Role of Col1a2 in Collagen Deposition in the Carapace of the Chinese Soft-Shelled Turtle (Pelodiscus sinensis): From Molecular Evolution to Expression Profile and Then to Function Validation

Abstract

The carapace of the Chinese soft-shelled turtle (Pelodiscus sinensis) is rich in collagen and stands as a crucial economic trait for assessing its quality, as well as a key indicator for selective breeding. However, current studies on the mechanisms underlying collagen deposition in the carapace remain severely limited, significantly hindering progress in selective breeding. Here, the Col1a2 gene of P. sinensis was molecularly characterized for the first time. Analysis of gene structure, phylogenetic tree, and amino acid sequence homology revealed that Col1a2 is relatively conserved among tetrapods but divergent from fishes. Collinearity analysis identified the BET1-COL1A2-CASD1-SGCE gene block shared across all 14 representative vertebrates and found that the Col1a2 is located on the Z chromosome of Thamnophis elegans. Tissue expression analysis showed that Col1a1 was highly expressed in the heart, gonad, and lung. Additionally, Col1a1 expression levels markedly increased during carapace development, exhibiting a strongly positive correlation with the changes in collagen content of the carapace. In situ hybridization results revealed strong signal for the Col1a2 transcripts in fibroblasts of the dermal layer of P. sinensis carapace. Knockdown of the Col1a2 gene in the carapace cells of P. sinensis significantly reduced collagen content. Transcriptome analysis following Col1a2 knockdown identified several differentially expressed genes associated with collagen deposition, including Fbln2, IL-11, and Rspo4, as well as significantly enriched pathways such as the JAK-STAT signaling pathway, the apelin signaling pathway, and the Hippo signaling pathway. Our findings offer a molecular basis for elucidating the mechanisms of collagen deposition in the carapace of P. sinensis, while also supplying a potential target for the selective breeding of collagen-rich strains of P. sinensis.

1. Introduction

The Chinese soft-shelled turtle (Pelodiscus sinensis) is the most widely farmed turtle species in China [1,2] with an annual commercial output exceeding 300,000 tons for 13 consecutive years and reaching 541,600 tons in 2024 [3]. The carapace of P. sinensis contains abundant collagen, serving as its primary edible part and the main raw material for producing soft-shelled turtle shell gelatin, powder, and collagen peptide [4,5]; it can also be utilized to develop value-added products for nutrition, health care, beauty care and biomaterial [6,7,8]. Meanwhile, the calipash of the carapace, as the major locus for collagen deposition, represents a critical economic trait for assessing the quality of P. sinensis and serves as a key indicator for the selective breeding of new varieties [9,10]. However, studies on the mechanisms underlying collagen deposition in the carapace of P. sinensis are severely lacking, significantly hindering the progress of its selective breeding.

Collagen (COL) is the most abundant protein type in living organisms, accounting for approximately 30% of total protein mass [11,12]. Type I collagen is the predominant type of collagen, essential for the structural integrity of biological tissues, and is present in nearly all connective tissues and plays a major role in the extracellular matrix [13,14]. The Col1a2 gene is a member of the COL gene family, and the polypeptide chain it encodes is an important component of type I collagen [15,16]. Previous studies have shown that high-temperature incubation promotes collagen deposition in the carapace of P. sinensis and upregulates the expression level of Col1a2 [9]. Col1a2-derived peptides from human promote the accumulation of type I collagen, cell proliferation, cell migration, and elastin synthesis [17]. Knockout of Col1a2 in mice myofibroblasts resulted in a decrease in total collagen content [18]. The above findings suggest that the Col1a2 gene may be a key regulator of collagen deposition. However, there have been limited reports on the physiological functions of the Col1a2 gene in collagen deposition within the carapace of P. sinensis.

