Sex-Specific Marker Development and Identification of an XY Sex Determination System in the Tire Track Eel (Mastacembelus favus) via Whole-Genome Resequencing

TL;DR

This study identifies a male-specific genetic system in tire track eels and creates tools for sex identification to improve aquaculture.

Contribution

Development of two PCR-based sex markers and evidence for an XY system in Mastacembelus favus.

Findings

Two PCR-based markers (W5 and W14) reliably distinguish male and female tire track eels.

Evidence supports an XY sex determination system in Mastacembelus favus.

Male-specific sequences show high conservation in related species despite non-transferable markers.

Abstract

This study developed a rapid, accurate, and cost-effective genetic sex identification method for the tire track eel (Mastacembelus favus). Whole-genome resequencing identified 69 male-specific genomic regions, from which two PCR-based markers (W5 and W14) were designed. These markers reliably distinguished males and females in genetic sex identification, indicating an XY sex determination system in M. favus. Sequence conservation was further examined by BLAST in the related species Mastacembelus armatus. This work provides practical tools for monosex breeding and advances the understanding of sex determination in Mastacembelus species.

The tire track eel (Mastacembelus favus) is a freshwater fish with high economic value and aquaculture potential. However, its sex determination mechanism remains unclear, which limits the development of monosex culture and sex-controlled breeding. To address this, preliminary female and male reference genomes were generated using second-generation sequencing, followed by whole-genome resequencing of four females and four males. Comparative analyses identified 69 male-specific sequences, with a total length of approximately 44.5 kb. Based on these sequences, two PCR-based sex-specific markers (W5 and W14) were developed. Both markers showed complete concordance with phenotypic sex in parental and offspring from controlled crosses populations, providing strong evidence for a male heterogametic (XY) sex determination system in M. favus. Although the markers were not transferable to the closely related species Mastacembelus armatus, the corresponding male-specific sequences exhibited high genomic conservation. In conclusion, this study provides reliable molecular tools for genetic sex identification in M. favus. These tools will support monosex aquaculture and sex-controlled breeding programs, while also offering insights into sex determination and sex chromosome evolution in the genus Mastacembelus.

Figures

Full text

1. Introduction

Fishes represent the most species-rich and ecologically diverse group of vertebrates, with over 32,000 described species worldwide. Their sex determination systems exhibit remarkable plasticity, diversity, and evolutionary complexity, making them important models for developmental and evolutionary biology [1,2,3,4]. However, in most teleosts, sex chromosomes are still at an early stage of evolution. Distinct sex chromosomes have been identified in only approximately 10% of fish species, and they are predominantly homomorphic, with limited regions of recombination suppression and low levels of structural differentiation [5,6,7]. These characteristics greatly limit the effectiveness of traditional cytogenetic approaches in accurately identifying sex determination systems in fishes.

In many cultured fish species, pronounced sexual size dimorphism (SSD) is observed during some or all life-history stages, resulting in substantial differences between sexes in growth rate, body size, feed conversion efficiency, and reproductive investment [5,8]. For example, males grow significantly faster than females in Nile tilapia (Oreochromis niloticus) [9], yellow catfish (Pelteobagrus fulvidraco) [10], and zig-zag eel (Mastacembelus armatus) [11]. In contrast, females exhibit superior growth performance in mandarin fish (Siniperca chuatsi) [12], largemouth bass (Micropterus salmoides) [13], and half-smooth tongue sole (Cynoglossus semilaevis) [14]. Consequently, clarifying sex determination systems and implementing sex-controlled breeding can effectively utilize SSD to enhance overall aquaculture efficiency.

In addition, to enable effective sex-controlled breeding, accurate identification of genetic sex based on molecular markers is essential. The earliest molecular marker technology can be traced back to restriction fragment length polymorphism (RFLP) analysis established by Botstein et al. [15]. Subsequently, a variety of DNA-based marker systems, including random amplified polymorphic DNA (RAPD) and amplified fragment length polymorphism (AFLP), were rapidly developed and successfully applied to several economically important aquaculture species, such as Yellow Rivercarps (Cyprinus carpio) [16], turbot (Scophthalmus maximus) [17] and O. niloticus [18], C. semilaevis [19]. These markers enable sex identification through simple PCR-based assays [20,21]. However, traditional marker systems are often characterized by low efficiency, limited reproducibility, and labor-intensive procedures [22].

