Structural–Molecular Linkage Shapes the Adaptive Traits of Nuptial Pads in Male Rana chensinensis

ABSTRACT

Nuptial pads represent a key sexually dimorphic trait in anurans, with male Rana chensinensis exhibiting these structures on the Finger I of forelimbs. However, the specialized mechanisms underlying their functional adaptation remain poorly understood. This study aimed to characterize the structural and molecular features of these nuptial pads by using Finger Ⅲ as a control. Histological staining and transcriptome analysis were conducted to compare glandular morphology, collagen fiber characteristics, and gene expression profiles between the two fingers. Results revealed significantly larger specialized mucous glands (SMGs) and thicker collagen fibers in the nuptial pads, along with downregulation of KRT‐related genes and catalase (CAT), and upregulation of genes associated with small heat‐shock proteins (sHSPs), melanogenesis, and motor proteins. These findings suggest that nuptial pads achieve functional adaptation through synergistic structural (mechanical support via enlarged glands and thickened fibers) and molecular (metabolic regulation and secretory optimization) mechanisms. Such adaptations align with the demands of successful amplexus and ovulation stimulation, associated with collectively ensuring reproductive success in anurans.

Summary

This research focuses on the nuptial pads of Rana chensinensis, aiming to analyze the specialized mechanisms underlying their functional adaptation. Comparative analyses reveal that the nuptial pads exhibit enlarged mucous glands, increased collagen fiber thickness, and enhanced metabolic regulation and secretory activity at the transcriptional level. These adaptations directly support reproduction (amplexus, ovulation stimulation), filling a knowledge gap and providing a foundation for better understanding of the mechanisms underlying nuptial pad function in anuran amphibians.

1. Introduction

Nuptial pads, a typical secondary sexual characteristic in male anuran amphibians, serve dual functions during amplexus by providing mechanical grip and facilitating the transmission of chemical signals. Stronger grip brings the cloacal openings of males and females closer, allowing for direct release of semen on eggs, thereby increasing fertilization success [1]. Furthermore, it reduces interference from rival males, thus securing the male's reproductive opportunities [2]. The secretions act as attractant pheromones that are infiltrated directly into the female's circulatory system through the pectoral regions of skin attrition [3]. Additionally, some researchers link grip strength to high secretory activity, believing that these secretions may form glue‐like substances to reinforce the males grip on the females [4, 5]. These multifunctional roles make nuptial pads an ideal model for investigating the reproductive behaviors in anurans.

Research on nuptial pads focused on structural characteristics and functional mechanisms [6, 7]. Histological studies revealed that it has significant characteristics compared to normal skin, including thickening of the epidermis, distribution of multicellular exocrine specialized glands, and pigmentation [2, 8, 9]. Several studies further explored functional mechanisms, for example, immunohistochemistry and Western blot confirm highly expression of androgen/estrogen receptors in nuptial pads during amplexus [10]. The hormone treatment experiments identified androgens as key regulators of their development [11, 12, 13]. Most existing studies compared males and females, but these comparisons are often confounded by differences in sex hormone levels. As a result, there has been insufficient attention to the molecular distinctions between nuptial pads and adjacent non‐specialized structures when hormonal influences are controlled.

We hypothesize that nuptial pads exhibit structural and molecular adaptive specializations—distinct from the adjacent non‐nuptial portion on Finger III—that synergistically enhance mechanical grip strength and glandular secretory efficiency during amplexus. To verify this hypothesis and address limitations of previous cross‐sex comparative studies [14, 15], we conducted a rigorously controlled intra‐individual comparative analysis using male Rana chensinensis during amplexus, directly pairing nuptial pads with adjacent non‐nuptial portion on Finger III. This paired‐tissue design uses histologically homologous tissues from a single individual to control for consistency in genetic background, physiological status, and reproductive stage, with the presence or absence of nuptial pads as the sole variable [16], thereby eliminating inherent confounding factors in cross‐sex comparisons between male nuptial pads and female finger. By restricting comparisons to the same anatomical region within a single individual, our methodology enables more accurate attribution of observed differences to the specialization of nuptial pads, thereby offering a robust foundation for understanding their structural and functional significance.

In this study, we performed transcriptomic sequencing of the nuptial pads and Finger Ⅲ of R. chensinensis during the breeding season, enabling annotation of differentially expressed genes (DEGs) and functional enrichment analysis. Based on these data, we investigated the DEGs associated with keratin, melanogenesis, antioxidant activity, vascular tension, and motor proteins in both tissues. Furthermore, we compared structural characteristics of the nuptial pads and Finger Ⅲ via morphological, histological sections, and transmission electron microscopy (TEM) observations, including the thickness of collagen fibers and the type and size of multicellular exocrine glands. These findings revealed the multidimensional regulatory network of nuptial pads function beyond hormonal influences, providing valuable insights into the reproduction in anuran amphibians.

