Bigfin reef squid were purchased from a fisherman on Heping Island, Keelung. These squids were collected using hand jigging from a boat near Heping Island, located along the northeastern coast of Taiwan. Additionally, juvenile bigfin reef squids were provided as a gift by the Marine Research Station of the Institute of Cellular and Organismic Biology, Academia Sinica.

Sex identification was based on gonadal morphology and the presence of accessory reproductive organs, such as the nidamental gland in females and spematophoric gland in males, as previously described (Lee et al., 2025). When external morphology was insufficient to determine sex, gonadal histology was further used for confirmation. The gonadal status of females, as well as the relationship between oocyte size and developmental stage, was assessed using gross morphological characteristics and ovarian histology, respectively, as described in our previous work (Chen et al., 2018). Male gonadal status was classified as either immature or mature based on the presence of spermatophores. Squids were anesthetized with 5% ethanol in seawater at room temperature, and tissue samples were collected prior to decapitation. All experimental procedures were approved by the National Taiwan Ocean University International Animal Care and Use Committee and conducted in accordance with standard ethical guidelines.

Tissue histology was performed as described in our previous study (Chen et al., 2018). Gonads were fixed in 4% paraformaldehyde for 16 hours, dehydrated in methanol, and stored at -20°C. The tissues were then processed through an ethanol-xylene series and embedded in paraffin. Sections were dissected at 5-µm thickness, rehydrated, and stained with hematoxylin and eosin (H&E).

Nucleic acids were extracted using Trizol reagent (ThermoFisher Scientific), following the manufacture’s protocol. Trizol-homogenized tissues were mixed with chloroform to separate total RNA into the upper aqueous phase. RNA was then precipitated using a high-salt solution and isopropanol. The isolated RNA was quantified using a Nanodrop 2000 spectrophotometer (ThermoFisher Scientific) and used for subsequent cDNA synthesis.

cDNA was synthesized from 1 µg of total RNA using Oligo(dT)_12-18 primers (Company) and Superscript III (Company), following the manufacture’s protocol. The synthesized cDNA was then used for qPCR analysis, performed as previously described (Chen et al., 2018). Gene expression levels were quantified using the CFX Connect™ Real-Time PCR Detection System (Bio-Rad) with SYBR Green Master Mix (Bio-Rad). PCR specificity for each gene was confirmed by the presence of a single melting curve in both experimental samples and template-containing (positive) controls. No signal was detected in the non-templet (blank) controls. Reaction efficiency for each gene was initially assessed using serially diluted standards. The qPCR amplification efficiencies of fas and ef1a were 70.8% and 73.2%, respectively. Because the difference in amplification efficiency between fas and ef1a were lower than 2.5%, the relative expression levels of the fas gene (GeneBank accession no. PX526114) were calculated using the 2^-ΔΔCt method. Elongation factor 1 alpha (ef1a; GeneBank accession no. MG924746) was selected as the internal reference gene for qPCR normalization based on its narrow Ct distribution across all tissues (16.2 ± 0.75). The specific qPCR primers are listed in Table 1.

Digoxigenin-11-UTP (DIG)-labeled antisense and sense probes were synthesized from target fragments of the fas genes (Table 1). The DIG-labeled antisense probe was used to detect target mRNA localization, while the sense probe served as a negative control to assess non-specific signal. Wholemount in situ hybridization (WISH) was performed as our previous description (Li et al., 2022a). Fixed samples were rehydrated and sectioned into 0.1 mm slices using a vibratome (microslicer DTK-1000, Dosaka), followed by standard ISH procedures. Samples were incubated overnight at 50-60°C with DIG-labeled probes. mRNA expression was detected using an ovary-pre-adsorbed, alkaline phosphatase-conjugated sheep anti-DIG antibody (11093274910, Merck), and visualized using the NBT/BCIP Detection System (11681451001, Roche). Following color development, samples were post-fixed in 4% paraformaldehyde, dehydrated in ethanol, and embedded in Technovit 3040 plastic resin. Sections were dissected at 5-µm thickness, rehydrated, and counterstained with Nuclear Fast Red (H-3403-500, Vector Laboratories) to visualize cell nuclei.

