Multi-omics unravels testis-regulated genetic mechanism of musk secretion in muskrat (Ondatra zibethicus) Authors: Bing Song, Tingting Hong, Chenmiao Liu, Jie Wu, Shuhui Wang, Zhanjun Ren Published: 2026 ABSTRACT AbstractBackgroundDuring the scent secretion period, adult male muskrats (Ondatra zibethicus) produce a musk-like fragrance similar to that of the forest musk deers (Moschus berezovskii). However, the underlying mechanism of scent secretion in muskrats were not well understood. Scent glands and testicular tissues from 2-year-old muskrats during non-scenting and scenting periods were analyzed using transcriptomics and metabolomics to identify key genes, metabolites, and pathways involved in scent secretion.ResultsKey findings include: (1) Significant differences were observed in the tissue structure of scent glands and testes between the two phases. (2) In the scent glands, fifteen candidate genes were identified that primarily regulate the formation of major musk components through steroid biosynthesis, unsaturated fatty acid (UFA) synthesis, glyceride metabolism, cholesterol metabolism, and aldosterone-regulated sodium reabsorption pathways. (3) In the testes, seven key genes were identified that regulate muskrat sex hormone synthesis and action, thereby controlling musk secretion via steroid hormone biosynthesis, GnRH signaling, estrogen signaling, prolactin signaling, cholesterol metabolism, and aldosterone synthesis and secretion pathways. (4) Transcriptomic and metabolomic association analysis revealed that cholesterol metabolism, UFA biosynthesis, and fatty acid metabolism pathways played key roles in regulating scent secretion activity. Meanwhile, during the scent-secreting period, the testes increase the level of phosphatidylethanol through genes Itpr1, Plcb2, and Prkca, influencing the transport and biosynthesis processes of fatty acids and cholesterol in the scent glands via GnRH signaling transduction and prolactin signaling transduction pathways. In addition, the testes regulate the expression of Hsd17b3 and Cyp3a in the scent glands through genes Cyp11a and Hsd11b2 via the steroid hormone biosynthesis pathway, promoting the synthesis of sterols and steroids from cholesterol in the scent glands.Conclusions This study discovered that the testes regulate the synthesis and transport of key components in muskrat musk within the scent glands through genes (Itpr1, Plcb2, and Prkca, Hsd11b2, and Cyp11a) via GnRH signaling transduction, prolactin signaling transduction, and steroid hormone biosynthesis pathways.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12864-026-13019-9.Keywords: Muskrat, Scent gland, Testis, Genetic mechanism, Musk secretion, Omics correlation analysis. Background The muskrat (Ondatra zibethicus) is a semi-aquatic mammal that belongs to the order Rodentia and family Cricetidae [1]. Through natural dispersal and intentional introduction, muskrats are now widely distributed across North America, Europe, and Asia [2–4]. Adult male muskrats produce musk-like secretions during their scent-secreting periods, which makes them economically valuable [5]. These secretions contain key chemical compounds such as normuscone, fatty acids, and steroids, that exhibit pharmacological effects comparable to natural musk [6–8]. Musk is an essential ingredient in the perfume, pharmaceutical, and cosmetic industries. However, the unregulated harvesting of natural musk has severely endangered species like Chinese forest musk deer (Moschus berezovskii), creating an urgent need for sustainable substitutes [9]. Muskrat musk presents a promising solution due to its similar chemical composition and secretion patterns, which closely resemble those of forest musk deer [10–14]. Moreover, as domesticated animals, muskrats are not subject to wildlife protection regulations, allowing for broader utilization. Investigating the mechanisms of muskrat scent secretion holds significant potential for sustainable resource utilization while contributing to the conservation of endangered species. The anatomical location, morphological structure, and seasonal changes of the muskrat’s musk gland are highly similar to those of the forest musk deer, with the musk-secreting ability of both species closely linked to sex hormone levels [15, 16]. Muskrats exhibit distinct seasonal breeding patterns, with the breeding season typically occurring from March to August. Adult males produce musk-like secretions during the scent-secreting period, which is closely synchronized with testicular development and elevated sex hormone levels. The scent glands and testes enlarge simultaneously during the breeding season and regress during the non-breeding season, indicating strong hormonal regulation of male reproductive and scent-marking behaviors [17]. Therefore, this growth pattern suggests that the musk gland may share similar gene expression profiles with the testis, as observed in the forest musk deer [18]. However, the seasonal expression profiles of the musk gland and testis during different reproductive periods remain largely unexplored. The development and function of the musk gland are likely regulated by testosterone produced in Leydig cells of the testis [17, 19]. Additionally, the androgen receptor (AR), along with the cooperative effects of prolactin (PRL) and androgens, may influence male courtship behavior by modulating the function of scent glands [20, 21]. Endogenous lipid metabolites may also play a role in regulating seasonal changes in musk gland activity across reproductive stages [5]. Despite these findings, the exact metabolites and metabolic pathways mediating the musk gland-testis relationship remain poorly understood. To bridge these knowledge gaps, future research should analyze musk gland and testicular tissues at various reproductive stages using transcriptomics and metabolomics. In the investigation of musk secretion in muskrats, transcriptomic technologies have elucidated the gene expression profiles of musk gland tissues. This research has identified pivotal genes associated with secretion activity and has provided insights into the alterations in regulatory pathways within the cells of the musk gland [22]. For example, Li et al. (2017) compared gene expression profiles of musk glands across different periods and found significant differences in the expression levels of key genes related to calcium ion and TGF-β signaling pathways between March and November [12], emphasizing the importance of apoptosis in musk gland changes. The atrophy and inhibition of musk glands during the non-secreting phase may be associated to decreased growth hormone receptor and insulin-like growth factor