Molecular Identification of HR97 in the Swimming Crab Portunus trituberculatus and Its Potential Involvement in Ovarian Development

TL;DR

This study identifies HR97, a nuclear receptor gene in crabs, and shows it plays a role in ovarian development, with arachidonic acid affecting its activity.

Contribution

The study identifies and characterizes HR97 in the crab Portunus trituberculatus and reveals its role in ovarian development and gene regulation.

Findings

HR97 is predominantly expressed in crab ovaries and is linked to ovarian development.

Arachidonic acid suppresses HR97 expression and partially overlaps with its transcriptional responses.

Reducing HR97 activity disrupts ovarian development and affects genes related to transport, signaling, and transcription.

Abstract

Nuclear receptors are DNA-binding gene regulators that can be influenced by hormones or fats. This study aimed to identify and explore the function of HR97, a nuclear receptor gene found in many arthropods but not in insects, in an economically important marine crab. HR97 was most active in the ovary, and temporarily reducing its activity in living crabs disrupted ovarian development and altered genes involved in nutrient transport, cell surface signaling, and nuclear transcriptional control. We also found that arachidonic acid, an important unsaturated fatty acid in crustacean ovaries, reduced HR97 gene expression. However, arachidonic acid and HR97 reduction produced only partially overlapping transcriptional responses, suggesting that arachidonic acid can active additional pathways. These findings advance understanding of crab reproduction and may inform future broodstock management.

Nuclear receptor HR97 is considered as a non-insect arthropod–specific receptor, but its roles in decapod reproduction remain poorly understood. Here, we identified and characterized an HR97 ortholog from the swimming crab Portunus trituberculatus (PtHR97) and verified its placement within the NR1L nuclear receptor family by phylogenetic analysis. PtHR97 encodes a canonical nuclear receptor with a conserved DNA-binding domain (DBD) and ligand-binding domain (LBD). Quantitative PCR revealed predominant PtHR97 expression in the ovary and stage-dependent changes during ovarian development. Using an ovarian explant culture system, we found that arachidonic acid (AA) consistently suppressed PtHR97 transcript levels, while methyl farnesoate (MF) and pyriproxyfen (P) had no significant effect, indicating a potential inhibitory role for AA in PtHR97 expression. RNA interference of HR97 caused significant changes in ovarian development, including reduced GSI, smaller oocytes, and uneven eosinophilic granule distribution. Transcriptomic profiling of HR97-silenced ovaries indicated that the major responses involved genes associated with substrate transport/exchange, cell boundary–related signaling and transduction, and disturbed nuclear transcriptional regulation. Short-term in vivo perturbations (HR97 RNAi and AA treatment) further supported these expression changes and revealed that AA- and HR97 RNAi–elicited transcriptional responses only partially overlapped. Taken together, these results suggest that HR97 contributes to ovarian development, potentially through broad transcriptional responses related to transport, signaling, and gene regulation. Although AA may suppress HR97 expression, HR97 does not fully explain AA-mediated regulation of ovarian development.

Figures

Full text

1. Introduction

Nuclear receptor HR97 is regarded as a nuclear receptor family member (NR1L) unique to non-insect arthropods. Its existence was first identified through genomic analyses in branchiopod crustaceans, particularly in Daphnia species [1,2]. Subsequent studies have reported HR97 or HR97-like homologs across diverse crustacean taxa, including copepods [3] and decapods such as lobsters and crabs [4]. Notably, HR97 homologs have also been identified in the chelicerate Metaseiulus occidentalis and Ixodes scapularis, whereas no corresponding homologs have been detected in sequenced insect genomes [5]. This suggests that HR97 likely originated in the common ancestor of Pancrustacea and was subsequently lost during insect evolution. Phylogenetic analyses further place HR97 within the NR1 subfamily of the nuclear receptor superfamily, showing evolutionary relatedness to insect HR96 (NR1J) as well as vertebrate NR1I nuclear receptors, including PXR, CAR, and VDR [5].

The endogenous ligands of HR97 and its physiological functions remain insufficiently defined. Despite its sequence similarity with PXR/CAR/HR96, which function as broad-spectrum xenobiotic-sensing nuclear receptors [6,7], HR97 in Daphnia is insensitive to many ligands known to activate CAR, PXR, or HR96 [5]. Transcriptional activation assays using GAL4 chimeric receptors have shown that, among the large number of compounds tested, HR97 responds to only three: the juvenile hormone analog pyriproxyfen (P), the crustacean juvenile hormone precursor methyl farnesoate (MF), and the polyunsaturated fatty acid arachidonic acid (AA) [5]. Pyriproxyfen and methyl farnesoate activate HR97 transcriptional activity only at relatively high concentrations, whereas AA acts as an inverse agonist and significantly suppresses the basal activity of HR97 at physiologically relevant concentrations. In Daphnia, arachidonic acid, as one of the few unsaturated fatty acids retained in the ovary [8], increases reproductive output and represses pyriproxyfen-induced environmental sex determination under restricted dietary conditions [9]. Given the ovarian expression of HR97 and its upregulation during reproductive maturation [5], these findings suggest that HR97 may represent an important reproductive regulatory factor in Daphnia.

