Whole-Genome Identification and Characterization of the DKK Gene Family and Its Transcription Profiles: An Analysis of the Chinese Soft-Shell Turtle (Pelodiscus sinensis).

The DKK family is a canonical small family of WNT antagonists. Though recent studies have suggested that the DKK gene family may be involved in sex differentiation in Pelodiscus sinensis, there are still a lot of things about the DKK gene family that we do not know. In this study, we used bioinformatics methods to identify members of the DKK gene family in P. sinensis and analyzed their phylogeny, covariance, gene structure, structural domains, promoter conserved sites, signal peptides, gonadal transcription factors, transcriptional profiles, and tissue expression profiles. Additionally, qRT-PCR results were utilized for the validation and preliminary investigation of the function of the DKK gene family in P. sinensis. The results showed that the DKK gene family is divided into six subfamilies, distributed on six different chromosomal scaffolds containing different gene structures and conserved motifs with the same structural domains, and all of the members were secreted proteins. Our transcriptional profiling and embryonic expression analysis showed that DKKL1 and DKK4 were significantly expressed in the testes, whereas DKK1 and DKK3 were significantly upregulated in the ovaries. This suggests a potential function in sex differentiation in P. sinensis. Our results may provide a basic theoretical basis for the sex differentiation process in P. sinensis.

Identification of Sex-Specific Markers and Candidate Genes Using WGS Sequencing Reveals a ZW-Type Sex-Determination System in the Chinese Soft-Shell Turtle (Pelodiscus sinensis).

Male and female Chinese soft-shelled turtles (Pelodiscus sinensis) have sex-dimorphic growth patterns, and males have higher commercial value because of their larger size and thicker calipash. Thus, developing sex-specific markers is beneficial to studies on all-male breeding in P. sinensis. Here, we developed an accurate and efficient workflow for the screening of sex-specific sequences with ZW or XY sex determination systems. Based on this workflow, female and male P. sinensis reference genomes of 2.23 Gb and 2.26 Gb were obtained using de novo assembly. After aligning and filtering, 4.01 Mb female-specific sequences were finally identified. Subsequently, the seven developed sex-specific primer pairs were 100% accurate in preliminary, population, and embryonic validation. The presence and absence of bands for the primers of P44, P45, P66, P67, P68, and P69, as well as two and one bands for the PB1 primer, indicate that the embryos are genetically female and male, respectively. NR and functional annotations identified several sex-determining candidate genes and related pathways, including Ran, Eif4et, and Crkl genes, and the insulin signaling pathway and the cAMP signaling pathway, respectively. Collectively, our results reveal that a ZW-type sex-determination system is present in P. sinensis and provide novel insights for the screening of sex-specific markers, sex-control breeding, and the studies of the sex determination mechanism of P. sinensis.

Comparative genomic survey and functional analysis of DKKL1 during spermatogenesis in the Chinese soft-shelled turtle (Pelodiscus sinensis).

A feature of the Chinese soft-shelled turtle (Pelodiscus sinensis) is seasonal spermatogenesis; however, the underlying molecular mechanism is not well clarified. Here, we firstly cloned and characterized P. sinensis DKKL1, and then performed comparative genomic studies, expression analysis, and functional validation. P. sinensis DKKL1 had 2 putative N-glycosylation sites and 16 phosphorylation sites. DKKL1 also had classic transmembrane structures that were extracellularly localized. DKKL1's genetic distance was close to turtles, followed by amphibians and mammals, but its genetic distance was far from fishes. DKKL1 genes from different species shared distinct genomic characteristics. Meanwhile, they were also relatively conserved among themselves, at least from the perspective of classes. Notably, the transcription factors associated with spermatogenesis were also identified, containing CTCF, EWSR1, and FOXL2. DKKL1 exhibited sexually dimorphic expression only in adult gonads, which was significantly higher than that in other somatic tissues (P < 0.001), and was barely expressed in embryonic gonads. DKKL1 transcripts showed a strong signal in sperm, while faint signals were detected in other male germ cells. DKKL1 in adult testes progressively increased per month (P < 0.05), displaying a seasonal expression trait. DKKL1 was significantly downregulated in testes cells after the sex hormones (17ß-estradiol and 17α-methyltestosterone) and Wnt/ß-catenin inhibitor treatment (P < 0.05). Likewise, the Wnt/ß-catenin inhibitor treatment dramatically repressed CTCF, EWSR1, and FOXL2 expression. Conversely, they were markedly upregulated after the 17ß-estradiol and 17α-methyltestosterone treatment, suggesting that the three transcription factors might bind to different promoter regions, thereby negatively regulating DKKL1 transcription in response to the changes in the estrogen and androgen pathways, and positively controlling DKKL1 transcription in answer to the alterations in the Wnt/ß-catenin pathway. Knockdown of DKKL1 significantly reduced the relative expression of HMGB2 and SPATS1 (P < 0.01), suggesting that it may be involved in seasonal spermatogenesis of P. sinensis through a positive regulatory interaction with these two genes. Overall, our findings provide novel insights into the genome evolution and potential functions of seasonal spermatogenesis of P. sinensis DKKL1.