Polyclonal antiserum was generated by immunizing rabbit with a synthetic peptide (H₂N-CNKPLKIEVIDGNHESFIQGEYAQK-COOH) corresponding to the C-terminal region of bigfin reef squid FAS, conjugated to ovalbumin. The immunization and antiserum collection were performed by Yao-Hong Biotechnology Inc. The specificity of the antiserum was further confirmed by Western blot (WB) analysis using tissue samples expressing the target protein.

Protein analysis was performed as described in our previous study (Chen et al., 2018). Samples were homogenized in cell lysis buffer containing 25 µM Phenylmethanesulfonyl fluoride (PMSF) and cOmplete™ Mini protease inhibitor cocktail (4693124001, Roche) and centrifuged, and the supernatant was used for subsequent analysis. Protein concentrations were determined using a Pierce™ bicinchoninic acid (BCA) protein assay kit (A65453, ThermoFisher Scientific). Equal amounts of protein (up to 20 µg per sample) were separated on a 5% SDS-PAGE gel. For WB analysis, the separated proteins were transferred onto a 0.45 µm nitrocellulose membrane. After washing, membranes were blocked with 5% nonfat milk powder for 1 hour and then incubated overnight at 4°C with primary antibody, anti-FAS antiserum (1: 2000 dilutions in 1.5% nonfat milk). The membranes were subsequently incubated with an alkaline phosphate-conjugated goat anti-rabbit IgG secondary antibody (31340, ThermoFisher Scientific; 1: 10,000 dilutions in 1.5% nonfat milk) at room temperature for 1 hour. Protein expression was visualized using the 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium chloride solution (BCIP/NBT) Detection System (B1911, Merck). As a specificity control, the anti-FAS antiserum was preabsorbed with 1 µg/ml of the peptide antigen and used in place of the primary antibody.

Immunohistochemical (IHC) staining was performed as previously described (Chen et al., 2018). Section preparation performed the methods described in this study. For IHC staining, sections were treated with HistoVT One (Nacalai Tesque) to expose the antigens and then incubated with 3% H₂O₂ to block the endogenous peroxidase activity. To detect nucleus-incorporated bromodeoxyuridine (BrdU), sections were further treated with 2N HCl to dissociate proteins bound to chromatin. After blocking with 5% nonfat milk powder for 1 hour, sections were incubated overnight at 4°C with primary antibody: anti-FAS antiserum (1: 2,000 dilutions with 1.5% nonfat milk powder) or anti-BrdU (MAB4072, Sigma-Aldrich; 1:1,000 dilution with 1.5% nonfat milk powder). The next day, sections were incubated with a biotinylated secondary antibody (1:1,000 dilution with 1.5% nonfat milk powder) at room temperature for 1 hour. Protein localization was visualized using the Avidin-Biotin Complex (ABC) kit (Vector Laboratories) and 3,3’-diaminobenzidine (DAB, Sigma-Aldrich), followed by hematoxylin counterstaining. As a specificity control, the anti-FAS antiserum was preabsorbed with 1 µg/ml of the peptide antigen and used in place of the primary antibody.

The FAS activity assay was performed as previously described with minor modification (Ross et al., 2008). Oocytes of various sizes with associated follicle cells were independently isolated and stocked at -80°C. Samples were homogenized in lysis buffer, centrifuged (16,000 x g, 15 minutes), and the resulting supernatant was used for analysis. Protein concentrations were determined using a BCA kit. For the assay, supernatant was added to a reaction mixture containing 200 mM potassium phosphate buffer (pH 7.0), 400 mM EDTA, 240 µM NADPH, and 30 µM acetyl-CoA, and incubated at 30°C for 10 minutes. Background NADPH oxidation was measured prior to the addition of malonyl-CoA. The enzymatic reaction was initiated by adding 100 µM of malonyl-CoA, and the decrease in optical density (OD) was recorded at 5 minutes intervals over 30 minutes using a microplate reader at 340 nm.