expression, with cell apoptosis and pyroptosis contributing to their rapid shrinkage [23, 24]. Metabolomics complements transcriptomics by highlighting subtle changes in gene and protein expression at the metabolite level. This approach deepens our understanding of musk gland cellular functions, identifies key metabolites associated with musk secretion, and reveals the metabolic characteristics of musk glands [25]. By integrating transcriptomic and metabolomic analyses, researchers can achieve a more comprehensive understanding of the molecular mechanisms driving musk secretion [26, 27]. The objectives and significance of this study were as follows: (1) This research aimed to explore the mechanisms underlying musk secretion in muskrats and the regulatory role of testis in scent gland secretion activity, contributing to the improvement of the yield and quality of muskrat musk. The study analyzed scent gland and testicular tissues from muskrats during both scent-secreting and non-secreting periods to identify candidate genes and key pathways related to musk secretion. This approach revealed the musk secretion mechanism in muskrats at the transcriptional and expression levels. (2) The study provided a reference for using muskrats as a model animal to replace the endangered forest musk deer in research. As farmed animals with abundant resources, muskrats offered a feasible model for scientific studies on musk species. This foundation guided musk production, enhanced musk yield and quality, and contributed to the conservation of endangered wild musk species. (3) Most importantly, DEGs and differential metabolites (DMs) provided critical information to the scientific community and accelerated research on other mammals with scent glands, such as forest musk deer, civets, and beavers. Materials and methods Experimental animals and tissue collection Twelve healthy, two-year-old muskrats at their peak scent-producing age were donated from Jinmu Muskrat Farm located in Xinji, Hebei, China. The animals were acquired in December 2021 (during the non-scent-producing phase, denoted as N phase, n = 6) and June 2022 (during the scent-producing phase, denoted as S phase, n = 6). After a two-day acclimation period, the scent glands (referred to as GN and GS groups) and testes (referred to as TN and TS groups) were collected from six muskrats in each group (Fig. S1). They were euthanized by intravenous injection of pentobarbital sodium (30 mg/kg body weight), followed by exsanguination. All procedures were performed in accordance with approved protocols (Permit No: 202205A29) by the Northwest A&F University Animal Care and Use Committee. Using sterile surgical instruments, both testes and scent glands were separated and collected from each side of the muskrats. One side’s tissue was immediately placed in a 4% paraformaldehyde solution for histological observation, while the other side was cut into small pieces, individually packaged, snap-frozen in liquid nitrogen, and then transferred to a -80℃ ultralow temperature freezer for storage. Due to the smaller size of the scent gland tissue during the non-scent-producing period, excess tissues such as fascia and fat adhered to the surface were removed to minimize interference. Additionally, the proximal and distal ends of each tissue sample were marked to ensure sample consistency. Morphological and histological observation The length, width, thickness, and weight of the scent glands and testes were measured during both the scent-secreting and non-secreting periods, with results expressed as mean ± standard deviation (Mean ± SD). These morphological parameters were statistically analyzed using IBM SPSS Statistics V21.0 (Armonk, NY: IBM Corp.) with an independent samples t-test to evaluate differences between the two periods. Additionally, paraffin-embedded sections of muskrat scent gland and testis tissues from the respective periods were stained with hematoxylin and eosin (H&E) for histological observation of tissue morphology. Transcriptome sequencing and analysis Due to the unavailability of a complete genome assembly or annotation for the muskrat, the qualified libraries were sequenced using reference-free eukaryotic transcriptome sequencing on the Illumina NovaSeq 6000, generating 150 bp paired-end reads. Raw data were filtered to remove reads with adapters, ambiguous bases (N), and low-quality reads using fastp v0.19.7 [28]. Clean data quality was assessed based on Q20, Q30, and GC content. Q20 and Q30 indicate base error probabilities of 1% and 0.1%, respectively, with higher values reflecting greater base accuracy. GC content (%) measures the proportion of guanine (G) and cytosine (C) bases, with quality standards typically requiring Q30 > 85% and GC content between 40 and 50%. Clean reads were assembled using Trinity v2.4.0 [29] to generate reference sequences, which were then hierarchically clustered using Corset v1.09 [30] to obtain the unique reference sequences (unigenes). Unigenes were used for functional annotation, CDS prediction, and expression quantification. Clean reads from each sample were mapped to the unigenes using RSEM v1.2.15 [31], and reads with mapping quality scores below 10, unpaired mapped reads, and reads mapped to multiple genomic regions were filtered out. Gene expression was quantified as FPKM (Fragments Per Kilobase of transcript sequence per Million base pairs sequenced) [32]. DESeq2 v1.20.0 [33] was used to perform inter-group differential expression analysis on the raw read counts of expressed genes in the scent glands and testes with |log2(Fold Change, FC)| > 1 and P < 0.05 between the scenting and non-scenting periods. DEG functions were annotated using seven databases: Nr [34], Nt (https://www.ncbi.nlm.nih.gov/nucleotide/), Pfam [35], KOG [36], Swiss-Prot [37], KEGG [37], and GO [38]. The expression of annotated DEGs was analyzed, and their overlap among groups was displayed in a venn diagram. RT-qPCR validation RT-qPCR was performed to verify the differentially expressed mRNA. Total RNA was extracted, quality-tested, and reverse-transcribed using the Evo M-MLV Reverse Transcription Kit (Hunan Aikerui Biotechnology Co., Ltd.), which included gDNA Clean Reagent to remove genomic DNA contamination from the RNA samples. The CDS sequences of differentially expressed mRNA were obtained, and primer sequences were designed using the NCBI Primer-BLAST online tool. Primers were synthesized by Sangon Biotech (Shanghai) Co., Ltd., using GAPDH as the reference gene. The SYBR® Green Premix Pro Taq HS qPCR Kit (Hunan Aikerui Biotechnology Co., Ltd.) served as the fluorescent dye for RT-qPCR analysis, with an amplification reaction system