2. Materials and Methods

2.1. Experimental Animals

Nine pairs of adults of R. chensinensis (nine males and nine females) were randomly captured from a breeding site in the Qinling Mountain, Shaanxi Province, China (109°06′52″E, 34°00′56″N). They were captured during amplexus and each pair was held separately in an aquarium (50 cm × 20 cm × 20 cm) containing 5 cm of dechlorinated tap water. The dechlorinated water was obtained by aerating tap water for 3 days. All water temperature was maintained at room temperature (22 ± 1°C) under a 12L:12D photoperiod. We analyzed morphology of male frogs using the TesaCal Dura‐Cal Digital electronic calipers (1E‐2, Qingdao, China) and an analytical balance (1E‐4, METTLER‐TOLEDO, Zürich, Switzerland). Weight, snout‐vent length (SVL, distance from the tip of the snout to the posterior edge of the vent), hindlimb length (HLL, maximum length from to vent to the distal tip of toe Ⅳ), hindlimb thickness (HLT), lower arm and hand length (LAHL, from elbow to the tip of Finger Ⅲ), lower arm thickness (LAT) [17]. Male frogs develop nuptial pads on the Finger Ⅰ during the annual breeding season (Figure 1A). The surface of the nuptial pads was easily distinguishable from the rest of the skin by the presence of the dark (usually black) spherical bulges and spine‐shaped epidermal protrusions (EPs), and length and width of the nuptial pads were measured as shown in Figure 1B. All experimental procedures involving animals were approved by the Experimental Animal Care Regulations of Shaanxi Normal University and the measures of Shaanxi Province for the administration of affairs concerning experimental animals (approval number 202612056).

Figure 1.

Morphological maps of nuptial pads of Rana chensinensis during the breeding season. (A) The entire right limb of the male frogs. Ⅰ、Ⅱ、Ⅲ, and Ⅳ indicates the order of manual digit. The break arrow indicates the nuptial pads. Finger numbers follow Luna et al [2]. (B) The enlarged portion of the nuptial pads, illustrating the measurement of its length (L) and width (W). The total length of the nuptial pad was calculated as the sum of the lengths of the two separate protrusions (L = L1 + L2). The width was measured as the sum of the maximum transverse diameters of each protrusion (W = W1 + W2).

2.2. Experimental Design

Given that the skin of the nuptial pads shares a similar epidermal structure with Finger III and was easier to obtain in sufficient quantities, it was selected as the control tissue for the nuptial pads. The experiment was comprised of the following four parts: (i) morphological observations, (ii) microstructural observations, (iii) ultrastructural observations, and (iv) transcriptome analysis. After measuring morphological parameters, the nuptial pads (Tissue 1) and Finger III (Tissue 2) were excised from the subjects. The three tissue samples were immediately placed into sterile tubes, flash‐frozen in liquid nitrogen, and stored at −80°C for subsequent transcriptomic analysis. Additional samples were fixed in Bouin's solution for 1 day and then transferred to 70% ethanol for histological sectioning and microstructural observations. For TEM ultrastructural observations, the tissues were fixed in 3% glutaraldehyde.

2.3. Histological Analysis and Microstructural Observations

We gradually transferred the samples immersed in 70% ethanol into a graded alcohol series (80%, 90%, 95%, and twice in 100%), treating them for 1 h at each time to ensure complete dehydration, followed by xylene treatment to achieve tissue transparency. Subsequently, the samples were immersed in a wax solution maintained at 60°C for 10 h and then embedded in paraffin. Sections of 5 μm thickness were stained using both Masson's trichrome and hematoxylin‐eosin (H&E) staining methods. After staining, the stained slides were scanned using a fully automated panoramic digital slide scanning system (Axioscan 7) and the acquired images were analyzed using ZEISS ZEN 3.11 software (Windows version) to analyze histological changes.

2.4. TEM and Ultrastructural Observations

For electron microscopy, samples initially fixed in 3% glutaraldehyde were post‐fixed in 1% osmium tetroxide. The specimens were then dehydrated using a graded acetone series (30%, 50%, 70%, 80%, 90%, 95%, and 100%, with three times in 100% acetone), embedded in Epon‐812 resin, and stained with uranyl acetate and lead citrate. Ultrathin sections were examined using a transmission electron microscope (TEM; JEM‐1400FLASH, JEOL, Japan).

2.5. RNA Extraction, cDNA Synthesis and cDNA Library Construction

According to the manufacturer's instructions, total RNA was extracted using Trizol Reagent (Invitrogen, GenStar). The RNA concentration and purity were assessed by measuring the absorbance ratios at 260/230 and 260 nm/280 nm using a NanoPhotometer spectrophotometer. RNA integrity and the presence of genomic DNA contamination were evaluated through agarose gel electrophoresis. Subsequently, RNA was reverse transcribed into cDNA using Illumina TruSeq RNA Sample Prep Kit (Illumina) method. The cDNA library construction procedures were as follows: (1) mRNA containing poly‐A tails was enriched using oligo(dT)‐conjugated magnetic beads. (2) The enriched mRNA was fragmented into short fragments via ultrasound. (3) Using the fragmented mRNA as a template and random oligonucleotides as primers, first‐strand cDNA was synthesized in an M‐MuLV reverse transcriptase system. (4) The RNA strand was degraded using RNase H, and the second‐strand cDNA was synthesized in a DNA polymerase I system with dNTPs as substrates. (5) The resulting double‐stranded cDNA underwent end repair, a single “A” nucleotide was added to the 3′ ends to facilitate adapter ligation, and fragments of appropriate size (~200 bp) were selected using AMPure XP beads. (6) The cDNA library was amplified via PCR. (7) The PCR products were purified again using AMPure XP beads. (8) Library quality was assessed using a Qubit 2.0 Fluorometer (for concentration) and an Agilent 2100 Bioanalyzer (for integrity).