HR97 has not yet been functionally characterized in decapod crustaceans, but homologous HR97 sequences have been annotated from transcriptomic datasets of multiple decapod species. A nuclear receptor identification and evolutionary analysis in the ornate spiny lobster Panulirus ornatus showed that decapod HR97s can be assigned to the NR1L group but are phylogenetically distinct from HR97 sequences identified in branchiopods and copepods [4]. In that study, three additional HR97-like receptors were identified and named molt-associated receptors MAR1, MAR2, and MAR3, based on their upregulated expression prior to metamorphosis. Phylogenetic analyses further indicated that the MAR receptors likely originated from within the HR97 lineage but form a distinct monophyletic clade with substantial divergence among members. Notably, transcriptomic datasets from Sagmariasus verreauxi suggest that MAR1–3 are detectable in both ovary and testis, with MAR1 highest in the ovary and MAR2/3 highest in the testis [4]. In decapod crustaceans, molting and ovarian development are regulated through partially overlapping endocrine pathways, including shared hormonal components such as methyl farnesoate (MF) and ecdysteroids [10,11,12], which provides a potential association between HR97-related receptors and reproductive process. Moreover, methyl farnesoate is widely recognized as an important regulator of ovarian development in decapod crustaceans [11,13], and increasing evidence also indicates that arachidonic acid is closely associated with ovarian development and reproductive performance in shrimps and crabs [14,15]. Therefore, if the responsiveness of Daphnia HR97 to MF and AA is conserved in decapod crustaceans, HR97-related nuclear receptors may warrant further attention for their potential physiological relevance to reproductive regulation in this lineage.

The swimming crab, Portunus trituberculatus, is an economically important cultured crab species in China, with the total national catch of swimming crabs reaching 450,000 tons and mariculture production reaching 100,000 tons in 2024 [16]. Ovarian development is widely regarded as a major determinant of broodstock reproductive performance and seed production efficiency [17]. However, the molecular mechanisms regulating ovarian development in decapod crustaceans, particularly those involving nuclear receptors, remain insufficiently characterized. In our previous ovarian transcriptome dataset, we identified an HR97-like nuclear receptor sequence in P. trituberculatus (Portunus trituberculatus). In the present study, we cloned the full-length cDNA of this gene and found that it is highly expressed in the ovary and shows marked stage-dependent expression changes during ovarian development, suggesting a potential role in ovarian regulation. We then used in vitro assays to characterize its responsiveness to pyriproxyfen, methyl farnesoate (MF), and arachidonic acid (AA). Finally, RNA interference followed by ovarian transcriptomic analysis was performed to systematically evaluate its potential function during ovarian development.

2. Materials and Methods

2.1. Experimental Animals and Tissue Sampling

Wild-caught female swimming crabs (Portunus trituberculatus; 90–300 g) were purchased at regular intervals from a local fish market in Zhenhai, Ningbo, Zhejiang Province, China, according to their ovarian developmental cycle [18]. In brief, the ovarian development of P. trituberculatus can be divided into five stages based on gonadosomatic index (GSI) and ovarian features. Stage I (GSI < 0.6%) is characterized by a transparent, ribbon-like oviduct. Stage II (GSI 0.5–1.0%) shows a milky-white ovary and is dominated by endogenously vitellogenic oocytes. Stage III (GSI 0.9–5%) presents a light yellow to orange-yellow ovary, marking rapid ovarian development and dominated by exogenously vitellogenic oocytes. Stage IV (GSI 5–10%) represents the near-mature stage, with an orange-red ovary dominated by exogenously vitellogenic oocytes and near-mature oocytes. Stage V (GSI 12–15%) corresponds to the mature stage, in which oocytes are distinct and relatively uniform in size [18]. For biological replicates, four crabs were used for each stage. Prior to sacrifice, crabs were anesthetized on ice for 10 min. Tissues from ovary, thoracic ganglion, epidermis, muscle, hepatopancreas, eyestalk, intestine, brain, gill, heart and Y-organ were collected. All tissues were immediately frozen in liquid nitrogen and stored at −80 °C in RNA stabilization solution (Kangwei Century, Taizhou, Jiangsu, China) for future use.

2.2. RNA Extraction and cDNA Synthesis

Total RNA was extracted using the TRIzol reagent (Omega, Norcross, GA, USA) according to the manufacturer’s instructions. RNA integrity was assessed by agarose gel electrophoresis, and the purity and concentration were determined using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). First-strand cDNA was synthesized from 1 μg of total RNA using the ABScript III RT kit (ABclonal, Wuhan, Hubei, China) according to the manufacturer’s protocol. Reverse transcription was performed using a mixture of random primers and Oligo(dT)18 primers.