Reproductive Output Reveals the Maternal Effects on Offspring Size-Number Trade-Off in Cultured Asian Yellow Pond Turtle (Mauremys mutica).

Offspring size-number trade-off is a critical component of life-history theory and is important for further understanding the reproductive strategies of animals. The relationship between this trade-off and maternal size has been explored in several turtle species, except for the Asian yellow pond turtle, Mauremys mutica. To investigate how the maternal condition affects offspring size and number, we explored the relationships among the maternal body size and the number and size of cultured M. mutica hatchlings using a 4-year dataset. Our results showed that different females not only produced different sizes of offspring but also produced different numbers of offspring. No trade-off in egg size number was detected. According to regression analysis, we did not find that the maternal body size significantly influenced the offspring mass; however, we detected that the offspring size was significantly correlated with the clutch size and maternal age. The mean body mass of offspring increased with maternal age, and the clutch size varied significantly over four years, which was correlated with offspring size, maternal body size and age. However, the number of offspring per female increased with the maternal plastron length rather than age. Our results were inconsistent with the optimal offspring size theory in that females did not increase their offspring size but rather increased the offspring number to increase their fitness, which will also provide a basis for the efficient cultivation management of turtles.

Single-Cell Atlas Reveals the Hemocyte Subpopulations and Stress Responses in Asian Giant Softshell Turtle during Hibernation.

Hibernation in turtle species is an adaptive survival strategy to colder winter conditions or food restrictions. However, the mechanisms underlying seasonal adaptions remain unclear. In the present study, we collected hemocytes from Pelochelys cantorii and compared the molecular signature of these cells between the active state and hibernation period based on single-cell RNA sequencing (scRNA-seq) analysis. We found six cell types and identified a list of new marker genes for each cell subpopulation. Moreover, several heat shock genes, including the Hsp40 family chaperone gene (DNAJ) and HSP temperature-responsive genes (HSPs), were upregulated during the hibernation period, which predicted these genes may play crucial roles in the stress response during hibernation. Additionally, compared to hemocytes in the active state, several upregulated differentially expressed immune-related genes, such as stat1, traf3, and socs6, were identified in hemocytes during the hibernation period, thus indicating the important immune function of hemocytes. Therefore, our findings provide a unified classification of P. cantorii hemocytes and identify the genes related to the stress response, thereby providing a better understanding of the adaptive mechanisms of hibernation.

Genome-wide identification, evolution and expression analysis of bone morphogenetic protein (BMP) gene family in chinese soft-shell turtle (Pelodiscus sinensis).

Introduction: Bone morphogenetic proteins (BMPs) play a crucial role in bone formation and differentiation. Recent RNA-Seq results suggest that BMPs may be involved in the sex differentiation of P. sinensis, yet more relevant studies about BMPs in P. sinensis are lacking. Methods: Herein, we identified BMP gene family members, analyzed the phylogeny, collinear relationship, scaffold localization, gene structures, protein structures, transcription factors and dimorphic expression by using bioinformatic methods based on genomic and transcriptomic data of P. sinensis. Meanwhile, qRT-PCR was used to verify the RNA-Seq results and initially explore the function of the BMPs in the sex differentiation of P. sinensis. Results: A total of 11 BMP genes were identified, 10 of which were localized to their respective genomic scaffolds. Phylogenetic analysis revealed that BMP genes were divided into eight subfamilies and shared similar motifs ("WII", "FPL", "TNHA", "CCVP", and "CGC") and domain (TGF-ß superfamily). The results of the sexually dimorphic expression profile and qRT-PCR showed that Bmp2, Bmp3, Bmp15l, Bmp5, Bmp6 and Bmp8a were significantly upregulated in ovaries, while Bmp2lb, Bmp7, Bmp2bl and Bmp10 were remarkable upregulated in testes, suggesting that these genes may play a role in sex differentiation of P. sinensis. Discussion: Collectively, our comprehensive results enrich the basic date for studying the evolution and functions of BMP genes in P. sinensis.

