Introduction

Turtles are among the most at-risk groups of vertebrates, even more so than birds, mammals, fish, or the similarly endangered amphibians1. The critical threats facing turtles globally include familiar issues affecting other species, such as habitat loss, unsustainable hunting, and climate change, which affects them particularly since many turtles rely on environmental factors for sex determination. Prior to the human impact on their populations, turtles were characterized by their large numbers and the high biomass they contributed to ecosystems, making them some of the most abundant animals. Because of their large populations, turtles played significant roles as soil bioturbators, sea floor diggers, seed dispersers and enhancers of germination, nutrient recyclers, and consumers2. The global decline of turtle populations has significantly weakened their ecological functions3. Fossil reports indicate that sea turtles have belonged to five families, of which three families, including Protostegidae, Toxochelyidae, and Plesiochelyidae have been extinct since the Jurassic period, while two families, Cheloniidae with seven species, and Dermochelyidae with one species, have survived to the present4.

Among the eight living species, five species, including the green turtle (Chelonia mydas; Linnaeus, 1758), hawksbill turtle (Eretmochelys imbricata; Linnaeus, 1766), loggerhead turtle (Caretta caretta; Linnaeus, 1758), olive ridley turtle (Lepidochelys olivacea; Eschscholtz, 1829) from the Cheloniidae, and leatherback turtle (Dermochelys coriacea; Linnaeus, 1766) from the Dermochelyidae, have been observed in the Persian Gulf5. Among these species, nesting of the hawksbill, green, and olive ridley turtles in the Persian Gulf and the Sea of Oman has been reported, with the highest nesting related to the hawksbill and green turtles6,7,8.

Hawksbill turtles usually have smaller bodies compared to other species and undergo long migrations between foraging and nesting areas, and in some cases, their nesting migrations can extend to several thousand kilometers9. In recent years, the population density of hawksbill turtles worldwide has decreased, and since 2008, they have been endangered species and listed in the International Union for Conservation of Nature (IUCN) Red List10. Multiple factors contribute to a decline in sea turtle density, with the most critical being the destruction of nesting habitats by human activities, fishing and hunting, oil and industrial pollution, tourism, and vessel collisions. Alongside anthropogenic factors, natural elements such as predation of eggs and hatchlings by mammals like jackals and foxes, as well as seabirds, nesting site destruction by waves, and annual beach erosion also play significant roles11,12,13.

This species is distributed across the northern and southern coasts of the Persian Gulf, with the highest populations concentrated in the northern Iranian coastal regions. It has been recorded in various locations in Iran, including the Kish, Larak, Hormoz, Shidvar, Qeshm, Abu Musa, Nakhiloo, Ommolgorm, Bani Farur, Hengam, and others. There have also been reports of sightings along Iran’s coasts of the Oman sea, particularly in southern Sistan and Baluchestan. Key nesting sites on the southern coast of the Persian Gulf are primarily located in the United Arab Emirates, on islands such as Koryan, Hargus, Jumah, and Karan. Additionally, nesting grounds have been identified in the Sea of Oman at Ras al-Hadd. Researches indicate that the coastal habitats of Australia and the Persian Gulf are among the most significant nesting areas for hawksbill turtles globally, with the Iranian coasts being particularly vital for this species6,12,14,15.

In recent years, the use of modern molecular genetics technologies has gained attention as an appropriate method for gathering information about the characteristics of populations15. One of the conditions that poses a risk to a population is the effective population size16. If it is low, or if the number of males or females is insufficient, it can lead to increased inbreeding and decreased genetic diversity within the population15. Genetic diversity encompasses any type of variation at the nucleotide, gene, chromosome, or whole genome level of an organism. Diversity can be evaluated within species and between populations17. Various methods have been used to assess genetic diversity to date. In recent years, genetic markers, particularly DNA-based markers, have become a suitable and reliable tool for studies related to genetic diversity and population genetics18.