2.3. Identification and ORF Validation of PtHR97

A putative PtHR97 sequence was first identified from our ovarian RNA-seq dataset by homology-based screening using the DNA-binding domain (DBD) sequence of Daphnia HR97 as the query. The predicted open reading frame (ORF) was then validated by RT-PCR using ovary-derived cDNA as the template. Gene-specific primers for ORF validation (Table S1) were designed with Primer Premier 5.0 and synthesized by Zhejiang Youkang Biotechnology Co., Ltd. (Hangzhou, China). RT-PCR was performed using 2 × Es Taq MasterMix (CWBIO, Taizhou, Jiangsu, China) under the following conditions: 94 °C for 3 min, followed by 35 cycles of 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 1 min, with a final extension at 72 °C for 10 min. PCR products were examined on a 1% agarose gel, purified using the DiaSpin DNA Gel Extraction Kit (Sangon Biotech, Shanghai, China), ligated into the PMD19-T vector (Takara, Shiga, Japan), and transformed into competent Escherichia coli DH5α cells. Transformants were selected on LB agar plates containing ampicillin, and six positive clones were further verified by colony PCR, and colonies yielding products of the expected size were considered positive. Six positive clones were selected and Sanger-sequenced by Zhejiang Youkang Biotechnology Co., Ltd. (Hangzhou, China).

2.4. Sequence Analysis and Phylogenetic Analysis

The open reading frame (ORF) was identified using the online ORF Finder (NCBI), and the corresponding amino acid sequence was translated. Basic protein properties of the deduced amino acid sequence were predicted using ExPASy ProtParam (https://web.expasy.org/protparam/). Transmembrane helices were predicted with TMHMM 2.0 (https://services.healthtech.dtu.dk/services/TMHMM-2.0/). Conserved domains were annotated using SMART (https://smart.embl.de/). Multiple sequence alignments were performed with Clustal X 2.0.11. Phylogenetic analysis was conducted in MEGA 7.0 based on ligand-binding domain (LBD) sequences of representative NR1 family nuclear receptors. A neighbor-joining (NJ) tree was constructed using the p-distance model with 1000 bootstrap replicates, and the resulting tree was visualized using ChiPlot (https://www.chiplot.online/).

2.5. Ovarian Explant Culture and Ligand Treatments

A female swimming crabs (GSI of 5.9% and body weight of 169 g) at the exogenous vitellogenic stage were selected for ovarian explant assays. An ovarian explant culture system was established according to Tu et al. [19]. Briefly, ovarian tissue was carefully dissected from female crabs under sterile conditions, and washed three times with cold phosphate-buffered saline (PBS) to remove residual hemolymph and impurities. Ovarian explants (50 mg) were then transferred into a 24-well culture plate containing 1 mL of medium per well. The medium consisted of 79% M199 (Corning, New York, NY, USA), 20% heat-inactivated fetal bovine serum (Bovogen, Melbourne, Victoria, Australia), and 1% penicillin–streptomycin solution (100×; Solarbio, Beijing, China). The plates were incubated at 25 °C in a shaker incubator under ambient atmospheric conditions. Prior to experimental treatments, explants were pre-cultured for 1 h to acclimate to the in vitro environment. Explants were assigned to 10 treatment groups, with four biological replicates per group. The control group was treated with crab saline, while the other groups received treatments with three concentrations each of methyl farnesoate (MF), pyriproxyfen (P), and arachidonic acid (AA). The working concentrations were 40 nM, 400 nM, and 4 μM for MF, 1 μM, 10 μM, and 100 μM for P, and 10 mM, 100 mM, and 1 M for AA. MF (Bailingwei, Beijing, China, S-0153), P (Maikelin, Shanghai, China, P834638), and AA (Maikelin, Shanghai, China, A875621) were dissolved in DMSO to prepare stock solutions, and then diluted with crab saline to obtain working solutions. The final DMSO concentration was kept low (≤0.1%, v/v) in all treatments, and the control group received vehicle containing 0.1% DMSO. Explants were collected 6 h after treatment and processed for qPCR analysis.