Due to limitations in surgical techniques and the need for accurate identification of sexual characteristics, one-month-old squids were used for organ culture. Sex determination was based on relative gonadal size, with females exhibiting larger gonads than males. Specimens with large gonads (presumed ovaries) were selected for organ culture (n = 9 per group). Three ovarian samples were placed in each well of a 6-well plate. Ovaries were maintained in M199 medium supplemented with 3% (w/v) sea salt, 5% fetal bovine serum, 100 IU penicillin, and 100 ug/ml streptomycin. Notably, ovarian tissue exhibited widespread structural degradation after one week in culture (Supplementary Figure S1).

Cell proliferation in the ovary was assessed by the incorporation of BrdU. Ovaries were incubated with BrdU (60 µg/ml) for 24 hours prior to sampling under two different conditions: control and FAS function inhibition groups. To evaluate the inhibiting efficiency of the FAS inhibitor TVB-3166 (SML1694, Sigma-Aldrich), different concentrations ranging from 200 nM to 1.6 µM were tested using juvenile female ovaries containing primary oocytes. The FAS activity assay revealed that enzymatic activity was reduced by approximately 50% in the 200 nM treated group compared with the control. Increasing the concentration beyond 200 nM produced no further significant suppression of FAS activity. Therefore, in the FAS inhibition group, ovarian tissues were incubated with the FAS inhibitor TVB-3166 (1 µM) to inhibit FAS activity. Treatments were administrated for two-day intervals, and ovaries were collected on day 5 for analysis. Cell proliferation was evaluated using IHC staining as described in this study. The number of BrdU-incorporated cells was quantified to assess proliferative activity. Ovarian cell proliferation was represented as the relative number of BrdU-positive germline and somatic cells, respectively. For each group, three tissue samples were analyzed, with triplicate stained (4-6 serial sections on each slide) per sample.

The data are presented as the mean ± standard deviation (SD). Prior to conducting parametric tests, all datasets were evaluated for normality using the Shapiro-Wilk test and for homogeneity of variance using Levene’s test. Values that met these assumptions were analyzed by one-way ANOVA, followed by a Tukey post hoc test, with P < 0.05 considered statistically significant. For comparisons between two groups, Student’s t-test was applied (P < 0.05), provided that normality and equal variance assumptions were satisfied. When assumptions were not met, appropriate non-parametric test, such as the Kruskal-Wallis test, were used.

The sequence of fas gene was identified from the genome-annotated gene database (In preparation) of the bigfin reef squid. To confirm the sequence, the full-length cDNA was obtained by cloning. The gene contains an open reading frame encoding 3,043 amino acid residues. The deduced amino acid sequence includes conserved functional domains, including the ketosynthase (KS; 556-1026 aa), acyltransferase (AT; 1047-1362 aa), dehydratase (DH; 1450-1651), enoylreductase (ER; 2089-2405 aa), ketoreductase (KR; 2434-2613 aa), acyl carrier protein (ACP; 2676-2744 aa), and thioesterase (TE; 2784-3038 aa) (Figure 2A), which are structurally similar to those found in other species (Lin et al, 2024).

Tissue distribution of the fas gene in a mature female showed predominant expression in the ovary, with weak expression in other tissues, including the lipid-rich digestive gland (Figure 2B). To investigate the reproductive-associated expression pattern of fas, qPCR analysis was performed on gonads and digestive glands from both sexes at immature and mature stages. The result showed that fas expression was female-biased in the gonads, with the highest expression observed immature ovaries (Figure 2C). In contrast, fas expression in the digestive gland remained unchanged across reproductive stages in females (Figure 2C). These findings indicate that fas mRNA is specifically expressed in the ovary and not in lipid-rich tissues such as the digestive gland, which functions similarly to the hepatopancreas in invertebrates and liver in vertebrates.

