Introduction

Effective broodstock management is essential for successful artificial reproduction in aquaculture, as it promotes gonadal maturity and gamete quality1,2. Both intrinsic and extrinsic factors, including water quality (e.g., temperature, salinity, dissolved oxygen), nutrition, and sex steroid hormones, significantly influence broodstock maturation, reproduction, and health3,4. Understanding the reproductive physiology, gonadal development stages and associated biomarkers is critical for developing optimized rearing strategies, such as tailored feeding regimes, and environmental adjustments of temperature, salinity or dissolved oxygen to support broodfish conditioning in captivity.

Key indicators of gonadal maturity in female fish include the gonadosomatic index (GSI), oocyte development, and plasma sex steroid hormone levels, particularly estradiol 17β (E2)5,6. An increase in GSI typically reflects gonadal development, while elevated plasma E2 levels signal the onset of sexual maturity, particularly during vitellogenesis7,8,9. These biomarkers enable aquaculturists to assess reproductive status, predict reproductive seasons, and thereby design strategic plans for artificial reproduction and seed production.

Reproductive hormones, notably gonadotropins (follicle-stimulating hormone, FSH; and luteinizing hormone, LH) play a crucial role in regulating gonadal development in fish10,11. FSH and LH stimulate the gonads to synthesize and secrete sex steroids, which, in turn, act on hepatic and gonadal cells to promote oocyte growth and maturation12,13. In female fish, plasma E2 is a key regulator of vitellogenesis, inducing hepatic synthesis of vitellogenin (Vtg), a critical yolk precursor14,15. During gametogenesis, surges in plasma FSH correspond closely with elevated E2 and Vtg levels, driving oocyte development, particularly during vitellogenesis16,17,18. The positive correlation between plasma E2 and vitellogenesis underscores its central role in reproductive physiology19,20, but this relationship has not been investigated in golden trevally.

Golden trevally (Gnathanodon speciosus) is a marine finfish species of high economic value and holds potential for commercial aquaculture in Southeast Asia. However, seed production remains limited, with many hatcheries relying on wild-caught fry due to challenges in artificial breeding21,22,23. In countries like Vietnam, G. speciosus is a priority species for marine aquaculture development, necessitating scalable seed production to reduce dependence on natural stocks. Comprehensive knowledge of its reproductive physiology, endocrinology, ovarian development, and reproductive season is critical for achieving consistent, commercial-scale artificial reproduction. However, such knowledge remains a major gap in seed production of G. speciosus.

This study investigated seasonal changes in GSI, ovarian development, and plasma E2 levels across one annual cycle of female G. speciosus, elucidating the relationship among these parameters. The findings provide robust data on reproductive physiology, reproductive seasonality, and the annual reproductive pattern of golden trevally. These findings contribute to improving broodstock management and enable hatchery operators to develop effective strategies for artificial reproduction and seed production.

Results

Seasonal changes in gonadosomatic index and ovarian development

From January to February, the GSI in female golden trevally was low and ranged between 0.76 ± 0.15–0.85 ± 0.21% with no statistically significant difference between months (P > 0.05). In March, GSI increased significantly to 2.66 ± 0.31% (P < 0.05) and remained elevated, fluctuating between 2.46 ± 0.6–3.05 ± 0.2% from April to October, with no significant differences among these months (P > 0.05). GSI values ​​decreased sharply to 1.12 ± 0.3 in November and further to 0.77 ± 0.3% in December (P < 0.05, Fig. 1).

Fig. 1
Fig. 1
Full size image

Monthly changes in gonadosomatic index (%) in female golden trevally (G. speciosus) during an annual reproductive season. Different letters indicate significant difference (P < 0.05), analyzed using one-way ANOVA, followed by Duncan tests.

GSI varied significantly across ovarian developmental stages (Fig. 2). Stage II ovaries exhibited the lowest GSI (0.86 ± 0.22%), while stage III ovaries showed a marked increase to 2.88 ± 0.68% (P < 0.05). GSI continued to rise gradually in stages IV and V, reaching 2.95 ± 0.8 and 3.08 ± 0.7%, respectively (Fig. 2), with no statistically significant differences among the ovary stages III, IV, and V (P > 0.05).

Fig. 2
Fig. 2
Full size image

Relationships between stages of ovarian development and gonadosomatic index (%) in female golden trevally (G. speciosus). Different letters indicate significant differences (P < 0.05), analyzed using one-way ANOVA, followed by Duncan tests.