In recent years, the rapid advancement of next-generation sequencing (NGS) technologies has greatly facilitated the development of sex-specific molecular markers, which have become indispensable tools in molecular-assisted breeding programs [23,24]. Among these approaches, whole-genome resequencing (WGS re-seq) has emerged as one of the most efficient strategies for developing genetic sex markers due to its high throughput and decreasing costs. By comparing genome-wide sequence differences between females and males, WGS re-seq enables efficient identification of sex-specific sequences, inference of sex determination systems, and localization of sex-determining regions (SDRs) [11,25,26]. Molecular markers developed using this strategy have been successfully applied in multiple fish species, including C. semilaevis [14], snakehead (Channa argus) [27], grass carp (Ctenopharyngodon idella) [28], large yellow croaker (Larimichthys crocea) [29], mandarin fish (S. chuatsi) [30], striped seabream (Oplegnathus fasciatus) [31], and zig-zag eel (M. armatus) [11].

The tire track eel (M. favus) is a widely distributed species across Thailand, the Lao People’ s Democratic Republic, Cambodia, Vietnam, and the Malay Peninsula, reaching a maximum total length of approximately 70 cm and possessing high economic value due to its superior flesh quality and nutritional content [32,33]. Small-scale aquaculture practices in China and Vietnam have revealed pronounced sexual size dimorphism in this species, with males exhibiting faster growth, higher feed conversion efficiency, greater disease resistance, and larger body size than females [34,35]. These traits indicate the potential advantages of monosex culture in M. favus. However, reliable sex identification based on external morphology is not feasible at early developmental stages [35].

Despite its aquaculture potential, genomic resources for M. favus remain limited. The absence of effective molecular markers for genetic sex identification, together with the unknown sex determination system and sex-determining (SD) genes, constitutes a major bottleneck for large-scale aquaculture development [36,37,38]. Previous studies have shown high cross-species transferability of microsatellite loci between M. favus and the closely related M. armatus [37]. Moreover, their strong morphological and karyotypic similarities have led to occasional misidentification, which may cause irreversible losses of germplasm resources if not properly managed [36].

In this study, WGS re-seq was employed to generate draft genome assemblies of female and male M. favus. Comparative analyses between female and male genomes were conducted to identify sex-specific sequences and develop reliable molecular markers for genetic sex discrimination. The conservation of these sequences and markers was further evaluated in M. armatus. The aims of this study were to: (i) elucidate the sex determination system of M. favus; (ii) develop sex-specific molecular markers to provide technical support for sex-controlled breeding; and (iii) assess variations in sex-determining regions among Mastacembelus species.

2. Materials and Methods

2.1. Sample Collection

Adult M. favus individuals were purchased from Guangdong Lianyi Aquatic Technology Co., Ltd. (Jiangmen, Guangdong Province, China). 24 adults (12 females and 12 males) were randomly selected and dissected for gonadal examination, and sex was determined by visual inspection of the gonads. Caudal fin tissues were collected from each individual and preserved in 95% ethanol for subsequent molecular analyses. The remaining adult M. favus individuals were acclimated to formulated feed and maintained for future breeding experiments.

In addition, 16 wild M. armatus individuals (8 females and 8 males) from the Tanjiang River basin were included for cross-species marker validation.

All animal handling procedures and experimental protocols were approved by the Experimental Animal Ethics Committee of the Guangzhou University of China.

2.2. Breeding of Self-Fertilized Offspring

Artificially fertilized offspring were obtained from known M. favus parents. Spawning was induced by a two-step hormone injection, with hormones fully dissolved in 0.7% physiological saline and injection protocols adapted from M. armatus [39]. Fertilized eggs were treated with red clay to remove adhesiveness and incubated in circulating water at 26–28 °C. Eggs were transferred to hatching ponds at the pre-hatching stage for larval emergence.

During incubation and larval rearing, dissolved oxygen was maintained at 6–10 mg/L, and approximately one-fifth of the water volume was replaced every two days with aerated, filtered water to control ammonia accumulation. Larvae were initially fed Artemia salina, followed by a gradual transition to commercial diets; Limnodilus spp. were avoided to prevent abnormal sex differentiation [40]. At 120 days post-fertilization (dpf), offspring were randomly sampled for sex-specific marker validation.