In this study, we analyzed the gene structure, collinearity, phylogenetic relationships, and amino acid sequence homology of P. sinensis Col1a2. Subsequently, its relative expression levels in adult tissues and embryonic carapaces were detected, along with its cellular localization within carapace tissues. Furthermore, the correlation between Col1a2 expression and collagen deposition was analyzed by measuring the total collagen content at different embryonic developmental stages. Finally, Col1a2 was knocked down in carapace cells, and changes in the total collagen content were measured. Transcriptome analysis identified key genes and candidate pathways affected by Col1a2 knockdown. Our findings establish a vital molecular foundation for elucidating the physiological functions of the Col1a2 gene in collagen deposition within the carapace of P. sinensis.

2. Results

2.1. Gene Structure and Collinearity Analysis

Analysis of the Col1a2 gene structure in 14 representative vertebrates using genomic annotation files revealed that mammals (Homo sapiens and Mus musculus), birds (Gallus gallus and Coturnix japonica), turtles (Chelonia mydas, Mauremys mutica, Mauremys reevesii, and P. sinensis), and snake (Thamnophis elegans) contain 52 introns and 51 coding sequences (CDS), respectively. The number of introns and CDS in fishes (Danio rerio and Oryzias latipes) is 50 and 49, respectively. Number of introns and CDS in Podarcis muralis, Xenopus tropicalis, and Pleurodeles waltl are 47 and 49, 51 and 50, and 47 and 48, respectively (Figure 1A). Collinearity analysis results indicated that Col1a2 is located on an autosome in H. sapiens, G. gallus, P. sinensis, X. tropicalis, and D. rerio, while it is uniquely situated on the Z sex chromosome in T. elegans. Additionally, the BET1-COL1A2-CASD1-SGCE gene block is highly conserved across all analyzed species. Meanwhile, the BET1-COL1A2-CASD1-SGCE-PPP1R9A gene block is conserved across species other than H. sapiens. The GNG11-ACAD6-BET1-COL1A2-CASD1-SGCE-PPP1R9A gene block is conserved in G. gallus, T. elegans, and P. sinensis (Figure 1B).

2.2. Phylogenetic Relationships and Amino Acid Sequence Alignment

Systematic evolutionary analysis showed that the Col1a2 gene clusters into six lineages, including mammals (H. sapiens and M. musculus), birds (G. gallus and C. japonica), squamates (T. elegans and P. muralis), testudines (C. mydas, M. mutica, M. reevesii and P. sinensis) amphibians (X. tropicalis and P. waltl), and fishes (D. rerio and O. latipes). Col1a2 gene of P. sinensis is most closely related to turtles, followed by birds, squamates, mammals, amphibians, and fishes (Figure 2). Analysis of amino acid sequence homology revealed that P. sinensis Col1a1 protein exhibits the highest identity with turtles exceeding 91%, followed by birds exceeding 85%, squamates exceeding 80%, mammals exceeding 77%, amphibians exceeding 75%, and fishes exceeding 70% (Figure S1).

2.3. Expression of Col1a2 in Adult Tissues and Embryonic Carapaces

Col1a2 is widely expressed in adult tissues, with the highest expression levels observed in the heart and gonad (p < 0.05), followed by the lung, muscle, and spleen, while its expression is relatively low in the liver and kidney (Figure 3A). Simultaneously, the expression of Col1a2 in the carapace of embryos at stages 14, 18, and 22 was detected and found that the expression level of Col1a2 significantly increased with the development of the embryonic carapaces (p < 0.01), reaching a peak at stage 22 (Figure 3B).

2.4. Cellular Localization of Col1a2 in the Carapace Tissues

The cellular localization of Col1a2 in the carapace was investigated using in situ hybridization experiments. The carapace of P. sinensis primarily consists of the epidermal layer, dermal layer, and subcutaneous tissue (Figure 4A). The epidermal layer is mainly composed of basal cells, which are neatly arranged in a short columnar or cubic shape, while the dermal layer is relatively loose and predominantly consists of fibroblasts, exhibiting an irregular spindle-shaped morphology (Figure 4B,C). The sense probe for the Col1a2 gene detected widespread signaling in epidermal and dermal cells, with the strongest signals observed in dermal fibroblasts (Figure 4B), whereas the antisense probe used as a control showed no detectable signal (Figure 4C).