2.6. Transcriptome Sequencing, De Novo Assembly, and Functional Annotation

Libraries were sequenced on the Illumina next‐generation sequencing (NGS) platform with high coverage [18]. Billions of raw image data were converted into sequence data through Base Calling, resulting in files in FASTQ format. Following sequencing, the Trimmomatic software (http://www.usadellab.org/cms/index.php?page = Trimmomatic) was employed for quality trimming, during which adapter sequences, reads with high N content, excessively short reads (< 75 bp), and low‐quality reads were removed to generate clean data. Subsequently, the clean reads were subjected to de novo assembly using the Trinity platform (http://trinityrnaseq.sourceforge.net/), followed by read realignment to refine the assembled unigenes. Functional annotation of unigenes was performed conducted by performing BLAST (Basic Local Alignment Search Tool) searches against five public databases: NR, KEGG, GO, Swiss‐Prot, and eggNOG. BLAST is a widely recognized algorithm for identifying sequence similarities [19].

2.7. Differential Gene Expression Analysis

The expression levels of the transcripts were quantified using Reads Per Kilobase of transcript per Million mapped reads (RPKM). Differential expression of genes was performed using the edgeR software package, with differentially expressed unigenes (DEGs) identified by a false discovery rate (FDR) ≤ 0.05 and |log2(fold change)| ≥ 1. These DEGs were subsequently subjected to functional annotation and enrichment analysis of Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways.

2.8. Statistical Analysis

Histological data were analyzed using IBM SPSS Statistics 27 software, and the results were presented as means ± standard deviation (SD). Prior to parametric testing, Levene's test was conducted to assess homogeneity of variances. For variables satisfying the assumption of equal variances (Levene's test p > 0.05; e.g., epidermal thickness), the independent‐samples Student's t‐test was applied. For variables violating this assumption (Levene's test p < 0.05; e.g., cross‐sectional long and short diameters of multicellular exocrine glands), Welch's t‐test was used as a robust alternative. All tests were two‐sided, with statistical significance defined as p < 0.05 and high statistical significant results as p < 0.001. Data visualization was carried out using OriginPro 2022 (OriginLab Corporation, Northampton, MA, USA) and Adobe Photoshop 2021 (Adobe Inc., San Jose, CA, USA).

3. Results

3.1. Morphological Observations and Measurements of the Nuptial Pads in R. chensinensis

In this study, male frogs (n = 7) with an average weight of 7.12 ± 1.81 g, SVL of 41.96 ± 1.29 mm, HLL of 75.47 ± 5.24 mm, HLT of 5.11 ± 1.47 mm, LAHL of 21.09 ± 2.09 mm, and LAT of 4.94 ± 0.64 mm were selected. The mean total length and width of the nuptial pads were 6.31 ± 1.32 and 2.18 ± 0.34 mm, respectively.

3.2. Histological and TEM Observations and Measurements

Masson's trichrome staining revealed significant disparities between the nuptial pads and Finger Ⅲ. In the nuptial pads, EPs were prominently observed (Figure 2A1), becoming gradually less conspicuous or even depressed towards adjacent skin (Figure 2A2). Conversely, the epidermis of Finger Ⅲ entirely lacked such EPs (Figure 2B1). The average epidermal thickness of the nuptial pads was 144.81 ± 30.05 μm, significantly greater than the 43.49 ± 22.95 μm measured in the epidermis of Finger Ⅲ (p < 0.001). Regarding dermal characteristics, both the nuptial pads and Finger Ⅲ exhibited a dense network of collagen fibers that stained blue. However, the multicellular exocrine glands surrounded by collagen fibers differed markedly. Ordinary mucous glands (OMGs) were present in the adjacent skin of nuptial pads or Finger Ⅲ (Figure 2A2,B2), while specialized mucous glands (SMGs) were present only in the nuptial pad area. The alveolar, ovoid‐shaped SMGs exhibited relatively large cross‐sectional dimensions, with long and short diameters averaging 228.59 ± 79.70 and 147.04 ± 39.45 μm, respectively, and their ducts were observed (Figure 2A1). In contrast, the OMGs were considerably smaller (p < 0.001), with cross‐sectional long and short diameters averaging 72.02 ± 18.94 and 30.76 ± 10.01 μm, respectively. Additionally, both the nuptial pads and Finger Ⅲ were surrounded by a high density of melanophores (Figure 2A1,B2).

Figure 2.

Histological sections of Rana chensinensis at Fingers Ⅰ and Ⅲ were stained with Masson's trichrome stain method (n = 3). (A) Cross‐sections of Finger Ⅰ showed nuptial pads. Scale bar = 300 μm. Dashed line separates nuptial pad from adjacent skin. (A1, A2) Enlarged views of squared areas show details of nuptial pads and adjacent skin. Note the increase in thickness of the dermis of the nuptial pad compared with the adjacent skin; dashed line separates nuptial pad from adjacent skin. (B) Cross‐sections of the finger Ⅲ. Scale bar = 150 μm. (B1, B2) Enlarged views of squared areas show details of Finger Ⅲ. de = dermis, du = duct, ep = epidermis, EPs = epidermal protrusions, m = melanophores, n = nucleus, OMG = ordinary mucous gland, SMG = specialized mucous glands.