2.3. Sex Identification of Self-Fertilized Offspring

Gonadal tissues from the 24 (12 females and 12 males) self-fertilized offspring were fixed in 4% paraformaldehyde for 24 h, dehydrated through a graded ethanol series, and embedded in paraffin. Sections of 5–6 µm thickness were prepared, stained with hematoxylin and eosin, and mounted with neutral resin. Phenotypic sex was determined by microscopic examination of the gonadal histology [41].

2.4. DNA Extraction and Library Construction

Genomic DNA was extracted from all samples using a commercial kit (Novizan, Nanjing, China). DNA quality and concentration were assessed by 1.2% agarose gel electrophoresis and a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA). Qualified DNA from four females and four males of M. favus was used for library construction. Library preparation was performed by Guangzhou Ruike Gene Co., Ltd. (Guangzhou, China), following standard Illumina protocols to generate paired-end libraries with ~350 bp inserts. Library quality was evaluated by qPCR and the Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA) [42,43]. Libraries passing quality control were sequenced on an Illumina NovaSeq 6000 (Illumina, San Diego, CA, USA) platform with 150 bp paired-end reads.

2.5. Genome Assembly and Evaluation Methods

After quality control, high-quality sequencing reads were assembled de novo using SOAPdenovo2 (v2.04) with a multi-Kmer strategy (K = 33, 43, 53, 63, 73, 83, 93), constructing scaffolds based on read overlaps and paired-end relationships to obtain optimal assemblies. GapCloser (v1.12) was then used for two rounds of gap filling and correction, generating the reference genome for subsequent analyses [44].

Genome assembly quality and characteristics were assessed by K-mer analysis. Clean reads were counted using Jellyfish (v2.3.0, https://github.com/gmarcais/Jellyfish, accessed on 6 May 2025) with K = 21, and genome size was estimated with GenomeScope2 (v2.0, https://github.com/tbenavi1/genomescope2.0, accessed on 6 May 2025) assuming diploidy (ploidy = 2, K = 21) [45,46].

2.6. Sequence Alignment and Identification of Sex-Specific Sequences

Resequencing reads from the remaining three male individuals were aligned to the male reference genome using bwa (v0.7.17-r1188) with the mem algorithm and default parameters [47]. Reads that failed to align as proper pairs were discarded. Sequences covered by all three male datasets were defined as male-shared sequences. Similarly, reads from three female individuals were aligned to the female reference genome, and sequences present in all three female datasets were defined as candidate female-specific sequences.

Aligned reads were sorted using SAMtools (v1.14) with the sort function. To improve alignment accuracy, properly paired reads were filtered using view –bF 12 for downstream analyses [48]. Sex-specific sequences were identified by comparing coverage between sexes, retaining sequences fully covered in all individuals of one sex but absent in all individuals of the opposite sex (Figure 1).

2.7. Primer Design and PCR Amplification Validation

Sex-specific sequences identified from high-throughput sequencing were used to design PCR primers with Primer3 (v4.1.0, https://bioinfo.ut.ee/primer3-0.4.0/, accessed on 16 June 2025) [49]. Design appropriate PCR protocols based on the melting temperature of primers. Then PCR products were visualized on 1.2% agarose gels.

The reliability of the markers was first validated in the remaining 8 female and 8 male parental individuals and was subsequently further evaluated in 24 (12 females and 12 males) self-fertilized offspring. Cross-species validation was performed on 16 individuals (8 females and 8 males) of M. armatus to assess marker transferability.

3. Results

3.1. Phenotypic Sex Identification

Phenotypic sex was determined by gonadal dissection in all individuals. For offspring whose sex could not be reliably identified based on external morphology due to incomplete gonadal development (Figure 2a,b), histological examination was performed. Gonadal sections stained with hematoxylin and eosin revealed individuals with abundant oocytes as females (Figure 2c) and those containing spermatocytes as males (Figure 2d). Using combined anatomical and histological analyses, phenotypic sex was accurately confirmed for all individuals used in genome sequencing and marker validation.

3.2. Sequencing Data Statistics

Raw sequencing reads were filtered using fastp (v0.23.2) with the parameters -q 20 -l 150, while all other settings were kept as default. In total, approximately 119.72 GB of high-quality clean data were obtained from eight M. favus individuals, including 60.73 GB from four females and 57.99 GB from four males. For all samples, the Q20 values exceeded 98% and the Q30 values exceeded 93%, demonstrating high sequencing accuracy and data reliability. These results indicate that the sequencing data were of sufficient quality for downstream genome assembly and sex-specific sequence identification. Detailed statistics are provided in Table 1.