Histological analysis revealed distinct seasonal patterns in ovarian development. From November to February, ovaries contained primarily previtellogenic oocytes, including primary and secondary oocytes (Fig. 3A). From March to October, ovaries exhibited asynchronous development, containing previtellogenic, vitellogenic, mature, ovulated oocytes and postovulatory follicles, coexisting, indicating a prolonged reproductive season with multiple spawning events (Fig. 3B-F).

Fig. 3
Fig. 3
Full size image

Histological sections of golden trevally (G. speciosus) ovaries in the annual reproductive cycle showing the presence of oocyte developmental stages. Lu, ovarian lumen; po, primary oocytes; so, secondary oocytes; vo, vitellogenic oocytes; pof, post-ovulated follicle; oo, ovulated oocytes; yg, yolk globules; gv, germinal vesicle; St II, St III, St IV, St V represent ovarian development stages II, III, IV and V respectively. A is representative section of ovaries collected during Nov-Feb. B-F are representative sections of ovaries collected during Mar-Oct. Scale bars: 100 μm.

Seasonal changes in plasma estradiol 17β

Plasma E2 levels were significantly lower (P < 0.05) from January to February, ranging from 156 ± 35 to 210 ± 45 pg/mL. From March, E2 levels increased rapidly to 460 ± 66 pg/mL and remained elevated through October, peaking at 712 ± 98 pg/mL in June. No statistically significant differences in plasma E2 levels were found between April to September (P > 0.05). E2 levels tended to decline in November to 321 ± 55 pg/mL and further to 160 ± 46 pg/mL (P < 0.05) in December (Fig. 4).

Fig. 4
Fig. 4
Full size image

Monthly changes in plasma estradiol levels (pg/mL) in female golden trevally (G. speciosus) during an annual reproductive season. Different letters indicate significant difference (P < 0.05), analyzed using one-way ANOVA, followed by Duncan tests.

Analysis of E2 levels across ovarian development stages revealed a peak of 750 ± 120 pg/mL during stage III (vitellogenesis, Fig. 5). E2 levels were lower in stages IV (670 ± 150 pg/mL) and V (450 ± 110 pg/mL), with no statistically significant differences among stages III, IV and V (P > 0.05). The lowest E2 level (360 ± 90 pg/mL) was observed in stage II (P < 0.05).

Fig. 5
Fig. 5
Full size image

Relationships between stages of ovarian development and estradiol level (pg/mL) in female golden trevally (G. speciosus). Different letters indicate significant differences (P < 0.05), analyzed using one-way ANOVA, followed by Duncan tests.

Discussion

Ovarian development and gonadosomatic index

The gonadosomatic index (GSI) and ovarian development stages are critical indicators of reproductive maturity in teleost fish5,24. In female golden trevally, GSI increased significantly from March to October (2.46–3.05%), coinciding with the reproductive season, and remained low (0.76–0.85%) from November to February during the post-spawning or resting phase. This seasonal pattern reflects active vitellogenesis during the reproductive season, where oocyte growth increases ovarian mass, leading to elevated GSI25,26. The highest GSI values (2.95–3.08%) were observed in ovarian stages IV and V, consistent with peak gonadal development prior to spawning, while the lowest GSI (0.86%) occurred in stage II, indicating minimal gonadal activity. These data indicate that the reproductive season of female golden trevally in captivity extends from March to October.

The asynchronous ovarian development observed histologically, with coexisting previtellogenic, vitellogenic, and mature oocytes from March to October, confirms G. speciosus as an asynchronous and multiple-spawning species with a prolonged reproductive season. This asynchrony explains the sustained high GSI during the reproductive season, as continuous vitellogenesis supports multiple spawning events27. The absence of stage I (found in juvenile fish) and stage VI (immediate post-spawning) ovaries in this study aligns with sampling mature adults and the transient nature of stage VI, characterized by disintegrated oocytes, blood vessels and degenerated tissue25,26,27.

Compared to other tropical marine fish species in the same region, such as rabbitfish (Siganus gusttatus), waigieu seaperch (Psammoperca waigiensis), the GSI value of the golden trevally was lower (maximum 3%), although their mature body size was much bigger than these fish. Waigieu seaperch has the average mature size between 200 and 300 g/fish, the highest GSI value was around 8%6, egg diameter ranged between 735 and 756 μm and absolute fecundity was between 100,000 and 120,000 eggs/female28. In rabbitfish, the average mature size was between 500 and 550 g/fish, the highest GSI value was around 5.8%29, egg diameter ranged between 550 and 580 μm and absolute fecundity was between 570,000 and 625,000 eggs/female30. In golden trevally, the average mature size was between 800 and 1000 g/fish, egg diameter was 328.5 ± 25.6 μm and the absolute fecundity was around 121,000 eggs/female31. This suggests species-specific differences in reproductive investment, potentially influenced by egg size or spawning frequency. Generally, when the GSI is low, the ovary size is small, which can lead to lower fecundity. However, the fecundity could be affected by the size of the egg. From available GSI values, fecundity could be estimated, which can help determine the number of eggs in female fish in each spawn, thereby building appropriate seed production plans.