2.6. RNA Interference In Vivo

Wild-caught female swimming crabs (Portunus trituberculatus, 140–250 g) were obtained from Xianxiang, Ningbo, Zhejiang Province, China, in November and acclimated for 7 days in a recirculating aquaculture system at Ningbo University. To ensure the used females are typically in a rapid ovarian development period during this season, individuals were pre-screened by trans-illuminating both sides of the carapace with a flashlight to assess ovarian fullness, and crabs without obviously full ovaries were selected for the in vivo RNAi assay. During acclimation, each crab was housed individually in a 32 × 20 × 14 cm tank to prevent aggressive behavior. Water temperature was maintained at 25 ± 1 °C and salinity at 24 ± 1 ppt. Dissolved oxygen was kept above 6.0 mg L^−1^, and both ammonia nitrogen and nitrite were maintained below 0.5 mg L^−1^. Crabs were fed fresh razor clams (Sinonovacula constricta) twice daily, and seawater was renewed at the same frequency. Gene-specific primers containing T7 promoter sequences (Table S1) were designed, and dsRNA targeting PtHR97 or GFP was synthesized following the procedures described by Tu et al. [19]. For the long-term RNAi experiment, healthy crabs were randomly assigned to the dsPtHR97 group or the dsGFP control group (n = 8 per group) and injected with 100 μL of dsRNA at 1 μg/g body weight at the base of the swimming leg every 48 h for five rounds. Ovaries were sampled 48 h after the final injection for RNA-seq and H&E histology. For short-term validation, dsPtHR97 and dsGFP groups were established for qPCR analysis, and an additional AA-injected group was included as a pharmacological perturbation control (AA, 1.83 g). Each group contained four crabs (n = 4). Ovaries were sampled 72 h after treatment for qPCR analysis.

2.7. Histological Analysis

Ovary tissues collected from the in vivo experiment were washed with PBS and fixed at 4 °C with 4% paraformaldehyde (PFA) for 12 h, samples were gradient dehydration in 70–100% ethanol, cleaned, and balanced with xylene. Ovary samples were imbedded in paraffin, sectioned at 5 μm thickness and dyeing in hematoxylin and eosin (H&E) staining solution. The diameter of the oocytes was determined by measuring the long and short axis lengths of the largest cell in each section [20]. It was photographed using an optical microscope (Nikon Eclipse, Tokyo, Japan) to observe its histological morphology.

2.8. RNA-Seq

Total RNA was extracted using the RNA 6000 Nano Kit (Agilent Technologies, Santa Clara, CA, USA), followed by removal of genomic DNA with DNase I from the same manufacturer. RNA integrity and purity were assessed using the Agilent 2100 Bioanalyzer and NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA). Only samples with an OD260/280 ratio between 1.8 and 2.2 and an RNA integrity number (RIN) ≥ 8.0 were selected for library preparation. Libraries were generated using the TruSeq RNA Sample Preparation Kit (Illumina, San Diego, CA, USA) following the manufacturer’s protocol. Library concentration and insert size were quantified using the Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen, Waltham, MA, USA) with a Quantifluor-ST fluorometer (Promega, Madison, WI, USA), and confirmed on an Agilent 2100 Bioanalyzer using the High-Sensitivity DNA Kit. Final libraries were diluted to 4–5 pM and sequenced using the Illumina NovaSeq 6000 platform with 151 bp paired-end chemistry.

2.9. Functional Annotation and Differentially Expressed Genes (DEGs) Analysis

Raw reads were filtered for quality, removing adapter sequences, reads with more than 10% ambiguous bases, and low-quality bases with a Phred score below 20. Clean reads were aligned to the Portunus trituberculatus reference genome (NCBI Accession No. ASM1759143v1) using TopHat2 v2.1.1 [21]. Transcript reconstruction was performed with StringTie v2.2.1 [22]. Functional annotation was carried out by searching the assembled transcripts against various databases, including NCBI non-redundant protein (NR), Gene Ontology (GO) [23], eggNOG [24], Swiss-Prot, and the Kyoto Encyclopedia of Genes and Genomes (KEGG) [25]. Gene expression levels were quantified using RNA-Seq by Expectation-Maximization (RSEM) based on genome-aligned data and annotation files. Gene expression was normalized as fragments per kilobase of transcript per million mapped reads (FPKM). Differentially expressed genes (DEGs) were identified using DESeq2 with a threshold of adjusted p-value < 0.05 and |log2FC| > 1. The Benjamini–Hochberg (BH) method was applied for false discovery rate (FDR) adjustment. Functional enrichment analysis was performed using the GO and KEGG databases to identify significantly affected biological processes and pathways. Additionally, to better understand the DEGs, we manually classified them according to their annotations.

2.10. Quantitative PCR (qPCR)

Quantitative real-time PCR (qPCR) was used to assess relative mRNA expression levels with gene-specific primers listed in Table S1. All primers were synthesized by Zhejiang Youkang Biotechnology Co., Ltd. (Hangzhou, China). Primer amplification efficiencies were confirmed by standard curve analysis using a five-point, ten-fold serial dilution of cDNA prior to the experimental assays. qPCR reaction conditions were applied as previously described [26]. Relative transcript abundance was quantified using the 2^−ΔΔCt^ method [27], with β-actin serving as the internal reference gene. Each qPCR reaction was performed in triplicate to ensure accuracy and reliability of the results.