3.3. Genome Sequencing, Assembly, and Evaluation

Independent genome assemblies were generated for the female (F_WWCQF3) and male (M_WWCQM3) individuals using their respective sequencing datasets. Genome characterization indicated comparable genome sizes between sexes, with low heterozygosity and repeat content (Table 2), suggesting limited genomic complexity. The K-mer distributions exhibited a clear unimodal pattern (Figure 3), confirming that M. favus is a diploid species. Based on the SOAPdenovo2 assemblies (Table 3), the assembled genome sizes were 613.49 Mb for the female and 610.53 Mb for the male, with scaffold N50 values of 34.06 kb and 34.44 kb, respectively. The GC content was approximately 40.56%, and a substantial number of scaffolds exceeded 10 kb in length, indicating that these draft genomes were adequate for downstream sex-specific sequence identification and marker development.

3.4. Identification of Sex-Specific Sequences Supports an XY Sex Determination System in M. favus

Each resequenced individual was aligned to both the female and male reference genomes, with pair-matched read ratios exceeding 85% for all samples (Table 4). A total of 72 sex-specific sequences were subsequently identified, of which the vast majority were male-specific (69 sequences), with a cumulative length of 44,512 bp, ranging from 304 to 2838 bp and averaging 645 bp. In contrast, only three female-specific sequences were detected, with lengths of 1464 bp, 1127 bp, and 370 bp. As illustrated in Figure 4, both the number and total length of male-specific sequences were markedly greater than those of female-specific sequences, suggesting that M. favus likely follows an XY sex determination system. Detailed information on all sex-specific sequences is provided in Supplementary File S1.

3.5. Primers W5 and W14 Accurately Discriminate Sex in Both Parental and Offspring Populations

Based on the 69 male-specific sequences, a total of 54 pairs of male-specific primers were designed. Detailed information on all primer sequences is provided in Supplementary File S2. PCR amplification revealed that primer W5 consistently produced a single male-specific band of 361 bp in all male individuals, whereas no amplification products were detected in any female individuals (Figure 5a,b). Primer W14 consistently amplified two fragments (437 bp and 257 bp) in all males, while only a single 257 bp fragment was observed in all females (Figure 5c,d). Genetic sex assignments based on both markers were fully concordant with phenotypic sex in parental and offspring populations, and the amplification patterns remained stable in the self-crossed progeny, demonstrating the high reliability of W5 and W14 for sex identification.

3.6. Conservation and Cross-Species Applicability of Male-Specific Sequences and Sex-Specific Markers

To investigate the conservation and chromosomal distribution of male-specific sequences in a closely related species, the 69 male-specific sequences identified in M. favus were aligned against the M. armatus reference genome (NCBI: GCA_019455535.1) using BLAST + 2.17.0. Of these, 66 sequences were successfully mapped, with sequence identity ranging from 74.66% to 100% (Figure 6). Chromosomal distribution revealed that 45 sequences were located on the Y chromosome, six on the X chromosome, four on chromosome 15, two on chromosome 18, eight on other autosomes, and one on an unplaced scaffold. These results indicate that male-specific sequences in M. favus are highly conserved in M. armatus and are predominantly enriched on sex chromosomes.

In addition, to evaluate the cross-species applicability of sex-related molecular markers between M. favus and M. armatus, the M. favus sex-specific markers W5 and W14, as well as the M. armatus sex marker M2 [11], were reciprocally tested in the other species. As shown in Figure 7a–c, neither W5 nor W14 exhibited sex-discriminatory capacity in M. armatus, as similar amplification patterns were observed in both males and females, with no clear sex-associated differentiation. Likewise, the amplification pattern of M2 in M. favus was inconsistent: although all males produced amplicons ranging from 500 to 750 bp, the fragment sizes differed markedly from those observed in M. armatus, whereas females yielded only fragments smaller than 500 bp or showed no amplification [11].These results indicate that the currently developed sex-specific markers W5, W14, and M2 are species-specific and effective only within their respective taxa, and therefore lack cross-species universality.

