Bioinformatics analysis of key biomarkers for cryptorchidism and potential risk of carcinogenesis

doi:10.21203/rs.3.rs-1994766/v1

Download PDF

Article

Bioinformatics analysis of key biomarkers for cryptorchidism and potential risk of carcinogenesis

https://doi.org/10.21203/rs.3.rs-1994766/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Cryptorchidism is characterized by undescended or incomplete descent of the testes. The pathogenesis of cryptorchidism has not been fully elucidated. In addition, patients with cryptorchidism are at a higher risk of malignancy than normal individuals, and its etiology and underlying molecular mechanisms need to be further investigated.

Methods

Datasets GSE16191 and GSE25518 were downloaded from the Gene Expression Omnibus database to identify the common differentially expressed genes (DEGs). Gene enrichment analyses were performed using the WebGestalt database. A protein-protein interaction network of DEGs was constructed using the STRING database, of which hub genes were identified by Cytoscape software. The GEPIA database was used to validate the expression of the hub genes of cryptorchidism in testicular cancer, and then the protein level of genes was detected in the HPA database. The analysis of immune cell infiltration was conducted in the R package. The clinical significance of the selected genes was analyzed from four aspects: clinical correlation, overall survival (OS), recurrence-free survival (RFS), and receiver operating characteristic (ROC) curve.

Results

Of the 438 common DEGs identified, 134 were up-regulated and 304 were down-regulated. Biological functions analysis identified important signaling pathways, key functional modules, and co-expression networks in cryptorchidism. Nine hub genes (HNRNPM, SF1, U2SURP, SNRPA1, AQR, RBM39, PCBP2, RBM5, and HNRNPU) were identified in cryptorchidism, four (SF1, HNRNPM, RBM5, and AQR) of which were significantly expressed in testicular cancer. The high expression of the genes SF1 and HNRNPM predicted poor RFS in cancer patients. Moreover, genes AQR and HNRNPM may contribute to malignant transformation from cryptorchidism to cancer via the spliceosome pathway.

Conclusion

Our study revealed the potential molecular mechanisms under the pathogenesis of cryptorchidism and its carcinogenesis. The biomarkers identified in this study may provide a theoretical basis and new ideas for further mechanism research of cryptorchidism.

Cryptorchidism

Testicular Germ Cell Tumor

Gene Expression Omnibus

Differentially expressed genes

Bioinformatics

Cryptorchidism, including undescended testis (UDT), testicular hypoplasia, and ectopic testis, is the most common congenital malformation of the genitourinary system and one of the most common male endocrine gland diseases in children(1, 2). The incidence of cryptorchidism at the age of 1 year in full-term male infants is 1-4.6%. Research evidence shows that the occurrence of cryptorchidism is closely related to multiple factors such as endocrine regulation, genetics, environmental factors, and anatomical basis. In recent years, a large part of the research on cryptorchidism involves the relationship between causative/susceptibility genes and cryptorchidism, but the exact pathogenesis remains largely unknown. Moreover, it has been recognized that cryptorchidism is the most common risk factor for testicular germ cell cancer (TGCT), with the risk of malignant transformation of cryptorchidism being about 20–40 times higher than that of normal people. However, the mechanism of the malignant transformation of cryptorchidism into TGCT remains unclear and is generally thought to be related to factors such as abnormal spermatogenic cells, elevated testicular temperature, and gonadal dysplasia, which requires more in-depth research.

The pathogenesis of a disease is the result of the interaction between multiple biomolecules. The evolution of computer science and bioinformatics, as well as the accumulation and development of big data, provides a powerful means to study the complex process of disease(3). The Gene Expression Omnibus (GEO) database is an international public repository created and maintained by the National Institutes of Health (NIH), which stores large numbers of experimental data including genomic data provided by different research teams around the world(4). This paper aims to find out the important biomarkers of cryptorchidism through bioinformatics, and identify the biomolecules in the pathogenesis of cryptorchidism that may be closely related to the malignant transformation of cryptorchidism into TGCT, hoping that the results could provide theoretical clues for clarification of the pathogenesis of the disease. We present the following article in accordance with the STREGA reporting checklist.

Data acquisition

"Cryptorchidism" was used as the search term to find the published gene chip datasets GSE16191 and GSE25518 in the GEO database of the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/geo/). The GSE16191(5) dataset contains a total of 20 testicular tissue samples derived from 16 cases of cryptorchidism testis and 4 cases of control. The GSE25518(6) dataset contains a total of 23 testicular tissue samples including 19 cases of cryptorchidism testis and 4 cases of control. The control group was from the contralateral descending testicular tissue of patients with testicular hypoplasia. Both datasets were processed by Affymetrix Human Genome U133 Plus 2.0 Array platform.