volume of 10 µL. Each sample was tested in three technical replicates, and relative quantification was conducted using the comparative Ct method. The 2−ΔΔCt method was used to analyze the relative expression of each gene in different groups, and the results were normalized using log2(FC) for comparison with the transcriptome sequencing results. For the comparison of gene expression levels of the same DEG between the scenting group and non-scenting group, the group with the lower expression level was used as the reference and normalized to a value of 1. For the comparison of expression fold changes of the same DEG between RT-qPCR and transcriptome data, the values obtained from both methods were log2(FC)-transformed to assess whether the expression trends were consistent. Metabolomic sequencing and analysis Identification of metabolite category and function Scent gland and testicular tissues were collected from muskrats during scent-releasing and non-scent-releasing periods and sent to Beijing Novogene Co., Ltd. for non-targeted metabolomic sequencing. The category and function of the obtained metabolites were analysed using the KEGG database (https://www.genome.jp/kegg/pathway.html), HMDB database (https://hmdb.ca/metabolites), and LIPID MAPS database (http://www.lipidmaps.org). Screening and analysis of DMs MetaX v2.0.0 [39] was used to convert the annotated metabolomics data, followed by principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA). A PLS-DA model was constructed for metabolite levels and sample classification to test whether the samples were correctly classified and to validate the model’s accuracy through ranking verification. Additionally, the data were permuted 200 times and modeled to verify the model’s accuracy through ranking. Screening of DMs in the scent gland and testis of muskrats involved FC, P-value, and Variable Importance in the Projection (VIP) as the main reference parameters. DM screening was conducted with thresholds of VIP > 1.0, P-value < 0.05, and FC > 1.5 or FC < 0.667. Here, FC represents the ratio of the mean quantitative value of each metabolite between comparison groups; P-value indicated the significance level of difference calculated by T-test, with a threshold of 0.05; and VIP value represented the contribution of the metabolite to the grouping, with VIP > 1 generally considered to validate the analytical model. The obtained DMs were then subjected to KEGG enrichment analysis to investigate metabolites’ functions and pathways. The pearson correlation coefficient between all pairs of metabolites was calculated to examine whether the trends of metabolite changes were consistent. A positive correlation approached 1, represented by a brownish color in the illustration, when the linear relationship between two metabolites strengthened, while a negative correlation approached − 1, shown in blue. Additionally, a threshold of P-value < 0.05 was used for significant correlations, and a correlation analysis chart was created for the top 20 metabolites with the most significant correlations. Comparison of KEGG annotation and enrichment for DEGs and DMs DEGs and DMs obtained from the two omics analyses were mapped to the KEGG database. By comparing the obtained pathways, the main biochemical and signal transduction pathways commonly involved by DEGs and DMs were identified. Further screening was conducted based on the significance of enrichment (P < 0.05), and the results were sorted by P-value in ascending order. The top 20 DEGs and DMs in the positive and negative ion modes were presented, highlighting the major biochemical and signal transduction pathways they commonly participate in. Results Seasonal morphological and histological changes in scent glands and testes During the scent secretion period, the scent glands were full, swollen, and had a moist surface covered with granular protrusions (Fig. 1a, c). In contrast, during the non-secretion period, the scent glands were significantly shriveled, difficult to observe, and covered with fat and fascia (Fig. 1b, d). The length, width, thickness, and weight of the GS group were significantly higher than those of the GN group, respectively (P < 0.001) (Fig. 1e-h). Similarly, during the scent secretion period, muskrat testes were full and swollen. From the secretion period (Fig. 2a, c) to the non-secretion period (Fig. 2b, d), there was a noticeable shrinkage, with a significant reduction in size and weight. The length, width, thickness, and weight of the TS group were significantly higher than those of the TN group, respectively (Fig. 2e-h), reflecting coordinated growth with scent glands during the breeding season. Comparison of the morphological structure of muskrat scent glands during (a, c) secretion and (b, d) non-secretion periods. Comparison of gland (e) length, (f) width, (g) thickness, and (h) weight between secretion and non-secretion periods Comparison of the morphological structure of muskrat testicular tissue during (a, c) secretion and (b, d) non-secretion periods. Comparison of testis length (e), width (f), thickness (g), and weight (h) between secretion and non-secretion periods Compared to the GN group, the GS group showed denser and more developed stromal cells surrounding the glandular cells, as well as a greater abundance of epithelial cells on both sides of the ducts (Fig. 3b). In contrast, the GN group exhibited significantly thicker muscle and connective tissues surrounding the glandular cells (Fig. 3a). Furthermore, during both periods, the glandular cells formed spherical clusters; however, in the GN group, these cells were smaller, with round, compact nuclei scattered throughout the tissue. In the GS group, the glandular cells appeared larger, featuring light-colored, homogeneously stained regions or granular substances matching the cytoplasmic color, along with numerous vacuolar structures that lacked visible nuclei. Additionally, the nuclei of the glandular cells in the GS group were mostly spindle-shaped or oval, larger in size, and often displaced toward the periphery of the cells. In the TN group, the seminiferous tubules of the testes were significantly retracted and sparsely distributed (Fig. 3c). Similarly, when compared at a 50 μm scale, the seminiferous tubules in the TS group were visibly larger than those in the TN group (Fig. 3d). Additionally, in the TN group, only spermatogonia and primary spermatocytes were observed, with no mature sperm present. In contrast, the TS group exhibited spermatogonia, primary spermatocytes, secondary spermatocytes, spermatids, and mature sperm, representing various stages of spermatogenesis. HE-stained sections of musk glands and