The histological presence of degenerative oocytes and postovulatory follicles from November to December suggests follicular atresia and resorption of unspawned oocytes. For hatchery management, temporarily reducing feeding post-spawning could facilitate resorption of residual oocytes, promoting new oocyte development and reducing feed costs. Additionally, tailored nutritional regimes targeting vitellogenesis (stages III–IV) could enhance oocyte quality, fertilization rate, and larval viability. Histological evidence confirms that final oocyte maturation is achieved in the ovaries of G. speciosus. The female spawners examined in this study, with mean body weights of approximately 900 g and total lengths of 40 cm, thus represent the size at sexual maturity. However, further research is needed to establish the size at first sexual maturation.

Environmental factors, particularly water temperature, significantly influence reproductive physiology. In Southeast Asian countries like Vietnam, higher temperatures (26–30 °C) from March to October likely accelerated metabolism and vitellogenesis, promoting gonadal maturation. In contrast, lower temperatures (20–25 °C) from November to February corresponded with reproductive quiescence32. Similar temperature-driven reproductive patterns are reported in other tropical fish species, including rabbitfish29, waigieu seaperch6, and longfin batfish Platax teira33. It appears that water temperature and seasonal changes influenced gonadal development, while photoperiod in the study area varied little across the year. These findings underscore the importance of aligning broodstock management with seasonal environmental cues to optimize spawning outcomes. Fisheries management could also leverage this knowledge to enforce seasonal fishing restrictions, protecting G. speciosus during its reproductive season to support sustainable populations.

Relationship between plasma E2 and ovarian development

In G. speciosus plasma E2 levels remained high, and peaked at 712 pg/mL during the reproductive season from March to October, particularly in ovarian stage III. This sustained high E2 reflects continuous vitellogenesis in a multi-spawning fish, where oocytes at various developmental stages coexist12. During this period, the temperature ranged between 26 and 30 °C. In contrast, E2 levels dropped to 156–321 pg/mL during the post-spawning season from November to February, consistent with reduced gonadal activity. These patterns highlight the critical role of E2 in driving vitellogenesis during the reproductive season.

Compared to regional species like rabbitfish (E2 ~ 1446 pg/mL29) and waigieu seaperch (E2 ~ 900 pg/mL6), G. speciosus exhibits lower peak E2 levels, potentially reflecting differences in reproductive strategies or hormonal regulation. Unlike single-spawning species, where E2 typically declines post-ovulation34, the sustained E2 in G. speciosus during IV - V suggests ongoing vitellogenesis to support multiple-spawning events35. This is consistent with observations in other multiple-spawning species, where E2 remains elevated to maintain oocyte development across spawning cycles12.

The complexity of steroid hormone dynamics in multiple-spawning species, as seen in G. speciosus, highlights the challenge of detecting peak E2 levels due to their transient nature6. In some species, androgens like testosterone may dominate during certain reproductive phases, complicating endocrine profiles36. Monitoring E2 and other steroid hormones provides a valuable tool for assessing physiological status of fish, thereby reproductive status and optimizing broodstock conditioning in aquaculture.

Conclusion

This study reveals that the golden trevally is a multiple-spawning fish species with a prolonged reproductive season from March to October, characterized by asynchronous ovarian development and sustained high GSI (2.46–3.05%) and plasma E2 levels (460–712 pg/mL). The peak E2 during vitellogenesis (stage III) underscores its critical role in oocyte development. These findings enhance understanding of G. speciosus reproductive physiology and endocrinology, providing actionable insights for aquaculture. Broodstock managers can optimize feeding and environmental regimes to align with seasonal reproductive cycles, improving gamete quality and seed production efficiently. However, this study is limited by its focus on captive, farmed female golden trevally in the south-central coastal waters of Vietnam over a single annual cycle, which may not fully reflect dynamics in wild populations. Future research, including comparative studies on wild fish across the natural Indo-Pacific distribution of the species, would strengthen these insights, and the applicability of these findings, particularly in supporting sustainable fisheries management by informing seasonal protections for wild populations, while advancing commercial aquaculture in Southeast Asia.