Data processing and screening of differentially expressed genes

GEO2R (https://www.ncbi.nlm.nih.gov/geo/geo2r/), as an interactive web tool, is used to compare two or more groups of samples in the GEO database through the GEOquery and Limma R packages to identify differentially expressed genes (DEGs) under different experimental conditions. Analysis on DEGs was performed in both datasets using the online tool GEO2R, with the filtering conditions of |log2FC| >1 and adjusted P-value<0.05. After obtaining the two sets of DEGs, the “dplyr” package of R language was used to process them separately and obtain their common DEGs. The screening conditions included: 1) deletion of genes without specific significance, such as LOC (non-coding genes), MIR (small RNA) and blank probes; 2) removal of duplicate genes: if a gene was highly or lowly expressed in both sets of DEGs, only one was retained; 3) if a gene had both high expression and low expression in the two groups of DEGs, it would be deleted; and 4) genes were screened according to FDR ≤0.05 and FC≥2. The DEGs of the two datasets were presented as volcano plots using the SangerBox tools (http://www.sangerbox.com/tool) and heatmaps in R, and the common DEGs were displayed in the form of a Venn diagram using the R language.

Enrichment analysis of DEGs

The WebGestalt (http://www.webgestalt.org/option.php) database(7), as a functional enrichment analysis web tool, was used to conduct Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis on the screened common DEGs, combined with the “dplyr” and “ggplot2” package of R language. P-value＜0.05 was considered statistically significant.

PPI network construction and module analysis

The STRING11.0(8) (https://string-db.org) is a database that searches for protein-protein interactions (PPI). The common DEGs were imported into the STRING11.0 database to construct a protein-to-protein interaction (PPI) network which was visualized with Cytoscape 3.9.1 software. The minimum required interaction score was set to medium confidence (0.4). Subsequently, key functional modules in the PPI network were screened using the plugin Molecular Complex Detection (MCODE) in Cytoscape, with the advanced options: degree cutoff=2, k-core=2, node score cutoff=0.2, and max depth=100. Then the KEGG analysis was performed by the KOBAS 3.0 (http://kobas.cbi.pku.edu.cn/) database(9) on genes of the key modules, with P-value <0.05 viewed as statistically significant.

Identification of hub genes

The plugin cytoHubba of Cytoscape was used to screen hub genes in the PPI network based on four algorithms (MCC, Degree, EPC, and MNC). Then the online website jvenn(10) (http://jvenn.toulouse.inra.fr/app/index.html) was used to draw the Venn diagram of the hub genes obtained by the four algorithms. Then KEGG pathway analysis was performed on the hub genes using KOBAS 3.0. We put the obtained hub genes back into the original datasets for expression validation to ensure the stability of the results. The comparison between the experimental and the control groups was performed using the Mann-Whitney U test (Wilcoxon rank sum test). P-value less than 0.05 was considered significant.

Subsequently, co-expression analysis of hub genes was conducted using the GeneMANIA (http://genemania.org/) database(11), which is an online tool for exploring the interactions and functions between genes. The database OmicsBean (http://www.omicsbean.cn/) was used to identify the upstream miRNAs of the hub genes and construct the miRNA-mRNA network.

Expression analysis of hub genes on the GEPIA

Gene expression profiling interactive analysis (GEPIA)(http://gepia.cancer-pku.cn/)(12) is an online database for normal and cancer gene expression profiling and interactive analysis, which is based on the Cancer Genome Atlas (TCGA) (https://portal.gdc.cancer.gov/)(13) and Genotype-Tissue Expression (GTEx)( https://commonfund.nih.gov/GTex)(14). The hub genes were imported into the GEPIA to obtain their expression levels in TGCT and other cancer types. After screening of the hub genes that were significantly expressed in TGCT (P-value < 0.05, FC ≥ 1.5), gene correlation analysis was performed using the Spearman rank correlation test based on RNAseq data from the TCGA database, and then the GO and KEGG pathway analysis was conducted using KOBAS 3.0.

HPA database analysis

The Human Protein Altas (HPA) (https://www.proteinatlas.org/) is an open-access database based on proteomics, transcriptomics, and systems biology data to map the human proteins in cells, tissues, and organs. It currently contains more than 26,000 antibodies, all of which are immunohistochemically stained and confirmed by professionals. We used this database to map the expression profile of hub genes in TCGT and normal tissues from two aspects of pathology and subcell.

Analysis of immune cell infiltration

Based on the RNAseq data of TGCT in the TCGA database, the ssGSEA algorithm in the R package "GSVA" was used to analyze the correlation between immune cell infiltration and the expression levels of hub genes. The analysis involved 24 common immune cells, including activated DC(aDC), B cells, CD8 T cells, cytotoxic cells, dendritic cells (DCs), eosinophils, immature DCs(iDCs), macrophages, mast cells, neutrophils, natural killer(NK) cells, NK CD56+ cells, NK CD56- cells, plasmacytoid DCs (pDCs), T cells, T helper cells, T central memory(Tcm), T effector memory(Tem), T follicular helper(Tfh), T gamma delta(Tgd), type 1 Th(Th1) cells, type 17 Th(Th17) cells, type 2 Th(Th2) cells, and regulatory T cells(Treg).