2.5. Collagen Measurement and Correlation Analysis

Meanwhile, we also measured the total collagen content in the carapaces of P. sinensis embryos at stages 14, 18, and 22, and the results showed that total collagen content significantly increased with the carapace development (p < 0.01) (Figure 5A). Correlation analysis indicates that the relative Col1a2 expression is significantly positively correlated with the total collagen content of embryonic carapace (p = 0.0109, R = 0.7924) (Figure 5B).

2.6. Knockdown of Col1a2 in Carapace Cells

Subsequently, we knocked down the Col1a2 gene in P. sinensis carapace cells, and results showed that transfection with siRNA_510, siRNA_1340, and siRNA_3621 significantly reduced the relative expression of Col1a2 in carapace cells (p < 0.01), with knockdown efficiencies of approximately 71, 42, and 58%, respectively (Figure 6A). The group with the highest knockdown efficiency was selected for total collagen content measurement, revealing that Col1a2 knockdown significantly reduced collagen content in carapace cells (p < 0.01) (Figure 6B).

2.7. Transcriptome Analysis

2.7.1. Identify Differentially Expressed Genes

Transcriptome sequencing was performed using the experimental group with the highest knockdown efficiency and control group. After filtering, a total of 6,226,039,664 to 7,291,093,372 clean reads were obtained, with Q30 values exceeding 94.83%, GC content ranging from 45.91 to 46.47%, and alignment rates to the reference genome exceeding 89.84% (Table S1). Principal component analysis results indicated a difference between the knockdown group (siRNA) and the control group (NC), while samples within each group showed good similarity (Figure 7A). The qPCR validation results confirmed that Col1a2 was significantly knocked down (p < 0.01) (Figure S2). The above results demonstrate that our transcriptomic sequencing exhibits high quality and reliable outcomes, making it suitable for subsequent bioinformatics analysis. A total of 109 differentially expressed genes (DEGs) were identified, with 58 upregulated and 51 downregulated (Figure 7B). Among these, genes associated with collagen synthesis included Fbln2, IL-11, Rspo4, and so on (Table S2).

2.7.2. GO and KEGG Analysis

The Gene Ontology (GO) analysis suggested that DEGs were primarily enriched in the cellular processes, the cell and the cell part, and the binding, of which 475 GO terms showed significant enrichment (Figure 8A) (Table S3). Additionally, the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis identified six significantly enriched pathways, including the herpes simplex virus 1 infection, the rheumatoid arthritis, the longevity regulating pathway-worm, the apelin signaling pathway, the JAK-STAT signaling pathway, and the hematopoietic cell lineage (Figure 8B) (Table S3).

2.7.3. GSEA Analysis

To further investigate the effects of knockdown on biological pathways in carapace cells, we performed a gene set enrichment analysis on all expressed genes. The Gene Set Enrichment analysis based on the GO database identified 295 significantly enriched GO terms (Figure 9A), while analysis based on the KEGG database identified 12 significantly enriched KEGG pathways (Figure 9B). Among these, terms and pathways associated with collagen deposition included the regulation of hippo signaling, the steroid biosynthesis, the oxidative phosphorylation, and so on (Table S4).