Expression and Characterization of the Spats1 Gene and Its Response to E2/MT Treatment in the Chinese Soft-Shelled Turtle (Pelodiscus sinensis).

Spats1 (spermatogenesis-associated, serinerich 1) has been characterized as a male-biased gene which acts an important role in the germ cell differentiation of mammals. Nevertheless, the function of Spats1 in the Chinese soft-shelled turtle (P. sinensis) has not yet been reported. To initially explore the expression of Spats1 in P. sinensis and its response to sex steroid treatment, we cloned the CDS of Spats1 for the first time and analyzed its expression profile in different tissues, including the testes in different seasons. The Spats1 cDNA fragment is 1201 base pairs (bp) in length and contains an open reading frame (ORF) of 849 bp, which codes for 283 amino acids. Spats1 mRNA was highly expressed in the testes (p < 0.01) and barely detectable in other tissues. In P. sinensis, the relative expression of Spats1 also responsive to seasonal changes in testis development. In summer (July) and autumn (October), Spats1 gene expression was significantly higher in the testes than in other seasons (p < 0.05). Spats1 mRNA was found to be specifically expressed in germ cells by chemical in situ hybridization (CISH), and it was mainly located in primary spermatocytes (Sc1), secondary spermatocytes (Sc2) and spermatozoa (St). Spats1 expression in embryos was not significantly changed after 17α-methyltestosterone (MT)and 17ß-estradiol (E2) treatment. In adults, MT significantly induced Spats1 expression in male P. sinensis. However, the expression of Spats1 in testes was not responsive to E2 treatment. In addition, the expression of Spats1 in females was not affected by either MT or E2 treatment. These results imply that Spats1 is a male-specific expressed gene that is mainly regulated by MT and is closely linked to spermatogenesis and release in P. sinensis.

Whole-Transcriptome Analysis Identifies Gender Dimorphic Expressions of Mrnas and Non-Coding Rnas in Chinese Soft-Shell Turtle (Pelodiscus sinensis).

In aquaculture, the Chinese soft-shelled turtle (Pelodiscus sinensis) is an economically important species with remarkable gender dimorphism in its growth patterns. However, the underlying molecular mechanisms of this phenomenon have not been elucidated well. Here, we conducted a whole-transcriptome analysis of the female and male gonads of P. sinensis. Overall, 7833 DE mRNAs, 619 DE lncRNAs, 231 DE circRNAs, and 520 DE miRNAs were identified. Some "star genes" associated with sex differentiation containing dmrt1, sox9, and foxl2 were identified. Additionally, some potential genes linked to sex differentiation, such as bmp2, ran, and sox3, were also isolated in P. sinensis. Functional analysis showed that the DE miRNAs and DE ncRNAs were enriched in the pathways related to sex differentiation, including ovarian steroidogenesis, the hippo signaling pathway, and the calcium signaling pathway. Remarkably, a lncRNA/circRNA-miRNA-mRNA interaction network was constructed, containing the key genes associated with sex differentiation, including fgf9, foxl3, and dmrta2. Collectively, we constructed a gender dimorphism profile of the female and male gonads of P. sinensis, profoundly contributing to the exploration of the major genes and potential ncRNAs involved in the sex differentiation of P. sinensis. More importantly, we highlighted the potential functions of ncRNAs for gene regulation during sex differentiation in P. sinensis as well as in other turtles.

The Seasonal and Stage-Specific Expression Patterns of HMGB2 Suggest Its Key Role in Spermatogenesis in the Chinese Soft-Shelled Turtle (Pelodiscus sinensis).