Histological hematoxylin and eosin (H&E) staining revealed the morphological characteristics of the secretory portions of SMGs and OMGs, as well as their respective nuclear distribution patterns. The secretory portion of the SMG consists of a single‐layered, high columnar, radially‐arranged mucocytes, each with a round, basal nucleus (Figure 3A2) and a few nuclei distributed near the lumen (Figure 3A1). Conversely, the secretory portion of OMG consists of a single‐layered, cubical to cylindrical mucocytes, each with a rounded nucleus located at the basal pole of mucocytes and the duct of OMGs resemble those described for SMGs (Figure 3B2). Although melanophores were distinctly observable in both the nuptial pads and Finger Ⅲ, their distribution patterns differed significantly. In the nuptial pads, melanophores were identified not only in the epidermis and the collagen fibers of dermis but also within the SMGs (Figure 3A2). In Finger Ⅲ, however, melanophores were restricted to the collagen fibers of dermis and absent from the epidermis (Figure 3B1). Additionally, blood vessels immersed in collagen fibers were clearly visualized in both the nuptial pads and finger Ⅲ (Figure 3A1,B1).

Figure 3.

Histological sections of Rana chensinensis at the Fingers Ⅰ and Ⅲ were stained with hematoxylin and eosin stain method (n = 3). (A) Cross‐sections of Finger Ⅰ showed nuptial pads and adjacent skin. Scale bar = 300μm. (A1–A3) Enlarged views of squared areas show details of nuptial pads. (B) Cross‐sections of Finger Ⅲ. Scale bar = 150 μm. (B1, B2) Enlarged views of squared areas show details of Finger Ⅲ. bv = blood vessel, de = dermis, du = duct, ep = epidermis, EPs = epidermal protrusions, m = melanophores, n = nucleus, OMG = ordinary mucous gland, SMG = specialized mucous glands.

TEM demonstrated that there were no significant morphological differences in epidermis between nuptial pads and Finger Ⅲ (Figure 4A–D). Notably, the junction between the epidermis and dermis in the nuptial pads exhibited thick collagen fibers, whereas the corresponding regions in Finger Ⅲ showed a higher density of fibroblasts and less developed collagen fibers (Figure 4C,D). The apices of contiguous mucocytes in the SMG towards the lumen displayed recognizable tufts of microvilli, and some mucocytes were enriched with large numbers of mitochondria. Also, numerous granules were observed in the SMG, most of which were the electron‐opaque vesicles (Figure 4F). In turn, while the OMG also exhibited similar short microvilli, its granules were generally larger and less electron‐opaque compared to those in the SMG (Figure 4E,F).

Figure 4.

Transmission electron micrographs of nuptial pads and Finger Ⅲ (n = 3). (A, B) Cross‐sections of stratum corneum. Scale bar = 10 μm. (C, D) Cross‐sections of the junction between the epidermis and dermis. Scale bar = 10 μm. (C1–D1) Enlarged views of panels (C) and (D) show details of desmosomes between epithelial cells. Scale bar = 2 μm. (E, F) Cross‐sections of the secretory portion and lumen of the glands. Scale bar = 5 μm. fb = nucleus of fibroblast, mv = microvilli, L = lumen, mc = mitochondria.

3.3. Expression Profile of Keratin Family Genes

We used transcriptomic analysis to identify the average expression levels of the top 8 keratin family genes in the nuptial pads and Finger Ⅲ of R. chensinensis, quantified by reads per kilobase of exon model per million mapped reads (RPKM) (Table 1). Notably, these keratin genes, which belong to Type I keratins or Type II keratins, exhibited either high (> 50 RPKM ≤ 1000) or extremely high (> 1000) expression level in both the nuptial pads and Finger Ⅲ (Table 1). Subsequently, a significant difference analysis was conducted on these eight keratin family genes (Figure 5). Compared to Finger Ⅲ, the expression levels of KRT3‐1, KRT, KRT19, and KRT10‐1 were significantly downregulated in the nuptial pads (FDR < 0.001, |log2FC| > 1, Figure 5B,D–F), while the expression levels of KRT17, KRT13‐1, KRT45, and KRT4 showed no significant differences (FDR > 0.05 or |log2FC| < 1, Figure 5A,C,G,H). Overall, the expression levels of keratin family genes in the nuptial pads were significantly lower than those in Finger III.

Table 1.

RPKM of the corresponding sequence of DEGs associated with keratin between nuptial pads and Finger Ⅲ of Rana chensinensis.

ID	Gene	Product	RPKM		FDR	Log2FC
			Finger Ⅲ	Nuptial pads
TRINITY_DN48328_c0_g1	KRT17	Keratin, Type I cytoskeletal 17‐like	1896 ± 634	2446 ± 414	0.45	0.37
TRINITY_DN79904_c0_g1	KRT3‐1	Keratin‐3, Type I cytoskeletal 51 kDa‐like isoform X1	6277 ± 2978	1881 ± 1689	< 0.001	−1.7
TRINITY_DN52455_c1_g1	KRT13‐1	Keratin, Type I cytoskeletal 13‐like isoform X1	750 ± 248	962 ± 201	0.47	0.36
TRINITY_DN43615_c0_g1	KRT	Keratin, Type I cytoskeletal 47 kDa‐like	605 ± 219	228 ± 129	< 0.001	−1.4
TRINITY_DN65882_c1_g1	KRT19	Keratin, Type I cytoskeletal 19‐like	588 ± 400	165 ± 98	< 0.001	−1.8
TRINITY_DN52455_c0_g1	KRT10‐1	Keratin, Type I cytoskeletal 10‐like isoform X1	668 ± 1028	70.70 ± 67	< 0.001	−3.2
TRINITY_DN65439_c0_g1	KRT5	Keratin, Type II cytoskeletal 5‐like	1138 ± 648	917 ± 387	0.39	−0.31
TRINITY_DN60149_c9_g1	KRT4	Keratin, Type II cytoskeletal 4‐like	1151 ± 408	598 ± 267	0.002	−0.95

Figure 5.