2.11. Statistical Analysis

Data are presented as means ± standard error of the mean (SEM). Normality was assessed using the Kolmogorov–Smirnov and Cochran tests before performing any statistical analyses. Non-parametric tests (Mann–Whitney) were used to analyze data that did not follow a normal distribution. Data following a normal distribution were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s test or t-test (GraphPad Prism 10.1.2 software). Statistical significance was accepted at p < 0.05, which is indicated by letters or asterisks.

3. Results

3.1. Molecular Characteristics of PtHR97

The HR97 gene of Portunus trituberculatus (PtHR97) contains a 2745 bp open reading frame encoding a protein of 914 amino acids (Figure S1). SignalP and TMHMM analyses showed that the deduced PtHR97 protein lacks both signal peptide and transmembrane regions, consistent with its classification as an intracellular nuclear receptor. Domain prediction revealed a typical nuclear receptor modular architecture, comprising an N-terminal A/B region (residues 1–64), a DNA-binding domain (DBD; residues 65–138), a hinge region (residues 139–232), a ligand-binding domain (LBD; residues 233–401), and an extended C-terminal F region (residues 402–914) (Figure 1A). Multiple sequence alignment indicated that the DBD of PtHR97 is highly conserved among crustacean HR97 homologs, retaining the canonical C4 zinc-finger motifs, including the conserved P-box and D-box elements (Figure 1B). In contrast, the LBD showed moderate conservation, with preserving residues associated with ligand binding and coregulator interaction (Figure 1C). Phylogenetic analysis based on LBD sequences further placed PtHR97 within the NR1L subfamily, where it clustered tightly with other crustacean HR97 proteins and was separated from the MARs clades (NR1Q, NR1R, and NR1S) and other related NR1 subfamilies (Figure 1D), supporting its annotation as a bona fide HR97 ortholog in decapod crustaceans.

3.2. Expression Pattern and Ligand Responsiveness of PtHR97

Quantitative PCR analysis showed that PtHR97 transcripts are predominantly expressed in the ovary, with expression levels in ovarian tissue being markedly higher than those detected in other examined tissues, including the thoracic ganglion, epidermis, muscle, hepatopancreas, eyestalk, intestine, brain, gill, heart, and Y-organ (Figure 2A). During ovarian development, PtHR97 expression exhibited a dynamic stage-dependent pattern, increasing progressively from the previtellogenic stage to the late-vitellogenic stage, reaching a peak at the near-mature stage, and subsequently declining sharply at the mature stage (Figure 2B).

To further assess the responsiveness of PtHR97 to candidate ligands, ovarian explants were cultured in vitro and treated with pyriproxyfen (P), methyl farnesoate (MF), and arachidonic acid (AA), respectively. PtHR97 expression showed no significant changes following treatment with P or MF across the tested concentrations (Figure 2C,D). In contrast, AA treatment resulted in a significant downregulation of PtHR97 transcript levels at all examined concentrations (10 mM, 100 mM, and 1 M) (Figure 2E), indicating a specific transcriptional response of PtHR97 to AA in ovarian tissue.

3.3. Effects of PtHR97 Knockdown on Ovarian Development

To investigate the role of PtHR97 in ovarian development, RNA interference was performed in vitellogenic female P. trituberculatus. Hematoxylin–eosin (HE) staining revealed that ovaries from both dsGFP and dsHR97 groups were dominated by oocytes at the exogenous vitellogenic stage (Figure 3A). In the dsGFP group, oocytes exhibited relatively larger sizes, with eosinophilic granules evenly distributed throughout the cytoplasm. Notably, some oocytes exhibited nuclear envelope shrinkage, a morphological feature often associated with rapid oocyte growth and active yolk accumulation. In contrast, oocytes from the dsHR97 group were generally smaller, displayed unevenly distributed eosinophilic cytoplasmic granules with locally enhanced eosinophilia, and maintained relatively well-defined nuclear envelopes. This morphological feature may reflect disrupted cytoplasmic maturation or abnormal accumulation of protein-rich materials during vitellogenesis. Morphometric analysis further demonstrated that PtHR97 knockdown resulted in a significant reduction in gonadosomatic index (GSI) and oocyte diameter, while the nuclear-to-cytoplasmic ratio remained unchanged, suggesting impaired oocyte quality and developmental progression (Figure 3B). Subsequent qPCR analysis confirmed the PtHR97 knockdown efficiency (Figure 3C), and the transcript level of the vitellogenin receptor (VgR), an important marker of ovarian vitellogenesis, was not significantly affected (Figure 3D), suggesting that the morphological changes observed in PtHR97 knockdown ovaries may not be primarily due to alterations in vitellogenin uptake.