testicular tissues during the secretion and non-secretion periods (400X). (a) Musk gland section from the GN group; (b) Musk gland section from the GS group. GC: Glandular cells; EC: Epithelial cells; IC: Stromal cells. (c) Testicular section from the TN group; (d) Testicular section from the TS group. S: Spermatogonia; PS: Primary spermatocytes; SS: Secondary spermatocytes; Sp: Spermatids; Sz: Sperm; SC: Sertoli cells DEG analysis of muskrat scent glands and testes Transcriptome sequencing of the scent gland and testicular tissues yielded a total of 169,811 annotated unigenes, of which 51,966 (31.63%) were uniquely expressed in the scent glands, 55,816 (33.97%) were uniquely expressed in the testes, and 56,520 (34.40%) were shared between the two tissues (Table S2, Fig. S2). Seasonal comparisons identified 1,068 DEGs in scent glands (563 (52.72%) upregulated, 505 (47.28%) downregulated in GS vs. GN; Fig. S3) and 12,107 DEGs in testes (7,249 (59.87%) upregulated, 4,858 (40.13%) downregulated in TS vs. TN; Fig. S4), with 271 common DEGs (59 (21.77%) upregulated, 113 (41.70%) downregulated in both tissues during the scent-secreting period; Fig. S5). This disparity in DEG numbers highlights greater transcriptional dynamism in testes, potentially driving scent glands’ activity. GO enrichment analysis annotated 604 DEGs in scent glands to 126 terms (63 Biological Process, 28 Cellular Component, 35 Molecular Function), focusing on lipid metabolism, transmembrane transport, as well as tissue development and differentiation (Figs. 4a and 486 significant DEGs). Candidate DEGs Pla2g4e, Acacb, Apoa1, Krt19, Aldh3, and Msmo1 (Table 1) underscore patterns of enhanced lipid binding and metabolic processes during secretion, correlating with observed gland enlargement and vacuole formation. In testes, 5,227 DEGs enriched in 141 terms (Figs. 4b and 953 significant DEGs) emphasized reproduction, hormone synthesis and regulation, with candidate DEGs Pla2g4f, Itsn1, Hsd17b1, Srd5a, Zfp120, Plcb2, Cyp11a, and Prkca (Table 2) indicating upregulated hormone regulation during the scent-secreting period, aligning with active spermatogenesis. GO enrichment of DEGs in musk glands and testes between secreting and non-secreting periods. a GO enrichment of DEGs in musk glands between secreting and non-secreting periods. b GO enrichment of DEGs in testes between secreting and non-secreting periods Table 1GO terms related to scent secretion in muskrat scent glandsTermsNumberRepresentative genesLipid metabolic process31Pla2g4e, Fads2, Acacb, Gjb4, Sqle, Apoa1, Knstrn, Pla2g4f, Msmo1Small molecule metabolic process89Pla2g4e, Acacb, Anxa1, Apoa1, Knstrn, Krt19, Slc47a2, Elovl6Phosphatase activity17Ttn, Cdkn3, Nt5c1a, Myom1Lipid binding7Anxa1, Amz1, Pctp, Apoa1, Anxa8, Sbsn, BpifcTransmembrane transport57Aldh3, Cacna1s, Cacna1h, Krt19, Tmem40, Slc7a3, Slc46a2Nucleocytoplasmic transport12Myh4, Krt19, Krt17, Krt76Vesicle-mediated transport14Knstrn, Krt19Anatomical structure development31Gjb4, Itm2b, Foxd3, Tmem40, Col23a1Anatomical structure formation involved in morphogenesis3Foxd3, Gnat2, Itm2b GO terms related to scent secretion in muskrat scent glands Table 2GO terms related to scent secretion in muskrat testesTermsNumberRepresentative genesReproduction143Dhrs3, Krt79, Znf431, Srd5a, Cyp11a, Cmtm1Cellular protein modification process533Itsn1, Zfp120, Prkca, Ddr2, Bmpr1aLipid metabolic process230Hsd17b1, Zfp120, Pla2g4f, Plcb2, Srd5a, Fads1Lipid binding92Hsd17b1, Lire1, Gltp, Hip1Signal transduction847Itsn1, Zfp120, Pla2g4f, Plcb2, Srd5a, Bmp7, Trmt2bKinase activity340Itsn1, Itpr1, Prkca, Alox5DNA-binding transcription Factor activity237Zfp120, Runx3, Krt79, Dgcr2 GO terms related to scent secretion in muskrat testes KEGG pathway enrichment further revealed 16 significant pathways in glands (124 DEGs; Fig. S6), including steroid hormone biosynthesis, steroid biosynthesis, terpenoid backbone biosynthesis, linoleic acid metabolism, cortisol synthesis and secretion, biosynthesis of unsaturated fatty acids (UFAs), fatty acid biosynthesis, glycerolipid metabolism, aldosterone synthesis and secretion, cholesterol metabolism, GnRH signaling pathway, and estrogen signaling pathway (Table 3; candidate DEGs: Akr1b, Aldh9a1, Aldh3, Adh1, Lipa, Acacb, Fads2, Alox, Msmo1, Pla2g4, Calm, Cacna1h, Cacna1s, Cyp2j, and Krt19). In testes, 17 pathways (482 DEGs; Fig. S7) focused on steroid hormone biosynthesis, steroid biosynthesis, GnRH signaling pathway, estrogen signaling pathway, prolactin signaling pathway, cholesterol metabolism, renin-angiotensin system, aldosterone synthesis and secretion, and aldosterone-regulated sodium reabsorption (Table 4; candidate DEGs: Itpr1, Plcb2, Prkca, Cyp11a, Hsd17b1, Hsd11b2, Lipa, Srd5a, and Sult2b). Shared pathways suggest coordinated regulation, where testicular upregulation of hormone-related genes may influence gland lipid pathways, explaining the synchronized morphological changes. Table 3KEGG pathways related to scent secretion in muskrat scent glandsPathwaysNumberRepresentative genesSteroid hormone biosynthesis7Cyp2c70, Cyp2d10, Cyp2j, Hsd17b7, Cyp1A2, Cyp2d27, Cyp3a9, Cyp2d26, Aldh9a1Steroid biosynthesis4Msmo1, Sqle, Dhcr24, Hsd17b7Terpenoid backbone biosynthesis6Dhdds, Fdps, Dhdds, Mvd, IdI1, HmgcrLinoleic acid metabolism6Cyp2c70, Cyp2j, CYP1A2, Pla2g4, Cyp3a9, AloxCortisol synthesis and secretion6Cacna1s, Cacna1h, Adcy2, Cacna1c, Ldlr, Aldh9a1, Adh1Biosynthesis of unsaturated fatty acids2Elovl1, Fads2Glycerolipid metabolism6Tgfa, Aldh3, Mboat2, Akr1b, Aldh9a1, LipaFatty acid biosynthesis1 Acacb Aldosterone synthesis and secretion7Cacna1s, Cacna1h, Adcy2, Cacna1c, Ldlr, Calm, Adh1Cholesterol metabolism2Ldlr, Apoa1GnRH signaling pathway6Cacna1s, Adcy2, Cacna1h, Pla2g4, Calm, AloxEstrogen signaling pathway7Shc3, Adcy2, Krt23, Gabbr2, Krt19, Calm, Tgfa KEGG pathways related to scent secretion in muskrat scent glands Table 4KEGG pathways related to scent secretion in muskrat testesPathwaysNumberRepresentative genesSteroid hormone biosynthesis14Hsd17b1, Cyp11a, Srd5a, Sult2b, Hsd11b2, Cyp3a2, Cyp1b1Steroid biosynthesis8Msmo1, Nsdhl, Dhcr7, Lipa, Srd5a, Sult2b, Hsd17b1, Cyp11aGnRH signaling pathway34Itpr1, Plcb2, Prkca, Pla2g4f, Lipa, Cacna1sEstrogen signaling pathway30Itpr1, Plcb2, Krt19, Krt20, Shc3Prolactin signaling pathway26Prkca, Shc3, Kras, Mapk1Cholesterol metabolism12Srd5a, Sult2b, Mapk12, Mapk3, Mapk9, ApoeRenin-angiotensin system10Atp2a1, Apoae, LipaAldosterone-regulated sodium reabsorption11Mapk1, Mapk3, Hsd11b2, Nedd4l, Agtr2Aldosterone synthesis and secretion28Itpr1, Plcb2, Cyp11a, Hsd11b2, Krt19, Prkca, KEGG pathways related to scent secretion in muskrat testes As shown in Fig. 5, six candidate