Among molecular markers, Inter Simple Sequence Repeat (ISSR) markers amplify DNA fragments located between two repetitive microsatellite regions19. Similar to Random Amplified Polymorphic DNA (RAPD), ISSR is a dominant marker; however, it offers higher reproducibility and variability than RAPD, along with greater speed and ease of use20. These markers require minimal template DNA21. Due to their longer fragment length, ISSR markers exhibit reproducibility comparable to Simple Sequence Repeat (SSR) microsatellites19. This technique effectively combines the advantages of microsatellites while integrating features of Amplified Fragment Length Polymorphism (AFLP) and RAPD methods22. ISSR primers are non-specific, and the technique is simple, rapid, and does not require radioactive materials or the cost-intensive construction of a genomic library23. These markers have been widely used to assess genetic diversity in various plants and animals, including cattle, sheep, goats, fish, and honeybees19,20,21,22,23,24. ISSR-PCR is a DNA amplification method that uses primers designed based on repeat sequences of two, three, or four nucleotides, and therefore complements microsatellites24. ISSR-PCR utilizes the abundance of microsatellites in the genome. When two microsatellites are located within an amplifiable distance and are in reverse positions, a primer initiates the amplification of the DNA fragment between them25.

Since microsatellites are located at different points in the genome, various bands of different sizes are produced on the resulting gel. Scoring for ISSR is done as zero (absence of a band) and one (presence of a band). The result of this scoring is a binary matrix that is used to calculate other genetic variables such as genetic distance, heterozygosity, dendrogram of genetic similarity, and other analyzable variables26,27.

Researchers have classified hawksbill turtles in Iranian waters of the Persian Gulf based on their nesting location and have assigned a nesting group to each turtle according to site names such as Nakhiloo (N), Kharkoo (K), Ommolgorm (O), and Shidvar (S) Islands12,14,28. Numerous studies have examined the reproductive biology and important nesting and feeding regions of E. imbricata in the Persian Gulf. In addition, several genetic investigations using microsatellite and mitochondrial markers have revealed population structure and diversity patterns in the Persian Gulf and across the Indo-Pacific6,29,30,31,32,33,34. However, ISSR markers have not previously been applied to sea turtles. Our study therefore provides the first assessment of hawksbill turtles using ISSRs, offering a cost-effective and accessible approach that can serve as a foundation for future low-cost genetic monitoring in resource-limited regions. This work contributes new information on the genetic diversity and population connectivity of nesting groups in Iran, complementing earlier microsatellite and mtDNA studies.

Results

Utilizing six primers, various fragments were successfully amplified (Fig. 1). One of the markers, (AG)8C, was monomorphic and revealed a consistent pattern of monomorphic bands across all individuals analyzed. This uniformity indicates that all individuals share the same allele for this specific primer, while the other five markers were polymorphic. The highest percentage of polymorphism was observed for the marker (GACA)4 (70%), while the marker (GA)8AC had the lowest polymorphism (29%). Additional gel images of ISSR amplifications and the binary scoring matrix (0/1) for all individuals and markers are provided in the Supplementary Data. AMOVA analysis showed that 71% of the genetic variation was related to within-population diversity, and 29% was related to between-population diversity (Table 1). The results indicated that the percentage of polymorphic loci for the nesting groups Kharkoo, Nakhiloo, Ommolgorm, and Shidvar were 0%, 4.35%, 8.7%, and 30.43%, respectively. Other polymorphism indices, such as Shannon’s diversity index, confirmed the estimated diversity (Table 2).

Fig. 1
figure 1

Agarose gel example for the (AG)8G ISSR marker amplification, comparing the 14 hawksbill turtle individuals of the nesting groups Kharkoo (K1–K3), Nakhiloo (N11–N14), Ommolgorm (O1–O6), and Shidvar (S1–S4) of the Persian Gulf. The 50 bp ladder is also shown (M50).

Full size image
Table 1 AMOVA, based on six ISSR markers, comparing Hawksbill turtle individuals from the nesting groups Kharkoo, Nakhiloo, Ommolgorm, and Shidvar of the Persian Gulf.
Full size table
Table 2 Genetic diversity metrics estimated from six ISSR markers in 14 Hawksbill turtles, comparing nesting groups of the Persian Gulf.
Full size table

The dendrogram of genetic similarity (Fig. 2) presented strong clustering support based on Dice dissimilarity distances. The analysis distinguished all four nesting groups, with the largest separation observed for Shidvar individuals, while Kharkoo, Nakhiloo, and Ommolgorm formed distinct but closer clusters. This pattern indicates that each nesting group represents a genetically distinct population.

Fig. 2
figure 2

Dendrogram of genetic similarity among 14 hawksbill turtle individuals from four nesting groups in the Persian Gulf: Kharkoo (K1–K3), Nakhiloo (N11–N14), Ommolgorm (O1–O6), and Shidvar (S1–S4). The dendrogram was constructed using ISSR markers and Dice dissimilarity distances. The vertical scale represents genetic distance, with greater height indicating lower genetic similarity between individuals or groups.