To investigate the expression profile of fas during the reproductive cycle, ovaries at different development stages of ovaries were analyzed for fas mRNA levels. Based to the histological characteristics described in a previous study (Chen et al., 2018), four ovarian stages were defined in this study: juvenile females containing primary oocytes, immature females with previtellogenic oocytes, maturing females with early vitellogenic oocytes, and mature females with late vitellogenic oocytes. qPCR analysis revealed that fas mRNA expression was highest in juvenile females, decreased gradually during ovarian growth, and reached its lowest level in mature females (Figure 3A).

We next examined the fas mRNA expression profile across different oocyte size classes, which correspond to specific developmental stages as described previously (Chen et al., 2018). qPCR results showed that fas expression was high in oocytes smaller than 1 mm, including oogonia, primary oocytes, and multiple follicular oocytes (Figure 3B). Expression levels declined significantly in oocytes ranging from 1.5 - 4 mm, which include previtellogenic (1.5-2 mm), early vitellogenic oocytes (2.5-3 mm, and some late vitellogenic oocytes (3.5-4 mm) (Figure 3B). The lowest expression levels were observed in oocytes larger than 4.5 mm, corresponding to late vitellogenic oocytes (4.5-5 mm) and fully mature (ripe) oocytes (>5 mm) (Figure 3B). Consistently, FAS protein exhibited the lowest fatty acid synthesis activity in oocytes larger than 3 mm, including late vitellogenic oocytes and ripe oocytes (Figure 3C). Together, these findings suggest that fas mRNA expression and its enzymatic activity are not positively correlated with de novo fatty acid synthesis during oocyte vitellogenesis.

To examine fas mRNA expression in specific gonadal cell types, WISH was performed on juvenile individuals, including ovaries at the primary oocytes stage, when fas transcript levels at high, and testes at the spermatogonia stage. In the ovary, WISH results showed strong fas mRNA expression in primary oocytes, with slight to no expression detected in surrounding somatic cells (Figure 4A). In the testis, fas mRNA expression was weak or nondetectable in both somatic and germline cells (Figure 4B). No signal was detected in testis and ovary with sense probe, respectively (Figures 4C, D). These results indicate that fas mRNA is specifically localized on oocytes.

Based on analysis using the ExPASy compute pI/Mw tool (https://web.expasy.org/compute_pi/), theoretical molecular weight of the FAS protein was estimated to be approximately 337-kDa. The specificity of the anti-FAS antiserum was validated by WB analysis using oocyte extracts containing follicle cells from oocytes ranging in size from 1 to 2 mm. Immunoblotting revealed a single band larger than 240-D, corresponding to the predicted size of the FAS protein (Figure 5A). Notably, our protein ladder does not contain markers above 240-kDa, suggesting the detected band represents a high molecular weight protein. Furthermore, the FAS immunoblot signal was greatly reduced or absent when the antiserum was preadsorbed with its antigenic peptide (Figure 5B), confirming the specificity of the antibody. This result indicates that the anti-FAS antiserum specifically recognizes the FAS protein.

To examine the cellular localization of FAS protein, IHC staining was performed using a specific anti-FAS antiserum. In ovaries from juvenile females, FAS signals were strongly detected in primary oocytes, with weaker or nondetectable staining observed in oogonia (Figure 5C). In maturing females, strong FAS signals were observed in primary oocytes and multiple follicular oocytes, while weaker expression was detected in previtellogenic oocytes and oocytes at later stages (Figure 5D). No signal was detected when using the antigen-preabsorbed anti-FAS antiserum, confirming antiserum specificity (Figures 5E, F). Unexpectedly, IHC results show very weak FAS signals in previtellogenic oocytes, despite qPCR and enzymatic activity assay indicating a decrease in fas mRNA and FAS activity during later oocyte stages. This discrepancy may be due to the substantial increases in oocyte size, which could dilute the protein signal and reduce detection sensitivity.