3. Discussion

Collagen from aquatic animals holds broad application prospects in functional foods, cosmetics, and biomaterials [19,20]. Therefore, investigating the molecular mechanisms underlying collagen deposition is a crucial step in breeding aquatic species enriched in collagen. This study first identified the molecular characteristics of the Col1a2 gene, revealing that its gene structure is conserved across major tetrapods including mammals, birds, and turtles, which may be related to the critical role of Col1a2 in maintaining the precise assembly of the triple helix structure of type I collagen [21,22]. Collinearity analysis detected a highly conserved gene block comprising BET1-COL1A2-CASD1-SGCE across 14 representative vertebrates, indicating this block originated from their common ancestor and suggesting this topological structure may be crucial for the coordinated regulation of genes within this region [23]. Bet1 is a gene member involved in the transport of secretory proteins from the endoplasmic reticulum to the Golgi apparatus [24,25]. The protein encoded by the Casd1 gene possesses a serine-glycine-asparagine-histidine hydrolase domain [26] and the Sgce depletion affects the accumulation of extracellular matrix [27]. Col1a2 is an extracellular matrix protein that requires intracellular synthesis, processing, and transport to the extracellular space [28,29]. Glycine is a key component of collagen and also the primary product during collagen synthesis and hydrolysis [30,31]. Therefore, we speculate that this gene block may function in the production, transport, and hydrolysis of collagen. Remarkably, the Col1a2 gene evolved from an autosome to the sex chromosome Z in T. elegans, implying that its function is likely linked to sex. 17β-Estradiol sex-specifically regulates type I collagen expression in cardiac fibroblasts via estrogen receptors [32]. Gender-specific transcription factor Dsx regulates collagen expression and secretion in the basement membrane through dPlod-dependent and independent pathways [33]. Meanwhile, similar findings were observed in phylogenetic tree and amino acid homology alignment, with Col1a2 exhibiting closer genetic distance and higher sequence identity among tetrapods.

Subsequently, the tissue expression profiles of Col1a2 in P. sinensis were investigated, showing significantly high expression of Col1a2 in adult heart, gonad, and lung. Collagen is primarily produced by fibroblasts [13], which are essential components of the heart, playing a vital role in shaping its structure and enabling its function [34]. In situ hybridization results also revealed a similar finding that Col1a2 is primarily localized in fibroblasts within the dermal layer of the carapace. Type I collagen is present in nearly all connective tissues, which provide crucial structural support and functional connectivity within the lung [13]. Collagen is the major extracellular matrix component of the ovary [35] and also constitutes the main component of the seminiferous tubule basement membrane, playing a role in maintaining the blood-testis barrier and supporting the proliferation and differentiation of Leydig cells [36,37]. Therefore, the significant enrichment of Col1a2 in these tissues may be related to the essential role of collagen synthesized by Col1a2 in the functions of these organs. Moreover, as the carapace develops, the expression level of Col1a2 significantly increases, showing a marked positive correlation with the trend of collagen deposition in the carapace, which matches the functions of Col1a2 in collagen deposition. The p300/CBP co-activator enhances TGF-β/Smad3-induced Col1a2 promoter activity and promotes type I collagen expression [38]. Cardiac fibrosis is characterized by increased deposition of fibrous collagen, and Scleraxis transactivates the Col1a2 gene, thereby accelerating the progression of fibrosis [39].

Furthermore, we investigated the functions of the Col1a2 gene in P. sinensis carapace cells. Knockdown of Col1a2 significantly reduced collagen content in carapace cells, indicating its role as a key regulator of collagen deposition in the carapace of P. sinensis and transcriptome analysis following Col1a2 knockdown also identified several genes and pathways associated with collagen deposition. Knockout of Fbln2 in mammary epithelial cells of mouse disrupts collagen structure [40], while knockdown of Fbln2 inhibits TGF-β1 by downregulating Vtn, leading to collagen metabolism imbalance in lung fibroblasts from human [41]. Oxygen-glucose deprivation promotes fibroblast senescence and collagen expression by inhibiting IL-11 [42]. IL-11 induces myofibroblast differentiation and stimulates excessive collagen deposition in the lung [43]. Rspo4 is a critical activator of Wnt/β-catenin signaling, which is a core pathway regulating collagen deposition [44]. Overexpression of Vgll3 in scar cells increases Wnt2 and β-catenin protein levels, thereby promoting the deposition of type I and III collagen [45]. Additionally, the JAK-STAT signaling pathway is also involved in collagen deposition. PDGF, a JAK-STAT activator, increases Stat3 expression and phosphorylation levels in cardiac fibroblasts while promoting type I collagen secretion [46]. Apelin-13, a key effector in the apelin signaling pathway, regulates type I collagen expression in retinal pigment epithelial cells via the PI3K/Akt and MEK/Erk pathways [47]. Type I collagen induces Ddr1 activation, which antagonizes the Hippo signaling pathway by promoting the recruitment of Pp2aa to Mst1 [48], and steroids can influence the extracellular maturation process of collagen [49]. Fibroblasts increase ATP production through oxidative phosphorylation to supply energy for sustaining collagen synthesis [50]. Taken together, the key functions of Col1a2 in collagen deposition of P. sinensis carapace may be mediated by the aforementioned genes and pathways. However, these findings remain preliminary, and the physiological function of Col1a2 in collagen deposition within the carapace of P. sinensis requires further in vivo validation. Concurrently, the transcriptional regulatory relationships between Col1a2 and DEGs, as well as the regulatory cascades linking Col1a2 to candidate pathways, may represent future research directions.