DEGs - Pla2g4, Cyp2j, Aldh3, Alox, Msmo1, and Ldlr - were selected from the key pathways identified in the GO and KEGG analyses. These were analyzed together with six randomly selected significantly DEGs from the transcriptome sequencing results - Aldph1a2, Gstm1, Slc7a3, Eln, Gstk1, and Atp2a1 - via RT-qPCR. The log2(FC) values for each gene from the RT-qPCR results were calculated and compared with those obtained from the transcriptome sequencing. The results showed that all six candidate genes were significantly differentially expressed between the scenting and non-scenting period (Fig. 5a), while the expression trends for all 12 genes were consistent between the RT-qPCR and transcriptome sequencing results (Fig. 5b). This indicates that the transcriptome sequencing results were highly reliable and the potential important roles of these candidate genes in the musk secretion and testicular activity of the muskrat. RT-qPCR verification of DEGs. (a) DEG expression levels; (b) Verification of DEGs. As shown in b, six candidate DEGs - Pla2g4, Cyp2j, Aldh3, Alox, Msmo1, and Ldlr were selected from the key pathways identified in the GO and KEGG analyses. These were subjected to RT-qPCR along with six significantly DEGs randomly selected from the transcriptome sequencing results: Aldph1a2, Gstm1, Slc7a3, Eln, Gstk1, and Atp2a1. The log2(Fold Change) values of each gene from the RT-qPCR were compared to the log2(Fold Change) values obtained from the transcriptome sequencing DM analysis of muskrat scent glands and testes A total of 387 metabolites were identified in the KEGG database (Fig. 6), with 23 linked to Cellular Processes, 51 to Environmental Information Processing, 13 to Genetic Information Processing, 507 to Metabolism, and 129 to Organismal Systems. In the HMDB database, 648 metabolites were annotated, including seven alkaloids and derivatives, 55 benzenoids, 174 lipids and lipid-like molecules, 52 nucleosides/nucleotides/analogues, 153 organic acids and derivatives, 16 organic nitrogen compounds, 63 organic oxygen compounds, 102 organoheterocyclic compounds, one organosulfur compound, and 24 phenylpropanoids and polyketides (Fig. S8). In the LIPID MAPS database, 224 metabolites were identified (Fig. 7), comprising 84 fatty acyls (FA), 89 glycerolipids (GL), 11 glycerophospholipids (GP), five polyketides (PK), three prenol lipids (PR), and 32 sterol lipids (ST). KEGG pathway annotation of metabolites LIPID MAPS annotation of metabolites PLS-DA analysis showed clear separation between S and N phases in both tissues (Fig. S9-10), validating distinct metabolic states without overfitting. In scent gland tissue, 180 DMs were identified under positive ion mode during the scent-secretion period compared with the non-secretion period (111 upregulated, 69 downregulated; Fig. S11), and 68 DMs under negative ion mode (33 upregulated, 35 downregulated). The top 30 upregulated and downregulated metabolites are presented in Tables S3 and S4, respectively. In testicular tissue, 139 DMs were identified under positive ion mode (19 upregulated, 120 downregulated; Fig. S12) and 55 DMs under negative ion mode (10 upregulated, 45 downregulated). The top 30 are shown in Tables S5 and S6, respectively. In positive ion mode, there were 27 common DMs between testicular tissues and musk gland tissues; in negative ion mode, there were 21 common DMs between testicular tissues and musk gland tissues(Fig. S13). Correlation analysis revealed strong positive correlations among lipids and lipid-like molecules in scent glands (including androstenone, dodecanedioic acid, all-trans-13,14-dihydroretinol, and pentadecanoic acid; Fig. S14). In testes, strong positive correlations were observed among steroids/steroid derivatives, fatty acyls, and flavonoids (including glycoursodeoxycholic acid, 10-undecenoic acid, lactobionic acid, sebacic acid, and diosmin), whereas phosphatidylcholine showed significant negative correlations with many metabolites (Fig. S15). DMs in scent glands from different periods were co-enriched in 67 pathways. In positive ion mode, 36 KEGG pathways were involved, with six significantly enriched (including steroid hormone biosynthesis, cortisol synthesis and secretion, primary bile acid biosynthesis, and cholesterol metabolism). In negative ion mode, 30 pathways were enriched, with two significantly enriched (including bile secretion). The top 20 pathways (sorted by ascending P-value) were shown in Fig. S16. Table 5 listed the five key KEGG pathways related to scent secretion and their associated significantly changed DMs. DMs in the testis were enriched in 86 pathways. In positive ion mode, 46 pathways were enriched, with three significantly enriched (including thyroid hormone synthesis and primary bile acid biosynthesis). In negative ion mode, 40 pathways were enriched, with 4 significantly enriched. The top 20 pathways were shown in Fig. S17. Table 6 presented the five key KEGG pathways related to testis activity and their corresponding significantly changed DMs. Table 5Metabolites enriched in scent-secretion pathwaysPathwaysMetabolitesSteroid hormone biosynthesis6α-Hydroxydehydroepiandrosterone; Androsterone; Cholesterol; Cortisol; Cortodoxone; Dehydroepiandrosterone; Estriol; Etiocholanolone; Progesterone;HydrocortisoneCortisol synthesis and secretionCholesterol; Cortisol; Cortodoxone; Progesterone;HydrocortisonePrimary bile acid biosynthesisCholesterol; Cholic acid; Glycocholic acid;Chenodeoxycholic AcidCholesterol metabolism acidsCholesterol; Glycocholic acidBiosynthesis of unsaturated fattyDocosapentaenoic acid; Nervonic acid;Adrenic acid; Arachidic Acid; Erucic acid Metabolites enriched in scent-secretion pathways Docosapentaenoic acid; Nervonic acid; Adrenic acid; Arachidic Acid; Erucic acid Table 6Metabolites enriched into testis activity-related pathwayPathwaysMetabolitesCholesterol metabolism Glycocholic acid Phenylalanine metabolismPhenylacetylglycineTryptophan metabolismIndole-3-acetic acid; Kynurenic acidEstrogen signaling pathway4-Aminobutyric acidPhenylalanine, tyrosine and tryptophan biosynthesisQuinic acid; Phosphoenolpyruvic acidPrimary bile acid biosynthesisCholic acid; Glycocholic acid; Chenodeoxycholic Acid Metabolites enriched into testis activity-related pathway Integrated analysis of metabolic pathways in muskrat scent glands and testes The association analysis of the transcriptome and metabolome of the scent gland in muskrats during both the scent-secretion and non-scent-secretion periods revealed 51 common metabolic pathways involving DEGs and DMs (Fig. S18a). Similarly, the association analysis of the transcriptome and metabolome of the testis during these periods indicated