Full size image

Contrasting with the dendrogram, the principal coordinate analysis (PCoA) revealed no clear genetic pattern division, although some clustering was visible and consistent with the dendrogram results. The first two axes of the principal coordinate analysis (PCoA) explained 48.44% of the total genetic variation among individuals sampled from the four nesting groups (PCoA1: 26.23%; PCoA2: 22.21%) (Fig. 3). The results from STRUCTURE HARVESTER indicated that the highest ∆K value corresponded to K = 2 (Fig. 4). This informed the population structure analysis (Fig. 5), which identified two sub-populations with distinct genetic compositions (green, and red). As shown in Fig. 5, the Kharkoo and Shidvar nesting groups clustered within the same sub-population, and were also positioned closely in the lower right quadrant of the PCoA plot. In contrast, the Nakhiloo and Ommolgorm nesting groups formed separate genetic clusters. The sub-population assignments based on the kinship matrix (q) were consistent with the principal component analysis results.

Genetic differentiation (Fixation index, FST) among four nesting groups of hawksbill turtles, based on 14 individuals evaluated using six ISSR markers, is presented in Table 3.

Fig. 3
figure 3

Principal Coordinate Analysis, based on the amplification of six ISSR markers from 14 hawksbill turtles, comparing the nesting groups Kharkoo (K), Nakhiloo (N), Ommolgorm (O), and Shidvar (S) of the Persian Gulf.

Full size image
Fig. 4
figure 4

The graph depicting ∆K versus K was created using the Evanno method to identify the optimal K values for 14 individuals studied using six ISSR markers.

Full size image
Fig. 5
figure 5

Bar chart obtained from the population structure analysis of 14 individuals studied from Kharkoo (K1–K3), Nakhiloo (N11–N14), Ommolgorm (O1–O6), and Shidvar (S1–S4) hawksbill turtle nesting groups of the Persian Gulf, based on the kinship matrix.

Full size image

Discussion

Although our sampling size was relatively small (14 hatchlings collected from dead individuals during a single nesting season) this study provides useful insights into the genetic structure of hawksbill turtles (Eretmochelys imbricata) and their nesting groups in the northern Persian Gulf. We introduced five ISSR markers and used them to assess polymorphism, genetic diversity, and genotyping in these endangered turtles. ISSR markers are simpler and faster, and because they are generally more cost-effective than many other genetic tools, they can be particularly useful for genetic monitoring in regions with limited resources19,20,21,22,23,24,25. Given the constraints, we believe the results are informative for conservation planning.

Table 3 Genetic differentiation (Fixation index, FST ) among the Kharkoo, Nakhiloo, Ommolgorm, and Shidvar Hawksbill turtle nesting groups, based on 14 individuals evaluated using six ISSR markers.
Full size table

All ISSR primers amplified in our samples and produced clear bands. One primer, (AG)8G, was monomorphic; the remaining primers were polymorphic, which supports their utility for assessing diversity in this species. Although overall genetic similarity among individuals was high, turtles from different nesting habitats nonetheless grouped as distinct populations in our analyses. The dendrogram showed population-level clustering, with nesting groups forming separate branches. This general pattern is consistent with previous work in the Persian Gulf: several studies using microsatellites, SSRs, and mitochondrial markers have reported population-level differentiation despite low overall diversity6,35,36,37. Comparable fine-scale structuring has also been observed in hawksbill populations elsewhere, for example in Indonesia33 and Brazil34, suggesting that localized conservation measures are often warranted.

Our geographic analyses (dendrogram and PCoA) revealed two main clusters. Individuals from Ommolgorm and Nakhiloo (central Persian Gulf) grouped together, while Shidvar and Kharkoo (eastern and western sites) formed a separate cluster. Several, not mutually exclusive, explanations could account for this pattern: natal homing behavior, oceanographic features that limit gene flow, or historical connectivity and demographic events. The largest genetic split (FST) was observed between Shidvar and Kharkoo, indicating pronounced separation despite the geographic arrangement. That said, some branches in the dendrogram had limited statistical support, so we avoid overinterpreting every split. Overall, however, the geographic pattern we observe aligns with earlier studies reporting fine-scale differentiation among nearby nesting sites.