HMGB2, a member of the high-mobility group (HMG) proteins, was identified as a male-biased gene and plays a crucial role in the germ cells differentiation of mammals. However, its role in spermatogenesis of turtle is still poorly understood. Here, we cloned the Pelodiscus sinensis HMGB2 and analyzed its expression profile in different tissues and in testis at different developmental ages. P. sinensis HMGB2 mRNA was highly expressed in the testis of 3-year-old turtles (P < 0.01), but was hardly detected in ovaries and other somatic tissues. The results of chemical in situ hybridization (CISH) showed that HMGB2 mRNA was specifically expressed in germ cells, where it was mainly distributed in round spermatids and sperm, but not detected in somatic cells, spermatogonia, primary spermatocytes, or secondary spermatocyte. The relative expression of HMGB2 also responded to seasonal changes in testis development in P. sinensis. In different seasons of the year, the relative expression of HMGB2 transcripts in the testis of 1 year and 2 year olds showed an overall upward trend, whereas, in the testis of 3 year old, it peaked in July and then declined in October. Moreover, in April and July, with an increase in ages, the expression of HMGB2 transcripts showed an upward trend. However, in January and October, there was a decline in expression in testis in 3-year-old turtles. These results showed that HMGB2 is closely related to spermatogenesis in P. sinensis.

Transcriptome analysis reveals key genes and pathways related to sex differentiation in the Chinese soft-shelled turtle (Pelodiscus sinensis).

Most vertebrates exhibit sexual dimorphisms in size, colour, behaviour, physiology and many others. The Chinese soft-shelled turtle (Pelodiscus sinensis) male individuals reach a larger size than females which produce significant economic implications in aquaculture. However, the mechanisms of sex determination and plastic patterns of sex differentiation in P. sinensis remain unclear. Here, comparative transcriptome analysis on male and female embryonic gonads prior to gonad formation and stages mediated gonadal differentiation of P. sinensis were performed to characterize the potential sex-related genes and their molecular pathways in P. sinensis. A total of 6369 differentially expressed genes (DEGs) were identified from day 9 and day 16 and assigned to 626 GO pathways and 161 KEGG signalling pathways, including ovarian steroidogenesis pathway, steroid hormone biosynthesis pathways, and the GnRH signalling pathway (P < 0.05). Moreover, protein interaction network analyses revealed that Akr1c3, Sult2b1, Sts, Cyp3a, Cyp1b1, Sox30 and Lhx9 might be key candidate genes for sex differentiation in P. sinensis. These data provide a genomic rationale for the sex differentiation of P. sinensis and enrich the candidate gene pool for sex differentiation.

Identification and analysis of novel microRNAs provide insights to reproductive capacity of the cultured Asian yellow pond turtle Mauremys mutica.

The crucial roles of miRNAs in regulating animal growth, development, and disease resistance have been extensively reported, but their roles in relation to the reproductive capacity of aquatic animals (numbers of eggs laid and hatchlings), especially reptiles, remain unclear. In this study, high-throughput sequencing technology was used to screen miRNAs related to reproductive capacity based on the construction of a cDNA library of ovaries from higher-fecundity (HF) and lower-fecundity (LF) M. mutica. The results showed that 15,767,494 (93.98%) and 14,137,621 (94.17%) high-quality reads were obtained from the HF and LF groups, respectively. We screened 131 miRNAs that were differentially expressed between the HF and LF groups, of which 78 were upregulated and 53 were downregulated compared with the M. mutica reference genome. GO and KEGG pathway enrichment analyses of the target genes of differentially expressed miRNAs revealed significant differences in the enrichment frequencies of genes associated with ATP binding and proteolysis between the HF and LF groups, while the tricarboxylic acid cycle, glucagon signaling pathway and vitamin B6 metabolic pathway were shown to potentially help determine reproductive capacity. Ten miRNAs were verified by qRT-PCR to confirm the reliability and accuracy of the sequencing results, and a miRNA-mRNA target gene interaction network was constructed. These results will further our understanding of the regulatory mechanism of miRNAs in regards to turtle reproductive capacity.

BTOB: Extending the Biased GWAS to Bivariate GWAS.

In recent years, a number of literatures published large-scale genome-wide association studies (GWASs) for human diseases or traits while adjusting for other heritable covariate. However, it is known that these GWASs are biased, which may lead to biased genetic estimates or even false positives. In this study, we provide a method called "BTOB" which extends the biased GWAS to bivariate GWAS by integrating the summary association statistics from the biased GWAS and the GWAS for the adjusted heritable covariate. We employ the proposed BTOB method to analyze the summary association statistics from the large scale meta-GWASs for waist-to-hip ratio (WHR) and body mass index (BMI), and show that the proposed approach can help identify more susceptible genes compared with the corresponding univariate GWASs. Theoretical results and simulations also confirm the validity and efficiency of the proposed BTOB method.