4. Discussion

The identification of sex-specific molecular markers is fundamental for elucidating sex determination systems, particularly in teleosts where sex chromosomes are weakly differentiated and lack conspicuous heteromorphy [7,12]. Traditional marker systems such as AFLP are characterized by low throughput, non-targeted discovery, and limited integration with genomic resources, and have repeatedly failed to identify sex-linked markers in multiple fish species [50,51,52]. Although RAD-seq has expanded marker discovery in some taxa [53], its dependence on restriction sites and short sequence tags restricts genome coverage and hampers precise localization of SDRs [27,54,55,56].

In contrast, The WGS re-seq strategy employed in this study offers higher coverage and more comprehensive variant information [13,30]. With the support of reference genomes, comparative analyses of resequencing data from four females and four males enabled the efficient identified of 72 sex-specific sequences. This approach, based on multiple high-depth samples, effectively excludes false positives arising from individual genetic variation or sequencing artifacts, thereby yielding stable and broadly applicable markers [26,57]. Importantly, this strategy does not rely solely on a single-sex reference genome; instead, cross-validation using both male and female assemblies further enhances the reliability of identified sex-specific sequences [58].

Based on these genome-wide comparisons, the strong predominance of male-specific sequences, coupled with the stable male-specific amplification of markers W5 and W14 across both parental and offspring populations, supports at the molecular level that M. favus follows a male heterogametic (XX/XY) sex determination system [13,26,59]. Combined with the pronounced male-biased growth observed in this species, the PCR-based markers developed here will enable accurate genetic sex identification at early developmental stages, overcoming the absence of external sexual dimorphism and providing a practical foundation for marker-assisted breeding and the future production of all-male stocks [39].

Comparative analyses among closely related species have shown that transferable sex-specific markers can serve as molecular anchors for investigating the radiation and evolution of sex determination systems in fishes, as demonstrated in spotted mandarin fish (Siniperca scherzeri) and big-eye mandarin fish (S. kneri) [58], as well as in Sebastes carnatus and S. chrysomelas [60]. Our study reveals a notable contrast: While male-specific sequences of M. favus are highly conserved in the M. armatus genome and show significant enrichment on its sex chromosomes, particularly the Y chromosome. This conservation pattern may indicate a shared, homologous Y-linked region. However, the sex-specific markers developed here (W5 and W14) are not transferable between the two species. Conversely, the previously reported sex marker for M. armatus is also not applicable to M. favus. Similar patterns have been observed in blunt snout bream (Megalobrama amblycephala) and its congeners M. terminalis and M. hoffmanni [61]. Given that both M. favus and M. armatus are likely at an early stage of sex chromosome differentiation and lack heteromorphic sex chromosomes [38,62], this apparent paradox of sequence conservation coupled with marker non-transferability could be explained by a highly restricted SDR that may comprise only small, non-recombining genomic fragments.

Similarly to findings reported in C. maculata and M. salmoides [25,63], the sex-specific markers developed in this study also show no homology to well-characterized master SD genes such as Dmrt1 or Amhr2 [8,64]. This may be attributed to the current genome assembly, which is based on second-generation sequencing data, and the moderate sample size used for marker validation, imposing certain limitations on resolution and continuity. Future efforts should prioritize the construction of a chromosome-level or telomere-to-telomere (T2T) reference genome for M. favus to enable precise delineation of the SDR. Additionally, expanding validation across geographically diverse populations will further confirm the general applicability of these markers and aid in screening candidate SD genes. These steps will help clarify the origin and evolution of the sex determination system within the genus Mastacembelus.

5. Conclusions

In this study, a draft genome of M. favus was generated using second-generation sequencing, and comparative resequencing of four females and four males identified 44.5 kb of male-specific sequences, supporting an XX/XY sex-determination system. Two robust PCR-based male-specific markers (W5 and W14) were further developed and validated in both parental and offspring populations, demonstrating high specificity and reliability for genetic sex identification. Homology analyses with M. armatus revealed that a large proportion of male-specific sequences are evolutionarily conserved, providing additional insights into sex chromosome differentiation within Mastacembelus. Collectively, these results provide practical molecular tools for early sex identification and monosex breeding in M. favus, establishing a basis for further research. Future studies will focus on building a T2T genome for precise SDR mapping and validating these markers across diverse populations, thereby elucidating the evolutionary dynamics of sex chromosomes within Mastacembelus.