Polymerase chain reactions (PCR) with eight ISSR primers generated 86 amplicons, 76 (87.1%) of which were polymorphisms. The polymorphism information content (PIC) value of the ISSR marker system was 0.848. The percentage of polymorphic loci ( P _p ) ranged from 41.2% to 44.7%. Nei’s gene diversity ( H ) ranged from 0.149 to 0.186, with an average overall value of 0.170. The mean of Shannon’s information index ( I ) value was 0.250. Total genetic diversity values ( H _T ) varied between 0.356 (ISSR-1) and 0.418 (ISSR-16), for an average overall polymorphic loci of 0.345. Interlocus variation in within-species genetic diversity ( H _S ) was low (0.170). On a per-locus basis, the proportion of total genetic variation due to differences among species (G _ST ) was 0.601. This indicated that about 60.1% of the total variation was among species. Thus, about 39.9 of genetic variation was within species. The estimate of gene flow, based on G _ST , was very low among species of the genus Zoysia ( N _m = 0.332). The phylogenic tree showed three distinct groups: Z. macrostachya and Z. tenuifolia clades and other species were formed the separated clusters. In conclusion, the ISSR assay was useful for detecting genetic variation in the genus Zoysia , and its discriminatory power was comparable to that of other genotyping tools. The genus Zoysia Willd. (Family Poaceae , subfamily Chloridoideae , tribe Zoysiece ) includes about ten species [7] and is commonly known as zoysiagrass ( Zoysia sp.) or lawn. Zoysiagrass is found throughout East Asia (China, Korea and Japan) and Australasia, grown especially in warm regions [17 , 27] . Wild species belonging genus Zoysia have been survived though long-term natural selection [13] , and thereby can tolerate wide variations in temperature, sunlight, and water. Zoysiagrass is widely used for turfgass and forage grass in as Korea and other countries in East Asia. Zoysia grasses can diminish soil erosion on slopes, and are excellent at repelling weeds throughout the year. Some types of Zoysia are available commercially as making lawn ground, park grass, and garden grass. In addition, they are used on golf courses to create fairways and teeing areas after several hybridization processes. DNA markers have numerous applications in plant molecular genetic research [6] . The two most common uses of DNA markers have been the assessment of genetic diversity within plant germplasm and evolutionary relationship within or among species [12 , 19] . One major use of DNA techniques is to reveal genetic diversity within and between populations or species. Inter simple sequence repeat (ISSR) technique is a polymerase chain reaction (PCR) based technique, reported by Zietkiewicz et al. [34] . ISSR technique involves PCR amplification of DNA using a single primer composed of a microsatellite sequence. ISSR technique involves amplification of DNA segments between two identical microsatellite repeat regions oriented in opposite direction using primers designed from microsatellite core regions [29] . The ISSR has mild technical difficulty, good reproducibility and reasonable cost, permitting its use for genetic studies of population. PCR products were resolved by performing agarose gel electrophoresis. Compared to arbitrary markers such as random amplified polymorphic DNA (RAPD), ISSR markers are highly reproducible due to the use of longer primers. Considering the potentials of the ISSR marker based genetic diversity analysis, the present study aimed to evaluate the extent of genetic diversity and phylogenetic relationships among four species of genus Zoysia collected from Korea. Four species within genus Zoysia Willd. occur in Korea. Zoysia japonica Steud., Z. macrostachya Franch. et Savat., Z. sinica Hance, and Z. tenuifolia Willd. were selected. Plant materials collected from four large natural populations per species. To avoid including individuals from the same lineage by clonal reproduction, the distance between the selected individuals was about 5.0 m. Arundinella hirta (Thunb.) Koidz. was used as an outgroup species in this study. The total genomic DNA was extracted from the leaf tissues using the plant DNA Zol Reagent (Life Technologies Inc., Grand Island, New York, USA) according to the manufacturer’s protocol. DNA quantity was checked using a mini fluorometer (TKO 100 Mini-Fluorometer, Hoefer Scientific Instruments). ISSR primers (University of British Columbia, Canada) synthesized by