To further investigate the potential role of FAS in ovarian development, juvenile female ovaries were treated with TVB-3166. After incubation, histological analysis results revealed signs of tissue degradation in approximately 50% of ovaries across all groups. Only samples (n = 3 per group) that retained the most intact structures were selected for subsequent analyses. Cell proliferation was examined by BrdU incorporation using IHC staining. IHC revealed that BrdU-positive signals were mainly localized to somatic cells, with minimal labeling in germline cells across all groups (Figures 6A, B). The number of BrdU-positive somatic cells significantly increased in the FAS inhibition group (TVB-3166 treatment) compared to the control (Figure 6C). No significant difference in BrdU-positive germline cells were observed among the groups (Figure 6C). Altogether, these results suggest that de novo fatty acid synthesis in oocytes may regulate the proliferation of surrounding somatic cells.

Our findings indicate that fas mRNA is predominantly expressed in the ovary compared to other tissues, including the lipid-rich digestive gland, in squid. In contrast, fas mRNA is generally expressed at high levels in lipid-rich tissues such as the liver and adipose tissue in vertebrates (Cui et al., 2012; Mildner and Clarke, 1991; Peng et al., 2017; Sakae et al., 2020), and in the hepatopancreas and fat body of invertebrates (Nie et al., 2023; Song et al., 2022). Notably, only one study in crustaceans reported lower fas expression in the hepatopancreas compared to other tissues (Zuo et al., 2017). Therefore, ovary serving as the primary site of fatty acid synthesis in squid represents a unique case among animal species.

In most animals, excess energy derived from dietary intake is converted into lipid stores in lipid-rich tissues such as the liver in vertebrates (Arukwe and Goksøyr, 2003) or the hepatopancreas and fat body in invertebrates (Lee and Chang, 1999; Osada et al, 2004; Pan et al, 1969). These tissues function as major energy reservoirs that support reproduction, particularly during vitellogenesis in females. In contrast, cephalopods exhibit a distinct strategy: dietary lipids are rapidly transported from the digestive gland to other tissues for growth rather than being stored for later use in reproduction (Semmens, 1998). This immediate-use strategy may represent an evolutionary adaptation that promotes energy supply to oocytes through de no fatty acid synthesis.

Another possible reason for the active de novo fatty acid synthesis in squid oocytes may be linked to their unique folliculogenesis process. Unlike in many other animals, cephalopod oocytes are not only surrounded by follicle cells but also invaded by them, forming a special structure known as a follicular fold (Chen et al., 2018). This invasive process likely causes a substantial expansion of the oocyte’s surface area, thereby requiring enhanced lipid synthesis to support membrane formation. In mammals, FAS-synthesized fatty acids are known to play essential roles in membrane biosynthesis, particularly in rapidly proliferating cells such as cancer cells (Chen et al., 2014). Together, these findings suggest that the squid ovary serves as a primary site of fatty acid synthesis, likely supporting both oocyte development and membrane biogenesis in the absence of dedicated lipid storage tissues.

Specifically, we found that fas mRNA expression levels are elevated during the primary oocyte stage but significantly reduced at the later multiple follicular oocyte stage. Furthermore, inhibition of FAS activity at the primary oocyte stage promoted proliferation of ovarian somatic cells. Taken together, these findings suggest that reduced FAS activity, and the consequent decrease in fatty acid synthesis during oocyte development, may serve as a regulatory signal to enhance somatic cell proliferation.

Yolk protein accumulation in oocytes plays a crucial role in energy storage for embryogenesis and larva survival in oviparous animals. The primary precursor of yolk protein is vitellogenin (VTG) (Hayward et al., 2010). In most vertebrates (Arukwe and Goksøyr, 2003; Guiguen et al., 2010) and insects (Pan et al., 1969; Wojchowski et al., 1986), VTG is synthesized in lipid-rich organs such as the liver and fat body, respectively. In other invertebrates, VTG mRNA expression has been observed in extraovarian tissues (e.g., the lipid-rich hepatopancreas) and/or the ovary itself (Review in Chen et al., 2018). Notably, in cephalopods, VTG mRNA expressions are specifically localized on the ovary. VTGs synthesized by ovarian follicle cells are accumulated in oocytes during vitellogenesis (Chen et al., 2018; Gaudron et al., 2005; Kitano et al., 2017). However, VTGs are absent from the hemolymph in mature female squids (Kitano et al., 2017), suggesting that VTGs are immediately transferred from follicle cells to oocytes without entering the systemic circulation.