3.4. Transcriptomic Profiling and Quality Assessment Following HR97 Knockdown

RNA-seq was conducted on ovarian tissues from dsGFP- and dsHR97-injected female crabs to assess global transcriptional responses to HR97 knockdown. A total of 275.43 million raw reads, corresponding to 41.59 Gb of sequencing data, were generated (Table 1). The number of raw reads per sample ranged from 36.57 to 49.54 million, with a stable GC content averaging 46.78%, indicating good consistency in base composition among samples. Overall sequencing quality was high, with mean Q20 and Q30 values of 98.59% and 96.09%, respectively. After quality filtering, 272.08 million clean reads (40.96 Gb) were retained, accounting for more than 98% of the raw reads across all samples. Subsequent alignment showed an average mapping rate of 93.25% to the reference genome, with minimal variation among samples, demonstrating the reliability of the sequencing data and their suitability for downstream analyses. The distribution of FPKM values was highly consistent among biological replicates within each treatment group (Figure S2), and correlation analysis revealed strong within-group similarities, with correlation coefficients approaching 1.0 (Figure S3). Principal component analysis (PCA) revealed a clear separation between the dsGFP and dsHR97 groups along the first principal component (PC1, 26.8%), indicating that HR97 knockdown induced a distinct global transcriptional shift in ovarian tissue (Figure 4A). Biological replicates within each group clustered closely, supporting the robustness of the transcriptomic data.

3.5. Transcriptomic Responses to HR97 Knockdown

Differential expression analysis identified 134 up-regulated and 179 down-regulated genes following HR97 knockdown. Based on functional annotation, DEGs were manually classified into five major categories, including gene regulation and processing, neuroendocrine signaling, lipid metabolism and redox, membrane transporters, and extracellular matrix (Table 2). The volcano plot shows the overall distribution of DEGs (Figure 4B), in which HR97 itself was significantly down-regulated, confirming the effectiveness of RNA interference at the transcriptomic level. Representative DEGs with large fold changes and high statistical significance are highlighted. A heatmap constructed from selected genes across major functional groups revealed consistent expression differences between dsGFP and dsHR97 samples, with clear clustering by treatment (Figure 4C).

GO enrichment analysis indicated that DEGs were significantly enriched in cellular component, molecular function, and biological process categories (Figure 4D). Enriched cellular component terms were mainly associated with the plasma membrane, cell surface, and extracellular region. Molecular function terms were dominated by transporter activity, channel activity, signaling receptor activity, and kinase-related functions, while biological process terms were enriched in responses to chemical and xenobiotic stimuli, lipid metabolic processes, ion transport, cell–cell signaling, and regulation of responses to external and hormonal stimuli. Because only a limited number of DEGs could be assigned to individual KEGG pathways, KEGG BRITE classification was applied. BRITE analysis showed that HR97 knockdown affected multiple higher-order functional categories, including lipid metabolism, signal transduction, signaling molecules and interaction, transporters, xenobiotics biodegradation and metabolism, cell motility, cellular community–eukaryotes, endocrine system, and protein families involved in signaling and cellular processes (Figure 4E). Overall, GO and KEGG enrichment patterns were consistent with the manual functional classification of DEGs.

3.6. qPCR Validation of DEGs Under Short-Term RNAi and AA Treatment

To validate DEGs identified from the transcriptome, we performed a short-term in vivo RNAi assay targeting PtHR97 and detected the expression of 20 representative genes selected from five functional categories by qPCR (Figure 5). In addition, since AA suppressed HR97 expression in ovarian explants, an AA-treated group was included to confirm whether AA can downregulate HR97 in vivo and to serve as a pharmacological perturbation for comparison. PtHR97 transcripts were significantly reduced in both the dsHR97 and AA groups, indicating effective short-term in vivo manipulation and further supporting AA may suppress HR97 expression. As expected, the directions of change for the selected genes after dsHR97 treatment were broadly consistent with the transcriptomic results. When AA and dsHR97 were compared directly, MED12, NKX2.4, GLRA4, PDH2, SRD5A1, AVT1J, RhB, and MCT2 were downregulated in both groups, whereas LDLRAP1 and LRP1B were upregulated in both, suggesting that these genes may represent downstream outputs of the putative AA–HR97 axis. By contrast, NTF1, sNPFR, AC2, JHEH1, TRET1, and four ECM-related genes (CHAD, MUC2, LACH, and MFAP4) showed clear responses to dsHR97 but did not change significantly after AA treatment, suggesting incomplete overlap between AA- and RNAi-induced transcriptional outputs. Collectively, these short-term in vivo qPCR data further support the robustness of HR97 knockdown–associated transcriptional signatures and help prioritize candidate HR97-responsive targets for future functional characterization.