The top eight average expression of genes associated with keratin between the nuptial pads and Finger Ⅲ in R. chensinensis (n = 3). (A–F) The expression levels of Type I keratin family genes. (G, H) The expression levels of Type II keratin family genes. RPKM =Reads Per Kilobase of exon model per Million mapped FDR reads. Data are presented as mean ± standard deviation. ***FDR < 0.001, |log2FC| > 1.

3.4. Melanin Pigment

Focusing on the KEGG pathway annotation results for melanogenesis (Ko04916), genes associated with the RAS–MAPK and WNT‐β–catenin signaling axes were significantly enriched in R. chensinensis (Figure 6). DEGs of regulating melanogenesis are shown in Table 2. Notably, the expression levels of ERK, GSK3β, and DCT in the nuptial pads were significantly upregulated compared to Finger Ⅲ (FDR < 0.01, |log2FC|> 1). It is worth noting that the expression level of ERK in the nuptial pads reached 12.5 RPKM, whereas no expression was detected in Finger III. Conversely, the expression levels of FZD and TYRP1 in the nuptial pads were significantly downregulated compared to Finger Ⅲ (FDR < 0.05, |log2FC|> 1). However, the expression levels of FZD and TYRP1 was extremely low (< 5 RPKM), lower than that of ERK.

Figure 6.

Mechanism diagram of melanogenesis in Rana chensinensis. Melanin pigments are synthesized and stored in melanosomes before it is transferred to the surrounding keratinocytes. The canonical melanogenesis pathway involves MITF driven expression of TYR, TYRP1, and DCT, which catalyze melanin formation. The activity of MITF is known to be regulated by the MAPK and WNT β catenin signaling pathways in vertebrates. Red arrows indicate significant trends in genes (FDR ≤ 0.05, |log2FC | > 1). Data are from three biological replicates (n = 3).

Table 2.

DEGs of melanogenesis (Ko04916) signaling pathway between nuptial pads and Finger Ⅲ of Rana chensinensis.

			RPKM
KO_ID	Gene	Product	Finger Ⅲ	Nuptial pads	FDR	Log2FC
K04371	ERK	Mitogen‐activated protein kinase 1	0	12.5 ± 5.2	< 0.001	11.0
K02842	FZD10	Frizzled‐10	4.4 ± 1.9	1.5 ± 0.1	< 0.001	−1.55
K03083	GSK3β	Glycogen synthase kinase‐3 beta isoform X1	0.2 ± 0.1	0.8 ± 0.1	< 0.001	2.3
K00506	TYRP1	5,6‐Dihydroxy indole‐2‐carboxylic acid oxidase‐like	0.6 ± 0.3	0.3 ± 0.1	0.01	−1.1
K01827	DCT	l‐Dopachrome tautomerase	1.3 ± 1.0	3.0 ± 1.1	0.002	1.2

3.5. Resistance to Oxidative Stress

The heatmap results revealed that the expression levels of the majority of heat‐shock proteins (HSPs) in nuptial pads were higher than those in Finger III (Figure 7A). A subsequent statistical analysis was conducted on DEGs related to oxidative stress. The expression levels of HSPB6, HSPB8, and HSPB11 were significantly upregulated in nuptial pads compared to Finger III (FDR < 0.05, |log2FC| > 1). (Figure 7B–D). Among these, HSPB6 showed the top most upregulation, with a 1.91‐fold change increase in expression levels in nuptial pads relative to Finger Ⅲ (Figure 7B). Multiple antioxidant enzymes were detected in both the nuptial pads and Finger III, including glutathione peroxidase 1 (GPX1), glutathione peroxidase 3 (GPX3), superoxide dismutase (SOD), extracellular SOD (EC‐SOD), and catalase (CAT). Notably, the expression level of SOD in nuptial pads was remarkably high, reaching 197.08 RPKM. However, no significant difference in the expression of SOD was observed between nuptial pads and Finger III (FDR > 0.05, |log2FC| < 1), where the RPKM value was 207.10. In contrast, although the expression of CAT was lower in nuptial pads than in Finger III, its expression level was 89.99‐fold change lower than that of SOD (Figure 7E).

Figure 7.

The difference of genes related to oxidative stress between nuptial pads and Finger III. (A) Heat map of genes related to oxidative stress. The color intensity reflects row‐wise z‐score‐normalized RPKM expression values. (B–E) Differently expression genes related to oxidative stress. Data are presented as mean ± standard deviation (n = 3). *FDR ≤ 0.05; **FDR ≤ 0.01; ***FDR ≤ 0.001. Data are from three biological replicates (n = 3).

3.6. Functional Enrichment of DEGs

To better study the biochemical metabolic pathways of the DEGs in the nuptial pads and Finger Ⅲ, we performed KEGG enrichment analysis (Figure 8). The top 20 important KEGG‐enriched pathways were grouped into four categories: metabolism (M), genetic information processing (GIP), cellular processes (CP), and organizational systems (OS). KEGG results showed that metabolism‐related pathways accounted for three‐quarters of the top 20. Of these 15, more than half are related to glucose metabolism, such as amino sugar and nucleotide sugar metabolism, glycolysis, and glycosaminoglycan degradation.