4. Materials and Methods

4.1. Eggs Incubation and Sample Collection

A total of 100 fertilized eggs were collected from Xianghui Aquaculture Co., Ltd. (Guangxi, Yulin, China), and hatched using the medium-free incubation method previously developed in our study [51]. The incubation temperature of the constant-temperature incubator (LRH-150 F, China) is set at 31 ± 1 °C, with humidity set between 75 and 85%. Samples of embryonic carapace tissue were collected at stages 14, 18, and 22 of fertilized egg development, with at least three biological replicates per stage. Additionally, further embryonic carapace tissues were collected and fixed in Borne’s solution for preparation of paraffin sections. The embryonic development stages of P. sinensis were identified using the methods described in previous studies [52]. Additionally, six adult P. sinensis aged 3 years were humanely euthanized. Tissues including the heart, liver, spleen, lung, kidneys, intestine, muscle, brain, and gonad were sampled and preserved in liquid nitrogen.

4.2. Molecular Characterization of Col1a2

Based on the chromosome-level genome of P. sinensis constructed by our team (GCA_048772765.1), we obtained the gene information for Col1a2. Subsequently, the genomes and annotation files for H. sapiens, M. musculus, G. gallus, C. japonica, T. elegans, P. muralis, C. mydas, M. mutica, M. reevesii, X. tropicalis, P. waltl, D. rerio, and O. latipes were downloaded from the NCBI database. The gene structure of Col1a2 in 14 representative vertebrates was visualized using the TBtools-II software [53] based on genomic annotation files. Synteny analysis was processed via the NCBI genome browser and Ensembl [54,55]. The sequences of Col1a2 amino acids (aa) were aligned using MUSCLE [56] and a phylogenetic tree was constructed using MEGA X [57] with the maximum likelihood method and 1000 bootstraps. The phylogenetic tree was beautified using ChiPlot [58].

4.3. RNA Extraction and Reverse Transcription

Total RNA of all tissues was extracted using the FastPure Cell/Tissue Total RNA Extraction Kit (Vazyme, Nanjing, Jiangsu, China) according to the manufacturer’s instructions. The quality and concentration of total RNA were detected using 1% gel electrophoresis and NanoQ™ (Thermo Scientific, Madison, WI, USA), respectively. High-quality total RNAs were reverse transcribed into the first-strand cDNA using the HiScript IV 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, Jiangsu, China).

4.4. Quantitative Real-Time PCR

Quantitative real-time PCR (qPCR) was employed to detect the relative expression level of the Col1a2 gene in adult tissues, embryonic carapaces, and cells. The qPCR reaction program was as follows: 95 °C for 15 s, 60 °C for 20 s, 72 °C for 20 s, repeated for 40 cycles. Ef1α was used as the control gene [59,60], and the 2^−ΔΔCt method was applied for normalization of the target gene. The primers used for the experiment are shown in Table S5.

4.5. In Situ Hybridization

Based on the Col1a2 gene sequence of P. sinensis, specific primers ISH-Col1a2-F and ISH-Col1a2-R were designed (Table S5). Positive or antisense RNA probes with T7 or Sp6 promoters were synthesized via in vitro transcription using the DIG RNA Labeling Kit (Roche, Penzberg, BY, Germany). Paraffin sections of embedded embryonic carapace tissue were prepared at a thickness of 6 μm. The in situ hybridization and hematoxylin and eosin staining processes were performed according to the description in the previous studies [1,61]. Observation and photography using the Axio Observer Z1 microscope (ZEISS, Oberkochen, BW, Germany).