These genetic patterns are further contextualized by movement ecology. Satellite tracking by Pilcher et al. (2014) showed that Persian Gulf hawksbills spend only short periods at nesting sites and much longer periods at distant foraging grounds9. On average, turtles traveled substantial distances to foraging areas (mean 189 km), with Iranian turtles showing the longest migrations. Despite seasonal movements, individuals tended to show fidelity to particular foraging sites (notably around Abu Dhabi, southern Qatar, and the northern coasts of Bahrain, Saudi Arabia, and Kuwait). Taken together, natal homing plus site fidelity at foraging grounds can produce the kind of localized genetic structure we detected.

Our results are consistent with the idea that female hawksbills return to natal beaches to nest (Archie Carr, 1967)38. If so, each nesting habitat can act as a repository of particular genetic combinations. Loss or degradation of any nesting site could therefore remove unique genotypes and raise extinction risk. For that reason, protecting individual nesting habitats should be a conservation priority, even when overall genetic diversity appears low10,11,12,28.

In conclusion, molecular markers have greatly advanced population research. ISSR markers, which often show high polymorphism, are useful tools for investigating genetic diversity. To obtain more complete and robust results, future work should include a broader set of ISSR primers and, where possible, larger sample sizes. Understanding genetic relationships among populations is essential for effective management; ISSR markers can help identify genotypes and assess diversity at the molecular level, information that is important for long-term conservation and genetic management. Given their relative simplicity, speed, and cost-effectiveness, ISSR markers are a practical option for genetic monitoring in regions with limited resources for species management programs.

Methods

Sampling sites

The Persian Gulf is a shallow, semi-enclosed sea located in the semi-tropical region of the northwest Indian Ocean. It connects to the open waters of the Sea of Oman and the Indian Ocean through the Strait of Hormuz. The main currents enter the Gulf via the Strait of Hormuz, curve around the Iranian coastline at the northern end, and exit through the seabed of the Strait after passing along the coasts of the surrounding Arab countries. The Gulf spans approximately 251,000 square kilometers, with a length of 1,000 km and a width ranging from 200 to 300 km. Its geographic coordinates lie between 24°N to 30°N latitude and 48°E to 57°E longitude39.

Hawksbill turtle nesting sites have been reported across various habitats within the Persian Gulf6,7,8. In this study, sampling was conducted during the nesting season (March to June), late at night until before sunrise, when hatchlings were emerging from their nests. Three to four nests (one sample per nest) were sampled on four islands. Nests with emerging hatchlings were examined, and dead embryos were collected immediately. Sampling was conducted at four nesting habitats of the hawksbill turtle: Nakhiloo (N), Kharkoo (K), Ommolgorm (O), and Shidvar (S) Islands which was shown in Fig. 6). Nakhiloo and Ommolgorm Islands are part of the Deir-Nakhiloo National Park, located in the delta of the Mond River. Shidvar Island is internationally recognized for its seabird nesting habitats and coral reefs. All of these areas have been designated as protected marine zones by Iran’s Department of Environment (DOE)40.

Fig. 6
figure 6

Map of locations along the northern shores of the Persian Gulf, Iran, where sampling of hawksbill turtles (Eretmochelys imbricata) has taken place (S, Shidvar Island; O, Ommolgorm Island; N, Nakhiloo Island; K, Kharkoo Island). The map produced in https://www.simplemappr.net/ and modified in Adobe Photoshop 2022 v23.5.4.981.

Full size image

This study was conducted in accordance with the standards of the Ethics Committee of Shahid Bahonar University of Kerman, and all experimental protocols and sampling techniques were approved by this committee. In addition, the ARRIVE 2.0 guidelines (https://arriveguid.elines.org/) were followed in conducting this study. Since dead turtle hatchlings were used in this study, no turtles were killed or injured. Persian Gulf Mobin Energy Company and Shahid Bahonar University of Kerman have relevant laws and regulations that were followed in this research. All sampling procedures involving wild sea turtles were conducted in accordance with relevant national and institutional guidelines and regulations. Sampling permission was obtained from the Department of Environment of Bushehr Province under permit number 2/1403/1224.

DNA extraction

Sampling was conducted in 2022, during the hatching season. Fourteen dead hawksbill turtle hatchlings were collected from different nests across four nesting groups. Six samples were obtained from Nakhiloo (N = 3) and Kharkoo (N = 3), while four samples each were collected from Ommolgorm and Shidvar. After packaging and labeling, the samples were stored on ice and transported to the research laboratory.