4. Discussion

The present study cloned the full-length cDNA of HR97 from Portunus trituberculatus (PtHR97) and confirmed its placement within the NR1L nuclear receptor group by phylogenetic analysis. In contrast to Daphnia species, which typically possess three HR97 subtypes (HR97a, HR97b, and HR97g) [5], we detected only a single HR97 sequence in P. trituberculatus. Although our evidence is based on an ovary-derived transcriptome, available annotations across multiple decapod species likewise generally identify only one canonical NR1L/HR97 gene [4]. Decapods also include three additional HR97-like receptors, the molt-associated receptors (MAR1–3), which are assigned to NR1S, NR1Q, and NR1R rather than NR1L [4]. Accordingly, the sequence cloned here represents the canonical HR97 (NR1L) rather than a MAR. MARs show lineage-dependent tissue enrichment, being ovary-biased in the eastern spiny lobster Sagmariasus verreauxi but antennal-gland–enriched in the tropical land crab Gecarcinus lateralis (ovarian expression was not reported) [4]. As we did not detect MAR-like transcripts in the P. trituberculatus ovary transcriptome, PtHR97 likely represents the major detectable HR97-like member in the ovary under the conditions sampled, thereby reducing potential confounding in subsequent expression assays and functional perturbation experiments.

PtHR97 displays the canonical modular architecture of nuclear receptors, including typical DBD and LBD domains. Across HR97 homologs, conservation is strongest in the DBD and weaker in the LBD, while flanking regions are more variable, consistent with general nuclear receptor design principles [28,29]. The PtHR97 DBD retains key DNA-recognition determinants, including canonical P-box/D-box features [30], supporting broadly conserved response-element recognition and downstream transcriptional wiring across crustaceans. In contrast, the LBD shows appreciable interspecific variation, and residues shaping the ligand-binding pocket [31] and co-regulator docking surfaces [32] are not uniformly conserved. Such LBD divergence may contribute to the distinct ligand-dependent activation reported for different HR97 subtypes in Daphnia [5], but it should be interpreted cautiously when accounting for the discrepancy between our results and the Daphnia study. In Daphnia, MF and pyriproxyfen activate HR97 transactivation at high concentrations, whereas in our in vitro system they did not change PtHR97 transcript levels. This difference is not necessarily contradictory because the Daphnia assay reports ligand-dependent receptor activity at the protein level, whereas our explant assay measured PtHR97 transcript levels, and the ligand-driven changes in nuclear receptor activity can occur without parallel shifts in receptor mRNA abundance [33,34]. Thus, we cannot exclude that MF/pyriproxyfen modulates PtHR97 activity at the protein level without detectable changes in PtHR97 mRNA, as protein-level activity was not evaluated in our study. Notably, AA produces a consistent inhibitory trend across the two contexts, highlighting AA as a candidate negative regulator on HR97-related signaling.

As an important unsaturated fatty acid in crustacean ovaries, AA serves as a classical precursor of eicosanoid-derived signaling molecules, including prostaglandins (PGs) [35]. In decapod crustaceans, evidence from exogenous administration and dietary supplementation consistently indicates that AA/PGs can increase the gonadosomatic index, oocyte diameter, and ovarian vitellogenin levels, and are associated with improved reproductive performance and larval quality [14,15,36,37]. Therefore, given that AA exerts an inhibitory effect on PtHR97 expression, the pronounced ovarian developmental impairment observed after PtHR97 knockdown in the present study presents an apparent inconsistency worth discussing. One possible explanation is that AA may influence ovarian development through multiple routes, including the generation of diverse AA-derived lipid mediators and the engagement of intracellular signaling pathways [38,39]. Accordingly, its effects may not be mediated solely by HR97. Indeed, in our short-term in vivo assay, AA did not always elicit downstream expression changes that mirrored those induced by dsHR97. Notably, although dsHR97 caused a marked reduction in GSI and oocyte size, it also significantly upregulated several lipid related genes, including LDLRAP1 and LRP1B. AA treatment similarly upregulated these genes, with an even greater magnitude. We speculate that AA may affect ovarian development partly by suppressing HR97, but additional regulatory pathways may counterbalance the adverse consequences of reduced HR97.