A polynomial algorithm for best-subset selection problem.

Best-subset selection aims to find a small subset of predictors, so that the resulting linear model is expected to have the most desirable prediction accuracy. It is not only important and imperative in regression analysis but also has far-reaching applications in every facet of research, including computer science and medicine. We introduce a polynomial algorithm, which, under mild conditions, solves the problem. This algorithm exploits the idea of sequencing and splicing to reach a stable solution in finite steps when the sparsity level of the model is fixed but unknown. We define an information criterion that helps the algorithm select the true sparsity level with a high probability. We show that when the algorithm produces a stable optimal solution, that solution is the oracle estimator of the true parameters with probability one. We also demonstrate the power of the algorithm in several numerical studies.

Molecular cloning of ESR1, BMPR1B, and FOXL2 and differential expressions depend on maternal age and size during breeding season in cultured Asian yellow pond turtle (Mauremys mutica).

The reproductive capacity (egg size and egg number) of most of oviparous animals, including the Asian yellow pond turtle (Mauremys mutica), is constrained by the maternal age and body size, but the mechanism determining the maternal reproductive ability remains unclear. To disclose how maternal age and size affect reproductive ability of M. mutica, we first identified the full-length cDNAs from estrogen receptor 1 (ESR1), bone morphogenetic protein receptor 1B (BMPR1B), and forkhead box L2 (FOXL2). The ESR1 open reading frame (ORF) was 1, 767 bp encoding 588 amino acids. For BMPR1B, the ORF was 1599 bp encoding 532 amino acids, and an ORF of 906 bp encoding 301 amino acids was identified in FOXL2. The effects of maternal age and size on the expression of ESR1, BMPR1B, and FOXL2 in the ovary, brain, and uterus showed that ESR1 expression in large females was significantly lower than that in small females in the brain, but body size did not affect ESR1 expression in the ovary. The expression of ESR1 was significantly different in the different age groups and size groups, and there was interaction detected between maternal age and body size. However, BMPR1B expression in the ovary, brain, and uterus was independent of maternal age and size. In addition, we found different FOXL2 expression patterns between the brain and uterus, while detected interaction of female age and size in the brain and ovary. Our results imply the complexity and diversity of maternal age and size in regulating the expression of genes related to reproduction. These results provide more information for the maternal effects on the reproduction-related gene expression.

A univariate perspective of multivariate genome-wide association analysis.

Multiple correlated phenotypes are frequently collected in genome-wide association studies (GWASs), and a systematic, simultaneous analysis of multiple phenotypes can integrate the signals from single phenotypes, therefore increasing the power of detecting genetic signals. However, fundamental questions remain open, including the conditions and reasons under which the multivariate analysis is beneficial, how a highly significant signal arises in the multivariate analysis. To understand these issues, we propose to decompose the multivariate model into a series of simple univariate models. This transformation offers a clearer quantitative analysis of the circumstances under which a multivariate approach can be beneficial for the bivariate phenotypes case. A real data analysis is employed to illustrate how to interpret how the signals arising from multivariate GWASs.

Subtype classification and heterogeneous prognosis model construction in precision medicine.

Common diseases including cancer are heterogeneous. It is important to discover disease subtypes and identify both shared and unique risk factors for different disease subtypes. The advent of high-throughput technologies enriches the data to achieve this goal, if necessary statistical methods are developed. Existing methods can accommodate both heterogeneity identification and variable selection under parametric models, but for survival analysis, the commonly used Cox model is semiparametric. Although finite-mixture Cox model has been proposed to address heterogeneity in survival analysis, variable selection has not been incorporated into such semiparametric models. Using regularization regression, we propose a variable selection method for the finite-mixture Cox model and select important, subtype-specific risk factors from high-dimensional predictors. Our estimators have oracle properties with proper choices of penalty parameters under the regularization regression. An expectation-maximization algorithm is developed for numerical calculation. Simulations demonstrate that our proposed method performs well in revealing the heterogeneity and selecting important risk factors for each subtype, and its performance is compared to alternatives with other regularizers. Finally, we apply our method to analyze a gene expression dataset for ovarian cancer DNA repair pathways. Based on our selected risk factors, the prognosis model accounting for heterogeneity consistently improves the prediction for the survival probability in both training and test datasets.