A portion of each sample’s limb tissue was separated and finely powdered in liquid nitrogen. The powdered tissue was lysed with 500 µL of lysis buffer containing proteinase K (200 mg/mL), 100 mM NaCl, 10 mM Tris-HCl, 13 mM EDTA, 35 mM SDS, and 3.5 mM sodium citrate. Total DNA was extracted using chloroform41.

The quantity and quality of the extracted DNA were assessed by spectrophotometry using the NanoDrop Onec device, and by electrophoresis on a 1% agarose gel using the Owl™ DuoGel™ Electrophoresis System and Gel Doc™ EZ Imager System. The 260/280 absorbance ratio was 1.9, indicating high purity and good integrity of the DNA on agarose gel.

Selection of ISSR markers

Six ISSR markers were selected for this study: (AG)8C, (AG)8G, (GA)8AC, (GA)8AG, (GACA)4, and (GTG)5GC19,42. These primers are known to exhibit high levels of polymorphism and genetic diversity. Detailed characteristics of the selected primers are presented in Table 4.

Table 4 Characteristics of primers used to study the genetic variation among Hawksbill turtles nesting along the Persian Gulf coastline.
Full size table

Polymerase chain reaction (PCR)

Polymerase chain reactions were performed in a total volume of 20 µL using Taq RED Polymerase (Ampliqon A.S., Odense, Denmark). Each reaction contained 20 ng/µL DNA, 10 µL of PCR master mix (containing Tris-HCl pH 8.5, (NH4)2SO4, 3 mM MgCl2, 0.4 mM of each dNTP, 2× Taq DNA polymerase, inert red dye, and stabilizer), and 2 µL of 10 µM primer. Sterilized distilled water was added to reach a final volume of 20 µL.

The PCR temperature profile was as follows: initial denaturation at 94 °C for 5 min; followed by 45 cycles of 94 °C for 50 s (denaturation), varying annealing temperatures for each marker (Table 4) for 1 min, and 72 °C for 2 min (extension). A final extension step was performed at 72 °C for 5 min. Negative controls were included to verify the absence of contamination during the PCR procedure.

Amplified products were electrophoresed in a 2.5% agarose gel containing 1% TBE buffer at 90 V for 2 to 2.5 h, alongside a 50 bp DNA ladder (Cina Gen Co., Iran). Gels were stained with ethidium bromide and visualized under UV light using a gel documentation system.

Data analysis

The resulting fragments were scored visually, with the ISSR profiles documented as either 1 (presence of bands) or 0 (absence of bands). In ISSR analysis, each band at a specific locus is considered to represent a dominant allele.

To assess genetic structure and diversity within and among populations, a non-parametric analysis of molecular variance (AMOVA)43 was performed, along with principal coordinate analysis (PCoA) based on a pairwise distance matrix using GenAlex 6.501 software44. Key genetic parameters were calculated, including sample size mean (N), standard error (SE), number of average alleles (Na), number of effective alleles (Ne), Shannon’s diversity index (I), expected heterozygosity (He), and unbiased expected heterozygosity (uHe).

A dendrogram of genetic similarity was constructed using Dice dissimilarity coefficients, and a dendrogram was generated using the weighted pair group method with arithmetic mean (WPGMA), as implemented in DARwin 6.0.021 software45. The dendrogram was based on the genetic distance matrix derived from ISSR marker data.

Population structure was inferred using STRUCTURE software version 2.3.446, applying a burn-in period of 10,000 iterations followed by 100,000 iterations of the Markov chain Monte Carlo (MCMC) method under the admixture model. Analyses were conducted for genetic cluster values ranging from K = 2 to K = 9, with each value replicated five times. The optimal number of clusters was determined using the K method proposed by Evanno et al.47, based on the variation in the logarithmic probability of the data across consecutive K values. The K value was calculated using the equation:

$$\Delta K = \frac{{m \cdot \left| {L^{{{\prime \prime }}} (K)} \right|}}{{s\left[ {L(K)} \right]}}$$

where L(K) is the log probability of the data for a given K, L′′(K) is the second-order rate of change of L(K) between consecutive K values, s[L(K)] is the standard deviation of L(K) across replicate runs, and m is a scaling constant (usually set to 1).

STRUCTURE HARVESTER48 was used to visualize and interpret the results.