Figure 8.

KEGG analysis of the top 20 differentially expressed genes (DEGs). Dot plots of DEGs between nuptial pads and Finger Ⅲ. The x‐axis shows the gene ratio of each term, and y‐axis shows the KEGG annotation terms. The colors of dots represent the enrichment significance and dot sizes represent gene counts. CP = cellular processes, GIP = genetic information processing, KEGG = Kyoto Encyclopedia of Genes and Genomes, M = metabolism, OS = organismal systems.

3.7. The Expression of Genes Related to Motor Proteins

The structure of the skeletal muscle filament was illustrated in Figure 9A, which highlights two key components: the thick filament (myosin filament, colored purple) and the thin filament (actin filament, with a chain ‐ like morphology). In the magnified view of the thin filament, it comprises not only actin but also regulatory proteins such as tropomyosin (TPM) and troponin. Tropomyosin binds end‐to‐end along the actin filament, while troponin consist of three subunits: Troponin T (TNNT), Troponin C (TNNC), and Troponin I (TNNI). Additionally, calcium was depicted as a small green sphere associated with the troponin complex, emphasizing their critical role in the muscle contraction mechanism.

Figure 9.

Comprehensive analysis of muscle contraction‐related molecular mechanisms. (A) Schematic diagram of the interaction between actin‐containing thin filaments and myosin‐containing thick filaments during muscle contraction. (B) Sankey diagram depicting functional classification and expression level of genes associated with muscle contraction in nuptial pads. The color corresponds to each functional category, while the width represents gene expression levels, a wider band indicates higher expression. (C) Heatmap showing differential expression motor protein‐related genes in Finger Ⅲ and nuptial pads. The color intensity reflects row‐wise z‐score‐normalized RPKM expression values. **FDR ≤ 0.01; ***FDR ≤ 0.001. Data are from three biological replicates (n = 3).

We analyzed the expression of 18 genes related to skeletal muscle contraction in nuptial pads, and all these genes showed high expression levels, ranging from 41 to 830 RPKM (Figure 9B). The cluster heatmap indicated that samples within each treatment group clustered separately, suggesting consistent gene expression patterns within groups (Figure 9C). Notably, the expression levels of several troponin subunits, TNNC2, TNNI2, TNNT3, TNNC1, and TNNI1, were significantly upregulated in nuptial pads compared to Finger III (FDR ≤ 0.01, |log2FC|> 1). Among these, TNNI2 exhibited the most significant upregulation, showing a 2.82‐fold increase in expression in nuptial pads compared to Finger III (Figure 9B). Regarding the expression of actin, ACTA1, ACTC2, and ACTC1 were significantly upregulated in nuptial pads compared to Finger Ⅲ (FDR ≤ 0.001, |log2FC| > 1). Analysis of myosin revealed that the expression levels of MYL1, MYL10, and MYH6_7 were significantly upregulated in nuptial pads than Finger Ⅲ (FDR ≤ 0.001, |log2FC|> 1). Particularly, MYL1 showed a threefold increase in expression in nuptial pads (829.97 RPKM) compared to Finger III (276.37 RPKM) (Figure 9B,C).

3.8. Vasodilation

Esterified arachidonic acid (AA) on the inner surface of the cell membrane is hydrolyzed into its free form by phospholipase A2 (PLA2) (Figure 10). Free AA is subsequently metabolized by cyclooxygenase (COX) and cytochrome P450 (CYP) enzyme systems to generate a spectrum of bioactive mediators with vasodilatory effects, including prostaglandins (PGs), epoxyeicosatrienoic acids (EETs), 16(R)‐hydroxyeicosatetraenoic acid (16(R)‐HETE), and so forth. Transcriptome analysis revealed that the expression levels of genes encoding key enzymes involved in these metabolic pathways were significantly downregulated in the nuptial pads compared to Finger Ⅲ. Specifically, the expression level of cytosolic PLA2 (cPLA2) in nuptial pads was 5.01‐fold lower than those in finger III. In the COX pathway, the expression levels of COX2, thromboxane A synthase (TXAS), aldo‐keto reductase 1 (AKR1), and 15‐hydroxyprostaglandin dehydrogenase (HPGD) were downregulated by 17.51‐, 3.16‐, 2.85‐, and 2.64‐fold, respectively, in nuptial pads compared to Finger III. Regarding the CYP pathway, the six highly expressed members of the CYP2 family were identified in nuptial pads, among which CYP2C8, CYP2C15, CYP2G1, CYP2H2, and CYP2K1 exhibited significantly reduced expression levels in nuptial pads compared to Finger Ⅲ (FDR ≤ 0.001, |log2FC|> 1).

Figure 10.