that the enriched pathways of DMs in the testis aligned with the enriched pathways of significantly DEGs (Fig. S18). This finding suggested that the identified DEGs and DMs demonstrated consistency, highlighting shared biological activities. Pathways were screened based on the significance of enrichment (P < 0.05) and sorted by ascending P-values to display the top 20 primary biochemical and signal transduction pathways in the scent gland, where DMs and DEGs were jointly involved (Fig. S19). Among these, 14 pathways showed significant enrichment, including steroid hormone biosynthesis, steroid biosynthesis, bile secretion, cortisol synthesis and secretion, aldosterone synthesis and secretion, primary bile acid biosynthesis, cholesterol metabolism, arachidonic acid metabolism, and oxytocin signaling. The KEGG pathways jointly enriched in the transcriptome and metabolome of testicular tissue were further screened based on the significance of enrichment (P < 0.05). Eleven pathways showed significant enrichment, including the FoxO signaling pathway, bile secretion, sphingolipid signaling pathway, cAMP signaling pathway, taste transduction, purine metabolism, thyroid hormone synthesis, and primary bile acid biosynthesis. Fig. S20 presented the top 20 primary biochemical and signal transduction pathways in which DMs and DEGs in testicular tissues were jointly involved. Discussion Significant enlargement of scent glands and testes, developed glandular cells, and full-stage active spermatogenic cells during the scent secretion period When comparing muskrat scent gland tissues between non-secretion and secretion periods, the GS group contained a greater abundance of interstitial and epithelial cells, which support the increased nutrient requirements and the need for timely transport and excretion of secretions at this stage [40], providing a foundation for glandular expansion during the secretion period [41]. Moreover, the gland cells in the GS group were larger but exhibited a lower nuclear density, with vacuole-like structures lacking nuclei that may contain significant amounts of secretions in the cytoplasm. The larger gland cells in the GS group facilitate the synthesis and transport functions of organelles like the endoplasmic reticulum and Golgi apparatus, while the reduced nuclear density prevents excessive competition among cells, thus promoting secretion efficiency [42]. The development and function of exocrine glands are regulated by various factors, including age, sex, neural regulation, and environmental influences, with hormones significantly impacting changes in exocrine glands. For instance, female mammary glands undergo significant proliferation and increased secretion during pregnancy and lactation [43]; androgens and estrogens regulate sebaceous gland activity [44]. The scent-secreting period of muskrats coincided with the breeding season, and musk secretion activity was regulated by sex hormones. Notably, a co-developmental relationship between musk glands and testicular tissues existed in muskrats [17]. Therefore, we analyzed testicular changes during secretion and non-secretion periods. The testis plays an essential role in reproductive and endocrine functions [45]. The testis in males often exhibits morphological differences across physiological states [46]. For example, the testes of the wood mouse (Apodemus sylvaticus) become active during the breeding season and regress during the non-breeding season due to the physiological cycles, as well as environmental factors such as photoperiod and temperature [47]. In the present study, the testes of muskrats during scent-secreting period were significantly larger and heavier than those during the non-secreting period. Tissue sections revealed that the seminiferous tubules in the scent-secreting group were significantly larger than those in the non-secreting group, containing various stages of spermatogenic cells, including mature sperm, indicating an active sexual state with increased sperm production and androgen secretion [48]. These findings demonstrated that testicular morphological changes align with those in scent glands. The physiological function of the testis is also closely regulated by hormones. Spermatogenesis is influenced by multiple hormones, such as gonadotropins and inhibin, while testosterone production is stimulated by gonadotropins, which regulate the secretion of reproductive hormones through a feedback mechanism [49]. Consequently, focusing on related hormones and key genes involved in the synthesis and action processes is essential for understanding the regulatory mechanisms underlying musk secretion in muskrats. Differential gene expression in scent glands and testes reveals key synthesis and hormone regulatory genes and pathways By comparing gene expression in the musk gland tissues between the scent-secreting and non-secreting periods, we identified 15 key genes associated with muskrat musk secretion. These included genes involved in fatty acid synthesis (Acacb [50] and Fads2 [51, 52]), arachidonic acid metabolism (Alox5, Alox15b, and Pla2g4 [53]), hormone signaling (Calm), cholesterol biosynthesis (Msmo1 [54]), calcium signaling (Cacna1h and Cacna1s [55]), cytochrome P450 pathways (Cyp2j [56]), sebaceous gland cells (Krt19 [57]), along with five genes (Akr1b, Aldh9a1, Aldh3, Adh1, and Lipa) previously linked to musk secretion in forest musk deer [58, 59]. These genes were significantly upregulated during the scent-secreting period, indicating their involvement in lipid and hormone-related pathways critical for musk production. In testicular tissues, seven key genes (Itpr1, Plcb2, Prkca, Cyp11a, Hsd17b1, Srd5a, and Sult2b) associated with steroid hormone biosynthesis [60, 61] and GnRH/prolactin signaling pathways [62] were significantly upregulated during the scent-secreting phase. This upregulation pointed to the activation of pathways promoting LHβ (Luteinizing Hormone Subunit Beta) transcription, indicating the activation of sex hormone synthesis and functional pathways at this stage, aligning with the reproductive activity needs of muskrats during this period. By examining the biological and metabolic pathways enriched in the scent gland and testis, we found that both are jointly enriched in pathways such as steroid biosynthesis, terpenoid backbone biosynthesis, metabolic pathways, and linoleic acid metabolism. These pathways may contain key genes involved in the synthesis and regulation of major components of musk, such as lipids, fatty acids, amino acids, and terpenoids. Through these commonly enriched pathways, we identified six candidate genes that may be associated with testis-regulated musk secretion