4.6. Measurement of Total Collagen Content

Total collagen content in the carapaces of P. sinensis embryos at stages 14, 18, and 22 was measured using a hydroxyproline assay kit (Nanjing Jiancheng Biological Engineering Institute, Nanjing, Jiangsu, China). Hydroxyproline (HYP) is a marker for collagen, found almost exclusively in collagen, and therefore it is widely used to quantify total collagen content [62,63]. Concretely, an appropriate amount of carapace tissue was added to 1 mL of 4 mol/L NaOH solution and hydrolyzed at 110 °C for 20 min. The solution was adjusted to neutrality using 1 mol/L HCl. Chloramine-T was added to the hydrolysate for oxidation reaction, followed by the addition of dimethylaminobenzaldehyde with incubation at 60 °C for 15 min to develop color. The absorbance values of each sample were measured at 550 nm using a spectrophotometer (BioTek, Winooski, VT, USA) and the standard curve was generated using hydroxyproline standard samples.

4.7. RNA Interference

To investigate the physiological functions of Col1a2 in collagen deposition of the carapace in P. sinensis, three siRNAs targeting P. sinensis Col1a2 (siRNA_510, siRNA_1340, and siRNA_3621) were designed and synthesized (GenePharma, Shanghai, China). Negative control (NC) siRNA [64] and three siRNA sequences targeting Col1a2 are provided in Table S5. The carapace cells of P. sinensis were seeded into 12-well plates and transfected with siRNAs encapsulated in liposomes (Beyotime, Shanghai, China) upon reaching 90% confluence. After 48 h, total RNAs were extracted and reverse transcribed as described in Section 4.3. The efficiency of Col1a2 knockdown was analyzed via qPCR, and changes in the total collagen content were detected.

4.8. Transcriptome Sequencing

Total RNAs from the experimental group with the highest knockdown efficiency and the NC group were selected for transcriptomic sequencing using Illumina Novaseq 6000 (Gene Denovo, Guangzhou, Guangdong, China). After quality control using fastp [65], clean reads were aligned to the P. sinensis reference genome using HISAT2 2.4 [66] and assembled with StringTie v1.3.1 [67]. The FPKM (fragments per kilobase of exon model per million mapped fragments) values were calculated using RSEM [68], and differentially expressed genes were identified via DESeq2 [69] with the threshold set at absolute fold change (FC) > 1.5 and p value < 0.05. Functional enrichment analysis was performed using the GO [70] and the KEGG [71] databases, with a p-value < 0.05 indicating significant enrichment. GSEA [72] was performed on all expressed genes, with the significant enrichment threshold set at NOM p-value < 0.05 and |NES| > 1.

4.9. Statistical Analysis

Each experiment was conducted independently at least three times. All experimental data are presented as mean ± standard error of mean (SEM) and analyzed using SPSS Statistics 26. Significance testing was performed using one-way analysis of variance and t-tests, with p < 0.05 considered statistically significant. Pearson correlation analysis was used to evaluate the correlation between the relative expression of Col1a2 and the total collagen content in the carapace at different embryonic developmental stages, with the linear relationship between variables constructed using a linear regression model.

5. Conclusions

In summary, we have characterized the Col1a2 gene in P. sinensis for the first time, revealing its molecular evolutionary features. Expression profiling and cellular localization studies demonstrated that its expression pattern is significantly correlated with collagen deposition in the carapace of P. sinensis. Functional studies indicated that it serves as a key factor in collagen deposition of P. sinensis carapace by interacting with genes like Fbln2, IL-11, and Rspo4, and through regulatory relationships with pathways including the JAK-STAT signaling pathway, the apelin signaling pathway, and the Hippo signaling pathway. This study provides molecular evidence for the role of Col1a2 in regulating collagen deposition in P. sinensis carapace, offering a molecular target for breeding collagen-rich P. sinensis strain.