Sigma Aldrich Inc., were used for the polymorphism survey ( Table 1 ). PCR (polymerase chain reaction) amplification was carried out in 25 μl reaction containing 25 ng of genomic DNA, 1 μM of primer, 10 mM of dNTPs (2.5 mM each), 50 mM KCl, 10 mM Tris HCl (pH 8.3), 2.5 mM MgCl2, 0.3 units (U) of Taq DNA polymerase (Promega, USA). PCR conditions were programmed for initial denaturation at 94℃ for 5 min, 36 cycles of 1 min denaturation at 94℃, 45 sec annealing at 42-50℃, 2 min extension at 72℃, and final extension for 10 min at 72℃. The annealing temperature for PCR amplification was maintained based on the specificity of the primer pair used. PCR amplified products of ISSR primers were subject to horizontal gel electrophoresis using 2.0% agarose gel and stained with ethidium bromide. The bands on gels were documented using Alpha Imager 1200TM (Alpha Innotech Co., USA). All the reactions were repeated twice and only reproducible bands were scored for analyses. The unambiguous ISSR bands were scored visually as present (1) or absent (0) and the binary data matrix was constructed. Several standard genetic parameters, the percentage of polymorphic loci ( P _p ), mean number of alleles per locus ( A ), effective number of alleles per locus ( A _E ), Nei’s [21] gene diversity ( H ), and Shannon’s Information index ( I ) were estimated using the computer program, POPGENE ver. 1.31 [33] . Polymorphism information content (PIC) value was calculated using the formula PIC, 1 - Σ pi ² , where pi is the frequency of the i th allele [25] . The estimation of genetic similarity (GS) between genotypes was based on the probability that an amplified fragment from one individual will also be present in another [23] . GS was converted to genetic distance (1-GS) [15] . Species differentiation analyses have been assessed within and among according to Nei's gene diversity formulae ( H _T , H _S , and G _ST ) [22] . The G _ST coefficient corresponds to the relative amount of differentiation among populations or species. Furthermore, gene flow ( N _m ) between the pairs of species was calculated from G _ST values by Nm = 0.5(1/ G _ST -1) [20] . Shannon’s index of genotypic diversity ( H _O ) for ISSR was estimated to quantity the degree of within species diversity following the formula [3] : H _O = –Σ pi log pi , where pi is the frequency of a particular phenotype i . H _O can be calculated and compared for different populations [24] . Let H _POP = 1/n H _O be the average diversity over the n different populations and let H _SP =–Σp log p be the diversity of species calculated from the phenotypic frequencies p in all the species considered together [24] . Then the proportion of diversity within species, H _POP / H _SP , can compared with that between species ( H _SP - H _POP )/ H _SP . A phenetic relationship was constructed by the neighbor joining (NJ) method using the NEIGHBOR program in MEGA5 [28] . For analysis of variability of genus Zoysia , eight primers were used for studying the ISSR banding patterns across the entire samples. The eight primers generated a total of 85 consistently scorable bands, 76 of which were polymorphic (89.4% polymorphism) ( Table 1 ). The average number of amplification products was 10.6 per primer; the maximum was 15 with (GT) ₈ YT (ISSR-12), whereas the minimum was 8 with (GA) ₈ CG (ISSR-5). Polymorphism information content (PIC) for ISSR primers ranged from 0.791 to 0.896 with an average of 0.848 per primer ( Table 1 ). In a simple measure of inter-populations variability i.e. the percentage of polymorphic bands ( P _p ), Z. japonica exhibited the highest variation (47.1%) and Z. tenuifolia the lowest (41.2%) ( Table 2 ). The average number of alleles per locus ( A ) was 1.439 across species, varying from 1.412 to 1.471. The effective numbers of alleles per locus ( A _E ) at the lowest species and the highest species level were 1.253 and 1.334, respectively. The mean genetic diversity within species was 0.170. Overall, Z. sinica and Z. japonica exhibited high variation among species. Z. macrostachya and Z. tenuifolia was shown the low genetic variation. Total genetic diversity values ( H _T ) varied between 0.356 (ISSR-1) and 0.418 (ISSR-16), for an average over all polymorphic loci of 0.345 ( Table 3 ). Interlocus variation in the within-species genetic diversity ( H _S ) was low (0.170). On a per-locus basis, the proportion of total genetic variation due to differences among species ( G _ST ) ranged from 0.273 for ISSR-1 to 0.660 for ISSR-6, with a mean of 0.506. This indicated indicated that about 50.6% of the total variation was among species. Thus, about genetic variation (40.4%) resided within species. The estimate of gene flow, based on G _ST , was very low among species of genus Zoysia ( N _m = 0.488). Values of genetic distance (D) were <0.470 ( Table 4 ). Genetic identity values among pairs of populations ranged from 0.625 to 0.813. Genetic distances between species were high. There was not shown significant difference among four species. An assessment of the proportion of diversity present within species, H _POP / H _SP , indicated that about 71.0% the total genetic diversity was among species. Thus, the other portion of genetic variation (29.0%) resided within species ( Table 5 ). Clustering of four species of genus Zoysia , using the NJ algorithm, was performed based on the matrix of calculated distances ( Fig. 1 ). Four species of genus Zoysia were well separated each other. The phylogenic tree showed three distinct groups; Z. macrostachya and Z. tenuifolia clade and the other species. Genetic diversity of Korean Zoysia species is comparable with other species, although the use of different methods (e.g., isozyme [co-dominant marker] and RAPD [dominant marker], ISSR [sequence based marker system) may preclude meaningful direct comparisons. In ISSR analysis, nine species belonging to genus Zoysia maintain moderate or higher than mean level of genetic diversity compared with other plant species [10] . For example, the percentage of polymorphic loci at the species level for Zoysia is 43.9%, which is similar to species with temperature- zone distributions (48.5%), species with a sexual and asexual reproduction mode (43.8%), and those with a short-lived perennial herbaceous (41.3%) (Hamrick and Godt 1989). Its genetic diversity of 0.170 is higher than that for temperate-zone species (0.146), species with a sexual and asexual reproduction mode (0.138), and those with a short-lived perennial herbaceous (0.116) [10] . The results of other research for Zoysia were similar to the results of this study. Choi et al. [5] analyzed 93 native zoysiagrass lines collected from the southern and western coastal regions of South Korea. They also estimated genetic diversity of zoysiagrass using RAPD [4] . Weng et al. [30] also analyzed genetic diversity of 131 clones of Taiwan Zoysia species by using isozyme and RAPD. Li and Tong [16] reported genetic diversity of 105 plants from seven Z. sinica populations in different regions of China using ten RAPD primers. About 4.8, 30, and 70% of the genetic variation was detected among groups, populations, and within populations, respectively. In ISSR literature, polymorphism and genetic diversity ( H ) within species in Chinese zoysiagrass were 96.7% and 0.250, respectively [31] . The relatively high level of genetic variation found in genus Zoysia is consistent with several aspects of its biology. First, the particular breeding system that a plant possesses is an important determinant of variability at both the species and population levels. Flowering plants exhibit a spectacular diversity in reproductive systems, and this can have important effects on the amount and structuring of genetic variation within and among populations [9] . Species of the genus Zoysia are rarely reproduced sexually and pertetuates theyselves clonally by rhizomes [1] . Whereas sexual reproduction might initially act to enhance genetic variation, asexual reproduction can maintain this enhanced variability [2] . Second, a perennial and/or long-lived species generally maintains relatively higher levels of variation than do annuals [18] . Because individuals of genus Zoysia are so longlived, opportunities for the accumulation of mutations should be high [14] . Genetic differentiation among populations is principally a function of natural selection, genetic drift, and gene flow among populations via pollen and seed dispersal [26] . Of the total variation observed in Zoysia species about 20.0-27.7% was due to differences among species ( G _ST = 0.601, H _VAR )/ H _SP = 0.710). Predominantly wind-pollinated outcrossing species have on an average less than 10 % of the genetic variation between populations [11] . This high level of genetic differentiation also suggests that gene flow among species is low ( Nm = 0.332), though no chromosomal barrier of interspecific crossing [8 , 32] .