in scent gland tissues: Itpr1, Plcb2, Prkca, Hsd11b2, Cyp11a, and Cyp2j. In the shared GnRH and prolactin signaling pathways, Itpr1, Plcb2, and Prkca were significantly upregulated in the testes during the scent-secreting period, promoting glycerophospholipid and phosphatidylethanolamine synthesis and metabolism. The upregulation of downstream genes Calm and Pla2g4 in the scent glands suggested that testes may regulate musk synthesis and secretion in the scent glands via the expression of Itpr1, Plcb2, and Prkca, influencing the arachidonic acid and tyrosine metabolic pathways. More importantly, the gene Cyp11a was significantly upregulated in the testis. The expression level of the hydroxysteroid 11β-dehydrogenase encoding gene Hsd11b2 was significantly downregulated, inhibiting the conversion of cortisol to cortisone. However, in the scent gland, the downstream genes Hsd17b3 and Cyp3a of Cyp11a and Hsd11b2 were significantly upregulated, promoting the synthesis of 16α-hydroxyestrone, 16-glucuronidase estriol, and 11β-hydroxytestosterone. Additionally, the gene Cyp2j, which encodes a member of the cytochrome P450 family, showed differential expression in both testis and scent gland tissues at different periods, but with opposite expression trends. During the musk-secreting phase, Cyp2j was significantly downregulated in the testis, while it was significantly upregulated in the scent gland. This gene promoted the metabolism of arachidonic acid to produce prostaglandin precursors while inhibiting the metabolic process of CYP11A converting cholesterol to pregnenolone. This suggested that the testis may regulate the expression of Hsd17b3 and Cyp3a in the scent gland through the genes Cyp11a and Hsd11b2, thereby influencing the cholesterol metabolism pathway in the scent gland. Cholesterol in the scent gland primarily participates in the metabolism of other poly unsaturated fatty acids rather than in the synthesis of pregnenolone. This further supported the hypothesis that, although scent gland cells in muskrats have the ability to synthesize androgens during the musk-secreting phase [19], the sex hormones regulating musk secretion activity are likely more dependent on synthesis by testicular cells rather than autocrine production in the scent gland. DMs enriched in steroid hormone, cholesterol, and fatty acid pathways that regulate scent secretion Comparing small molecule metabolites in the scent glands between the scent-secreting and non-secreting periods, 248 significant DMs were identified, including signature components of musk such as muscone [63]. The concentrations of muscone, along with primary components of musk such as ergosterol peroxide, heptadecanoic acid, heptadecen-16-yne-1,2,4-triol, dibutyl adipate, and cholesterol, were significantly upregulated in the scent glands during the scent-secreting period. Additionally, the concentrations of dehydroepiandrosterone, cortisol, and androstene, which are involved in steroid and steroid hormone synthesis and metabolism, were also significantly elevated in the scent glands during this period [64]. Analyzing the enriched pathways of these DMs can further elucidate their interconnections and explore the interaction network among them. In the scent glands, KEGG pathway enrichment showed that DMs were enriched in pathways related to steroid hormone biosynthesis, cortisol synthesis and secretion, primary bile acid biosynthesis, cholesterol metabolism, and UFA metabolism - biochemical pathways that play essential roles in musk secretion and were closely related to the synthesis of key musk components and hormonal effects [65]. Analysis of small molecule metabolites in the testes between the scent-secreting and non-secreting periods revealed 184 DMs in both positive and negative ion modes. The concentrations of 22(S)-hydroxycholesterol, epitestosterone, and D-α-hydroxyglutarate, which are closely related to steroid hormone metabolism, were significantly increased in the testes during the scent-secreting period. Additionally, lysophosphatidylethanolamine, linoleic acid, phosphatidylcholine, and lysophospholipids also showed significant increases during this period, suggesting that fatty acid synthesis plays a crucial role in the function of muskrat testes when scent-secreting [66]. Furthermore, the enriched pathways of DMs in the testes were primarily associated with phenylalanine, tyrosine, and tryptophan biosynthesis, the estrogen signaling pathway, primary bile acid biosynthesis, and cholesterol metabolism. These findings may provide insights into how the testes influence musk gland secretion function in muskrats. Transcriptomics and metabolomics revealed cholesterol and fatty acid metabolic pathways and testicular gene regulatory roles DMs in both scent gland and testicular tissues were significantly enriched in cholesterol metabolism, with a marked increase in cholesterol levels. Cholesterol, the most abundant sterol in animal tissues, is essential for cell function and metabolism [67]. Transcriptomic results indicated that Apoa1 expression was significantly downregulated, while Ldlr was upregulated in this pathway. Apoa1 encodes apolipoprotein, which facilitates cholesterol efflux from tissues as an auxiliary factor for lecithin-cholesterol acyltransferase (LCAT) [50]. Ldlr, which encodes the low-density lipoprotein receptor, binds to the primary cholesterol-carrying lipoprotein LDL in plasma and mediates cholesterol uptake via endocytosis [68]. These findings suggested that, in addition to synthesizing cholesterol through metabolic pathways, the scent gland requires ongoing cholesterol uptake to maintain its levels, highlighting the regulatory role of cholesterol synthesis in scent secretion. Future research could investigate major cholesterol synthesis sites, such as the liver and intestinal mucosa [69], to better understand its regulation in musk production. In this study, DMs and DEGs were jointly enriched in the UFA synthesis pathway, suggesting that UFAs play an important role in musk secretion and reproductive activities in muskrats. In this pathway, the Fads2 gene encodes fatty acid desaturase 2, which is involved in synthesizing highly UFAs by introducing cis double bonds for desaturation. Elovl1 encodes ELOVL fatty acid elongase 1, which catalyzes the first and rate-limiting reaction in the elongation cycle of long-chain fatty acids. Transcriptomic results showed that Fads2 was significantly downregulated, while Elovl1 was significantly upregulated during the musk-secreting phase. This finding aligned with the increased levels of cis-15-docosenoic acid and cis-13-docosenoic acid, along with the decreased levels of docosanoic