Mechanism diagram of arachidonic acid metabolism in Rana chensinensis. Endogenous arachidonic acid is mainly derived from cell membrane phospholipids, which are processed by phospholipase A2 (PLA2). Free arachidonic acid serves as a precursor for a plethora of metabolites, including PGs, 16 (R) H–ETE and EETs. Each enzyme gene was represented by a rectangle. The color scale represents log2(fold change, FC) values on a linear scale. The values themselves are log₂‐transformed, but the color gradient is linear, not logarithmic. Red indicates positive log2FC (relatively higher expression in nuptial pad), blue indicates negative log2FC (relatively lower expression). prostaglandins F2α (PGF2α); prostaglandins E2 (PGE2); epoxyeicosatrienoic acids (EETs); 16(R)‐hydroxyeicosatetraenoic acid (16 (R) HETE). n = 3 biological replicates per group, total six samples. ***FDR ≤ 0.001, |log2FC|> 1. Data are from three biological replicates (n = 3).

4. Discussion

Nuptial pads are secondary sexual characters exclusive to anuran amphibians, typically distributed on the ventral surface of males [4]. Previous studies identified that the distribution of nuptial pads can be served as a criterion for diagnosing species [15, 17, 20, 21]. Our morphological observations revealed that the nuptial pads of R. chensinensis occur only on Finger Ⅰ, exhibit a dark (usually black) coloration, are characterized by 2–3 spherical bulges, and contain spine‐shaped EPs. Notably, similar nuptial pads characteristics have been reported in R. minuscula [17], Thoropa petropolitana [21], Gracixalus guangdongensis [22], and Phyllomedusa azurea [2]. However, Gracixalus gracilipes and Gracixalus jinggangensis develop nuptial pads on both Fingers I and II [22], while Rana dybowskii and Thoropa miliaris exhibit these structures on Fingers I, II, and III [10, 21]. Moreover, Gracixalus seesom lacks nuptial pads entirely [22]. Altogether, the distribution of nuptial pads cannot be taken as a criterion for diagnosing species, as it varies significantly among all anuran species.

Glands associated with reproductive function are known as sexually dimorphic skin glands (SDSGs), and the term “breeding gland” has been widely used to describe SDSGs located within nuptial pads [4, 23, 24]. Breeding glands usually consist of SMG or specialized serous gland (SSG), both of which are linked to hypertrophic integument [25]. The classification of mucous glands into OMGs and SMGs follows standard terminology for anuran skin glands. OMGs are small, widely distributed glands involved in general skin lubrication, while SMGs are larger, breeding‐related glands restricted to sexually dimorphic structures such as nuptial pads [24]. Our study revealed that the nuptial pads contained only one type of breeding glands, the SMG. In contrast, Finger III had no breeding glands, just a small amount of OMG. This finding is consistent with a substantial amount of research suggesting that SMGs are the predominant breeding glands type within nuptial pads [2, 9, 10, 26]. Furthermore, both the long and short diameters of the SMG were more than three times larger than those of OMG. The significantly larger size of the SMG is the primary reasons why finger I in the R. chensinensis exhibits several nuptial pads visible to the naked eye.

Ultrastructural observations revealed a significant increase in collagen fiber deposition at the epidermis‐dermis junction of the nuptial pads. In addition, the presence of tufts of microvilli, numerous mitochondria and electron‐opaque vesicles in the SMG supports its high secretion activity [26, 27]. Given that previous research has associated nuptial pads with enhanced grip and secretion‐related functions [4], and thicker collagen fibers are known to confer rigidity and tensile strength on tissue [28, 29, 30], our observations of thicker collagen fibers and active SMG secretion are consistent with these proposed roles. It is therefore reasonable to hypothesize that these structural and cellular features may contribute to improved grip strength during amplexus and glandular communication functions.

Nuptial pads are specialized keratinized skin tissues that are closely associated with the expression of keratin genes [31]. The Type I–Type II keratin pairs form heterodimers, which subsequently assemble into intermediate filaments (IFs) [30, 32]. IFs mediate cell–cell adhesion by binding to desmosomes between vertebrate epithelial cells [33]. A previous study on R. chensinensis showed that the expression of keratin genes was significantly upregulated in nuptial pads than in hind toe skin, and it was hypothesized that high expression levels of keratins drive keratinization of the nuptial pad epithelium [15]. Conversely, our transcriptomic data found that the expression levels of KRT3‐1, KRT, KRT19, KRT10‐1, and KRT4 were significantly downregulated in nuptial pads. Based on the above results, it is speculated that the lower expression levels of keratin genes may weaken the stability of desmosomes, potentially increasing intercellular spaces and loosening intercellular connections. This potentially allows the secretions from SMGs to reach the skin surface more rapidly. It is important to emphasize that these functional implications are currently speculative and await experimental validation. To rigorously define the role of keratins in nuptial pad biology, complementary lines of evidence—including subcellular protein localization, high‐resolution ultrastructural analysis of desmosomal architecture, and quantitative histological assessment of intercellular space morphology—are critically required.

During the breeding season, the nuptial pads of R. chensinensis typically appear darker and more pigmented compared to the adjacent skin. KEGG pathway annotation for melanogenesis (Ko04916) indicated that the expression level of ERK, GSK3β, and DCT was significantly upregulated in the nuptial pads. Previous research has demonstrated that melanophores are essential for establishing the pigmentation pattern of nuptial pads [4]. In Plectropomus leopardus, the phosphorylation of ERK positively regulates melanogenesis by activating MITF expression, enhancing specific melanogenic enzymes activity, thereby modulating the melanogenesis process [34, 35]. This mechanism is conserved in many species, including amphibians, fish, and mammals [36, 37]. Therefore, we hypothesize that the high expression level of ERK in nuptial pads is likely to promote the pigmentation pattern.