Compared with lipid metabolism, the ovarian enrichment profile after HR97 knockdown points more clearly to changes at the cell boundary, where cells import substrates, sense signals, and interact with their surroundings [40]. GO-CC terms were enriched in cell periphery, plasma membrane, cell surface, and extracellular region. In line with this, GO-MF highlighted transmembrane transporter activities and ion channel/receptor functions, while GO-BP emphasized transmembrane and ion transport, intercellular signaling, cell-junction organization, and responses to hormonal stimuli. Collectively, these patterns may indicate that the deficiency in HR97 may shift the ovary toward reduced transport capacity and perturbed signaling competence, potentially accompanied by remodeling of extracellular features and cell–cell interactions. Consistent with this interface-centered interpretation, oocytes from the dsHR97 group were generally smaller and displayed unevenly distributed eosinophilic cytoplasmic granules with locally enhanced eosinophilia, which may reflect altered substrate exchange and/or disrupted organization of cytoplasmic deposition processes. The KEGG BRITE overview also supports this interpretation, with prominent coverage of high-level categories such as Transporters, Signaling molecules and interaction, Signal transduction, and the Endocrine system. This overall trend also aligns with general features of nuclear receptors, which can act as lipid-soluble ligand sensors and help coordinate transcriptional programs linked to transport and environmental adaptation [41,42,43]. In addition, work in insects and other model systems suggests that oocyte growth relies not only on intracellular biosynthesis, but also on sustained substrate influx, ionic/osmotic homeostasis, and support from the follicular microenvironment, where the extracellular matrix (ECM) helps maintain tissue architecture, permeability, and local signaling [44,45,46]. Interestingly, in the short-term in vivo assay, AA induced opposite expression changes in all four ECM-related genes compared with HR97 knockdown. This consistent inverse pattern suggests that AA may regulate ECM-related processes through an HR97-independent route, and such ECM modulation may represent an important component of how AA supports oocyte growth and ovarian development.

Our results further suggest that HR97 knockdown may alter the ovarian endocrine state, with transcriptomic changes evident in both signal production and signal responsiveness. On the production side, transcripts of several ovary-expressed peptidergic factors were reduced, including GPB5 and PDH2, both of which have been implicated in autocrine/paracrine regulation of ovarian processes [47,48]. On the responsiveness side, the sNPF-system GPCR receptor (sNPFR), which is linked to energy-state control [49], was upregulated, and in P. trituberculatus sNPFR interference attenuates sNPF-induced Vg/VgR upregulation [26], supporting an association with vitellogenic regulation. In addition, adenylyl cyclase (AC), a key effector of the GPCR–cAMP signaling cascade [50], was also upregulated after HR97 knockdown, which is consistent with enhanced receptor-to-second-messenger transduction capacity. This shift was accompanied by changes in multiple receptor/ion-channel–related transcripts, including downregulation of the inhibitory receptor subunit GLRA4 [51] and upregulation of the excitatory glutamate receptor subunits GRIK5 [52] and GRIN2B [53]. Although crustacean neuroendocrine networks remain incompletely resolved and these markers may not capture the full set of endocrine pathways influenced by HR97, the overall profile points to reduced local signal output coupled with heightened signal reception/transduction in the ovary following HR97 knockdown, which may reflect a compensatory and feedback-like adjustment.

It should be emphasized that the current dataset cannot distinguish direct transcriptional effects of HR97 knockdown from indirect, cascade-type responses in the ovary. Nonetheless, the downregulation of MED12 suggests that HR97 deficiency may engage components of the Mediator–Pol II transcriptional machinery. Given the well-established functional coupling between nuclear receptors and the Mediator complex [54,55], MED12 may represent a proximal point at which HR97 influences overall nuclear transcriptional capacity. Meanwhile, the decreased expression of NTF1-like (also known as UBF/UBTF) may indicate alterations in nucleolar rRNA gene transcription and ribosome biogenesis [56], pointing to a broader disturbance of the nuclear gene-expression infrastructure. Consistent with this view, several additional genes with more upstream regulatory implications were also affected. The homeobox factor NKX2.4 and the odd-skipped family C2H2 zinc-finger transcription factor SOB may reflect shifts in ovarian cellular state and associated transcriptional programs [57,58], whereas ZCCHC24 is more plausibly linked to nuclear RNA binding and post-transcriptional processing [59]. Although the current data do not resolve the precise regulatory relationships among these factors, their coordinated changes support the notion that HR97 depletion may act beyond terminal effector pathways by perturbing higher-order layers of transcriptional regulation and RNA processing. This interpretation is also consistent with HR97’s identity as a nuclear transcription factor, and it provides a plausible framework for how HR97 knockdown could broadly influence the expression output required for ovarian progression during vitellogenesis.

5. Conclusions

Taken together, our data suggest that HR97 may contribute to maintaining a relatively coordinated ovarian functional state during vitellogenesis, rather than acting through a single downstream pathway. The transcriptomic response to HR97 knockdown primarily points to two levels of change. First, components related to exchange and signaling at the cell boundary show notable shifts. Second, elements supporting nuclear gene-expression capacity, including the transcriptional machinery and associated processes, appear to be affected. In combination, these changes may reduce the overall efficiency and stability required for sustained oocyte growth and orderly yolk deposition. Considering the AA results within this framework also helps explain the apparent inconsistency. The inhibitory effect of AA on HR97 expression may represent only part of its action, while AA may simultaneously engage parallel routes that do not depend on HR97 and provide compensatory or counterbalancing effects. Therefore, AA-induced responses are not expected to simply mirror those of dsHR97. Overall, HR97 is more likely involved in coordinating ovarian responses to multiple regulatory inputs, whereas the reproductive effects of AA may reflect its combined influences across several regulatory pathways.