Materials and methods

Broodstock rearing and sampling

Farmed female golden trevally (G. speciosus) were reared in floating sea cages (4 × 4 × 4 m dimension) in the south-central coastal waters of Vietnam (12.2099° N, 109.0929° E). Those females originated from the local hatcheries and have been reared in the aquaculture facilities of Nha Trang University for over six years. The stocking density was maintained at 3 kg/m3 with a 1:1 male-to-female ratio. Fish were held under ambient environmental conditions, with salinity ranging from 30–34‰, water temperature from 22 to 30 °C, pH from 7.8 to 8.6, and dissolved oxygen from 4.5 to 6.5 mg/L. Seawater environmental parameters were weekly measured during the study period. Broodstock were fed daily with raw trash fish at 2–3% of body weight. Monthly, 10 female fish (mean body weight: 900 ± 200 g and mean total length of 40 ± 5 cm) were randomly sampled for ovarian tissues and blood collection. Blood was collected from the caudal vein using syringes, transferred into heparinized tubes, and stored in -80 °C until analysis. The gonadosomatic index (GSI) was calculated as the percentage of gonadal weight to total body weight. Gonadal development was classified into six stages based on the histological criteria and ovarian morphology, following established criteria (Table 1, also refs25,26.

Table 1 Description of ovarian development stages in female golden trevally (Gnathanodon speciosus).
Full size table

Ethics declarations

All procedures involving live animals complied with the Vietnam Veterinary Law 2015 and Government Decree 90/2017. We confirm compliance with the ARRIVE guidelines for animal research. The research and experimental protocols were approved by Nha Trang University and the National Foundation of Science and Technology Development of Vietnam (NAFOSTED).

Histological analysis

The ovarian samples were removed from the fixative solution, washed, and dehydrated by soaking in absolute ethanol for 4–8 h, then cleared in methyl salicylate for 12–24 h. Thereafter, samples were infiltrated with molten paraffin at 65 °C for a minimum of 6 h, embedded in paraffin molds, and cooled for 30 min to solidify. Paraffin blocks containing the sample were trimmed into trapezoidal or rectangular shapes, mounted on wooden bases, and labeled. Sections were sliced at 5–7 μm thickness using a microtome, floated in warm water (40–45 °C) for 1–2 min to expand, and then dried at 45–60 °C for 1–4 h. After drying, the sample was deparaffinized by immersion in xylene solutions and then immersion in ethanol solutions at different concentrations for 2–3 min. Finally, the sample was stained in Hematoxylin-Mayer solution (4–6 min) and Eosin (2 min), then dried, covered with adhesive, and labeled. Gonadal histological sections were examined using an Olympus microscope (Japan), at 10× and 40× magnification to assess oocyte development stages.

Plasma estradiol 17β assay

We conducted the plasma estradiol 17β assay following the methodology described in our previous study37. Briefly, the plasma E2 levels were analyzed using enzyme-linked immunosorbent assay (ELISA) with Steroid hormone enzyme immunoassay (EIA) Kits (Cayman Chemical Company, Ann Arbor, MI, USA). Blood samples were centrifuged at 10,000× g for 10 min at 4 °C to isolate plasma. Plasma was purified through organic solvent extraction to eliminate interference. Specifically, we used a vortex mixer to mix plasma with 5 mL of diethyl ether. Following phase separation, the aqueous layer was frozen in an ethanol/dry ice bath, the lipophilic phase was transferred into a clean tube, and the ether phase was removed via vacuum centrifugation. The resulting dry extract was resuspended in 300 µl EIA buffer by vortexing. Enzyme immunoassays were performed per the manufacturer’s protocol with a 75-minute development time. Absorbance was measured at 405 nm using a Thermo Multiskan EX 96-well spectrophotometer (Netherlands). E2 concentrations were calculated using a standard curve linearized through a logit transformation of bound sample/maximum bound (B/B0). The intra- and inter-assay coefficients of variation for all EIAs were approximately 10% and 8%, respectively. Cross-reactivity for E2 EIAs was 2.2% with 100% specificity for their respective steroid. On average, both intra- and inter-assay variations for steroid analyses remained below 5%.

Statistical analysis

Data were analyzed using SPSS statistical software (version, IBM Corp., Armonk, NY, USA). Monthly variations in GIS, plasma E2 concentrations, and ovarian development stages were compared using one-way analysis of variance (ANOVA). Post hoc comparisons were conducted using Duncan’s multiple range test at a 95% confidence level (α = 0.05). All data were tested for normality and homogeneity of variance prior to analysis, with transformation applied as needed to meet ANOVA assumptions.