HSPB6, HSPB8, and HSPB11 are members of the small heat‐shock protein (sHSP) family, which plays a crucial role in maintaining proteostasis during oxidative stress [38, 39, 40]. On the other hand, CAT is a major antioxidant enzyme, which is able to defend cells against the harmful effects of reactive oxygen species (ROS) [41]. In our study, HSPB6, HSPB8, and HSPB11 showed a trend of upregulation in the nuptial pads, whereas CAT showed a trend of downregulation. Previous research has demonstrated that enhanced HSPB6 activity contributes to protein stability and protects against oxidative damage under stressful conditions [42]. Furthermore, CAT directly regulates H₂O₂ clearance efficiency and its low expression may potentially increase the risk of oxidative damage [43]. Based on these expression patterns and existing literature, we hypothesize that the upregulation of sHSPs may represent a potential compensator mechanism in response to CAT downregulation, although this remains to be functionally validated. Given our sample size, this interpretation remains speculative and requires further validation in future studies.

Muscle contraction occurs through the sliding of actin thin filaments along myosin thick filaments, a mechanism regulated by Ca²⁺‐induced repositioning of the troponin— tropomyosin complex [44, 45]. Skeletal muscle isolated from actin‐deficient mice has been shown to exhibit a marked reduction in force production [46]. Additionally, studies have reported a positive correlation between myosin heavy chain expression and the muscle bundle diameter of the medial rectus, a skeletal muscle responsible for controlling eyeball movement [47]. KEGG functional enrichment analysis revealed that DEGs were enriched in motor protein pathways, including significantly upregulated genes encoding myosin and actin, such as MYL1 and ACTA1, in nuptial pads. Therefore, given the known roles of these proteins in muscle function, the upregulation of myosin and actin in nuptial pads may be consistent with enhanced contractile properties and structural stability, which could contribute to the improved gripping capabilities required during amplexus. These correlative interpretations remain speculative and await validation through functional and morphological assays.

Vasodilation refers to the ability of blood vessels to widen their luminal diameter [48]. The dilation of the rat gingival arterioles was significantly attenuated following treatment with a COX‐2 inhibitors, and topical application of PGE2 on the gingiva elicited a dose‐dependent vasodilatory response in arterioles [49]. EETs, which are synthesized from AA via the CYP pathway, regulate ion transport and gene expression, thereby promoting vasodilation [50]. Our findings indicated that key enzymes involved in AA metabolism, such as COX‐2 and CYP2, were significantly downregulated in nuptial pads. Given these expression patterns and the known roles of AA metabolism mediators in vascular function, it is reasonable to hypothesize that reduced expression of these enzymes may lead to decreased production of downstream vasodilatory mediators. This, in turn, could potentially contribute to altered vascular tone and fluid balance, which may be associated with the morphological changes observed in nuptial pads.

5. Conclusion

In conclusion, the nuptial pads on the Finger Ⅰ of R. chensinensis, as a specialized reproductive structure, exhibit functional adaptability through the synergistic effect of multiple mechanisms. Structurally, they depend on enlarged glands volumes, thickened collagen fibers and muscle bundle, and vascular regulation that induces swelling, all of which provide stable mechanical support. Physiologically, enhanced oxidative metabolism during reproduction, together with cellular protective mechanisms such as pigmentation and increased expression of sHSPs, maintains metabolic homeostasis. Furthermore, downregulated keratin genes enhance secretory transport efficiency by reducing barriers to mucus release. Taken together, these adaptations, spanning structural stability, physiological regulation and metabolic homeostasis, are closely aligned with the core requirements of successful amplexus and ovulation stimulation, thereby ensuring reproductive success. These features clearly illustrate the evolutionary interplay between structure traits and reproductive functions, laying a foundation for better understanding of the mechanisms underlying nuptial pad function in anuran amphibians. Given that our study focuses on the breeding season, future comparative analyses of nuptial pad structure and molecular function between breeding and non‐breeding seasons will further clarify the seasonal plasticity of these adaptations, which remains an important and promising direction for our follow‐up research.

Author Contributions

Guoxiang Zhang: data curation, formal analysis, investigation, visualization, roles/writing – original draft, writing – review and editing. Guoxiang Zhang, Xinyi Li: experimental design and implementation. Shuangyan Luo: data analysis, investigation. Hongyuan Wang: funding acquisition, experimental design, software, investigation, methodology, conceptualization.

Ethics Statement

All experiment animals used in this study were in accordance with the Experimental Animal Care Regulations of Shaanxi Normal University and measures of Shaanxi Province for the administration of affairs concerning experimental animals.

Conflicts of Interest

The authors declare no conflicts of interest.

Declaration of AI Use

During the preparation of this manuscript, the authors used Doubao (Version 1.0)/DeepL in 2025 to assist with English language editing, grammar correction, and phrasing refinement. The AI tool was not involved in study design, data collection, data analysis, figure preparation, or the generation of any scientific conclusions. All content revised with the assistance of AI was carefully reviewed, edited, and approved by the authors, who take full responsibility for the accuracy and integrity of the work.

Zhang G., Li X., Luo S., and Wang H., “Structural–Molecular Linkage Shapes the Adaptive Traits of Nuptial Pads in Male Rana chensinensis ,” Cell Biochemistry and Function 44 (2026): e70268, 10.1002/cbf.70268.

Data Availability Statement

The data that has been used is confidential.