In cephalopod, lipids account for a relatively small proportion (~15%) of the egg yolk compared to total proteins and carbohydrates (Matozzo et al., 2015). Although our findings show that oocytes are a major site of fatty acid production, with fas mRNA and protein highly expressed at early oocyte stages, these levels progressively decline during vitellogenesis and reach their lowest in late vitellogenic oocytes. In contrast, VTGs mRNA and protein expression are low in early oocytes but increase significantly during vitellogenesis (Chen et al., 2018; Kitano et al., 2017). These results suggest that VTGs synthesized by ovarian follicle cells are the primary contributor to energy storage in cephalopod eggs, whereas fatty acids synthesized by oocytes play a minimal role in this process.

In contrast to the ovary, which exhibited markedly higher fas mRNA expression than the lipid-rich digestive gland in females, fas expression in the testis was comparable to that in the digestive gland in males. These results indicate that the ovary serves as a primary site of de novo fatty acid synthesis in females, whereas this lipogenic pathway is not preferentially activated in testis. Consistent with this interpretation, higher lipid content has been reported in the digestive gland of females compared with males in both octopus (Estefanell et al., 2015; Rosa et al., 2004) and bigfin reef squid (our unpublished data). In contrast, lipid content was higher in the testis than in the ovary in octopus (Estefanell et al., 2015). Collectively, these findings suggest that, in males, gonadal lipid accumulation may rely primarily on lipid transport from the hepatopancreas rather than local de novo synthesis, whereas females preferentially activate ovarian lipogenesis to support oocyte development.

Further evidence for sex specific metabolic strategies comes from differences in fatty acid composition. In octopus, males exhibited higher levels of oleic acid (18:1 n-9), whereas females showed elevated eicosapentaenoic acid (20:5 n-3) in the digestive gland, independent of feeding regime (Estefanell et al., 2015). Moreover, arachidonic acid (ARA), docosahexaenoic acid (DHA), and palmitic acid (16:0) were predominant in testes, whereas ovaries were enriched in 16:0, DHA, and ARA (Estefanell et al., 2015). Together, these observations support the existence of sex-dependent lipid metabolic pathways in cephalopods, in which male gonad development may preferentially utilize lipid uptake and remodeling pathways rather than de novo fatty acid synthesis.

Achieving full-life-cycle aquaculture of the bigfin reef squid remains a major challenging, primarily due to the unreliable production of viable offspring in captivity. This includes issues such as precocious females laying small eggs, abnormal egg case structures, and the resulting low survival rate in subsequent captive generations. Unlike in vertebrates, where the roles of estrogen and androgen in sexual reproduction is well established, the functions of vertebrate-type sex steroids in invertebrates remain poorly understood (Fodor et al., 2020). To date, no studies in cephalopod have reported the regulatory pathway governing puberty and later sexual maturation in both sexes. In this study, we demonstrate that ovary serves as a primary site for fatty acid synthesis in cephalopods. While our results suggest a potential link between fatty acid synthesis and somatic cell proliferation, this relationship is currently correlative, and future studies are needed to establish causality. Comparative transcriptome analysis of the lipid-rich digestive gland and ovary may further clarify differences in lipid metabolism-related processes between these tissues. Although fas/FAS expression profiles during oocyte maturation were examined, the lipid content of eggs were not measured, limiting the direct extrapolation of our finding to lipogenesis during oocyte maturation. Consequently, while these findings provide a basis for exploring metabolic influences on reproductive physiology, any links to abnormal egg case structure or reduced survival in captive squid remain speculative and require further investigation.