acid in this pathway during the musk-secreting phase. These results suggested that muskrats tend to favor carbon chain extension and monounsaturated fatty acid synthesis during the musk-secreting phase, which has important implications for further decoding the musk secretion mechanism and musk composition in muskrats. Metabolomic results further confirmed the influence of the testes on musk secretion activities in muskrats. We found that key genes Itpr1, Plcb2, and Prkca in testicular tissue promoted the actions of genes Calm and Pla2g4 in the scent gland through GnRH and prolactin signaling pathways, regulating the synthesis and secretion of musk in muskrats via pathways such as arachidonic acid metabolism and tyrosine metabolism. Similarly, the arachidonic acid content in scent gland tissue increased, and the downstream product phosphatidylethanolamine in testicular tissue was also significantly upregulated, which can further regulate the transport and biosynthesis functions of fatty acids and cholesterol [70]. Additionally, we found that the testes regulate the expression of Hsd17b3 and Cyp3a in the scent gland through the genes Cyp11a and Hsd11b2. The gene Cyp2j, closely related to cytochrome P450 oxidation and fatty acid metabolism, was significantly enriched in the steroid hormone synthesis pathway. During the musk-secreting phase, Cyp2j was significantly downregulated in the testes but upregulated in the scent gland, a trend that facilitates the activation of cholesterol metabolism by the gene Cyp11a in the testes. The levels of the metabolites dehydroepiandrosterone sulfate and testosterone were significantly elevated in the testes, consistent with the upregulation of Cyp11a and Hsd17b and the downregulation of Hsd11b2 during the musk-secreting phase. In the scent gland during the musk-secreting phase, the progesterone level, a product of Cyp11a activity, and its downstream product 11-deoxycortisol were both lower than in the non-musk-secreting phase, while levels of cholesterol, dehydroepiandrosterone, cortisol, androstenedione, and estradiol were significantly increased, consistent with the function of key genes. These results indicated that the steroid hormone synthesis pathway plays an essential role in the testes-regulated musk secretion of the scent gland. The testes regulate the expression of Hsd17b3 and Cyp3a in the scent gland through the genes Cyp11a and Hsd11b2, promoting the synthesis of sterols and steroid compounds from cholesterol in the scent gland. This provides further metabolic support for the hypothesis that, while the synthesis pathways of sex hormones are activated in the scent gland [19], the key hormone testosterone, which regulates musk secretion, is more reliant on synthesis in the testes. Conclusions This study conducted a joint analysis of the transcriptome and metabolome of muskrat scent gland and testicular tissues for the first time, uncovering that the testis regulates the synthesis and transport of essential components of muskrat scent within the musk gland tissue through genes (Itpr1, Plcb2, Prkca, Hsd11b2, and Cyp11a) via the GnRH signaling pathway, prolactin signaling pathway, and steroid hormone synthesis pathway. These findings were further supported by metabolites closely linked to phenotypes, providing robust evidence at the gene transcription level for understanding the mechanisms behind musk secretion activity and aiding in elucidating the regulatory processes of musk secretion in muskrats. Future research will include gene sequence analysis and protein expression localization studies on the identified key genes, offering insights for deeper exploration into scent secretion mechanisms. Additionally, primary secretory cells of muskrat scent gland tissue will be isolated and cultured for stable cell passaging to further strengthen the roles of key genes and metabolites in scent gland development, primary scent fluid synthesis, and scent secretion regulation at the cellular and molecular levels. Supplementary Information Supplementary Material 1. Acknowledgements We would like to thank our colleagues Gang Luo, Ruiguang Gong, Naqash Goswami, Guohua Song, Hui Wang, Tongyan Zhu, Xiaozhao Tao, Jiapeng Li, and Xiaoxiao Liu for their valuable assistance during the experimental operations and data analysis. We thank the High Performance Computing Center of NWAFU for providing computing resources. Abbreviations Differentially expressed genes Androgen receptor Prolactin Reverse Transcription quantitative PCR Transforming Growth Factor Beta Differential metabolites Hematoxylin and eosin Principal component analysis Partial least squares-discriminant analysis Fold Change Variable Importance in the Projection Unsaturated fatty acids Gonadotropin-Releasing Hormone Authors’ contributions **Z.R** administered the project, and conceived and supervised the study. **B.S** developed the storyline and wrote the original manuscript. **T.H** performed all bioinformatic analyses. **C.L** contributed to sample collection and revised the manuscript. **W.J** uploaded the dataset. **S.W** contributed to language editing and revised the manuscript. Funding This study was supported by the Special Fund for Science and Technology Innovation Program of the Shaanxi Academy of Forestry (Grant No. SXLK2020-0207) and the Wildlife Conservation Project Fund of National Forestry and Grassland Administration (Grant No. 2020070212). Data availability Raw transcriptomic sequences have been deposited in the NCBI publicly-available repository - Sequence Read Archive (SRA) under BioProject PRJNA1284884 ( http://www.ncbi.nlm.nih.gov/bioproject/1284884 ). The metabolomic sequences have been deposited in the OMIX, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (https://ngdc.cncb.ac.cn/omix: accession no.OMIX010917). Declarations Ethics approval and consent to participate All animal-related protocols described in this study were conducted in accordance with the guidelines established by the China Council for Wild Animal Care. All animal experiment procedures (Permit No: 202205A29) were approved by the Northwest A&F University Animal Care and Use Committee. Consent for publication No applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Bing Song and Tingting Hong contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement Raw transcriptomic sequences have been deposited in the NCBI publicly-available repository - Sequence Read Archive (SRA) under BioProject PRJNA1284884 ( http://www.ncbi.nlm.nih.gov/bioproject/1284884 ). The metabolomic sequences have been deposited in the OMIX, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (https://ngdc.cncb.ac.cn/omix: accession no.OMIX010917).