Clinical significance of hub genes in TGCT

Analysis of the clinical correlation between hub genes and TGCT was performed using R language with pathologic T stage as a variable and presented as a violin plot, with P-value<0.05 considered significant. Next, we used the Kaplan-Meier plotter(https://www.kmplot.com/analysis/) database to map the overall survival (OS) and recurrence-free survival (RFS) curves of the hub genes in TGCT, which was constructed based on gene microarrays and RNAseq data from multiple databases and can assess the correlation between gene expression and survival in more than 25,000 samples from 21 tumor types. The hazard ratio (HR) and 95% confidence intervals (CI) were also calculated. Subsequently, the receiver operating characteristic (ROC) curve was drawn using the R package "pROC" to evaluate the diagnostic value of the genes, and the area under the curve (AUC) ≥ 0.7 indicated good discriminatory performance.

Ethics approval

We ensure that all research reported in submitted papers has been conducted in an ethical and responsible manner, and was performed in accordance with relevant guidelines and regulations. In addition, this article is based on public online databases for data analysis, with the datasets publicly available.

Identification of DEGs

After processing of the microarray data, 672 DEGs were screened from GSE16191 dataset, including 189 up-regulated genes and 483 down-regulated genes (Figure 1A, B), and 498 DEGs were screened from GSE25518 dataset, including 176 up-regulated genes and 322 down-regulated genes (Figure 1C, D). The DEGs of the two datasets were intersected using the “VennDiagram” package of R language, and the total number of overlapping genes was 438, of which 134 were co-upregulated and 304 were co-downregulated (Figure 1E).

Functional enrichment analysis of common DEGs

The common DEGs were analyzed for GO enrichment in three aspects: biological process (BP), cellular component (CC), and molecular function (MF). The results showed that the common DEGs were mainly enriched in the insulin-like growth factor receptor signaling pathway (ontology: BP), PcG protein complex (ontology: CC), and insulin receptor binding (ontology: MF), as shown in Figure 2 (A, B, C). KEGG pathway analysis revealed that the common DEGs were mainly enriched in axon guidance, spliceosome, nicotinate and nicotinamide metabolism, and mTOR signaling pathway (Figure 2D). Tables 1 and 2 show the enrichment results ranked in the top 10 by P-value.

PPI network construction and module analysis

Based on the STRING database, the PPI network of the common DEGs was mapped, consisting of 299 nodes and 600 edges (Figure 3A), from which 11 functional modules were screened, and the top two scores are shown in Figure 3(B, D). Further KEGG pathway analysis showed that genes in module 1 were mainly enriched in spliceosome and ferroptosis, while genes in module 2 were mainly enriched in IL-17 signaling pathway, progesterone-mediated oocyte maturation, cell cycle, oocyte meiosis, and ubiquitin mediated proteolysis (Figure 3C, E).

Identification and analysis of hub genes

Four algorithms (MCC, Degree, EPC, and MNC) were used in the PPI network to screen the top 20 ranked hub genes (Table 3), and then the Venn diagram of the hub genes obtained by the four algorithms was drawn, which led to nine common genes HNRNPM, SF1, U2SURP, SNRPA1, AQR, RBM39, PCBP2, RBM5, and HNRNPU (Figure 4A). KEGG enrichment analysis showed that the nine hub genes were mainly enriched in the spliceosome (Figure 4B), and the details are shown in Table 4. To verify the reliability of the expression levels of the hub genes, we put the hub genes into the original datasets for validation, and the results showed that the expression levels of hub genes in the two datasets were relatively consistent (Figure 4C, D).

To further explore the interactions and functions between genes, we constructed a co-expression network of hub genes using the GeneMANIA database, as shown in Figure 4E. These hub genes presented a complex PPI network with physical interactions of 41.98%, co-expression of 33.46%, pathway of 18.88%, co-localization of 2.99%, and shared protein domains of 2.57%, with most gene functions were closely related to the spliceosome. In addition, an analysis of the miRNA-mRNA regulatory network was performed by the OmicsBean database (Figure 4F). The results showed that hsa-miR-495 regulated the largest number of target genes and could be regarded as the key miRNA in the network; RBM5 was regulated by the largest number of miRNAs, which was considered as the key gene in the network.

Expression analysis of hub genes in TGCT

Given the hypothesis that the hub genes in the pathogenesis of cryptorchidism may be associated with TGCT, we verified the expression of hub genes in TGCT using the GEPIA database (Figure 5A). It was found that SF1, RBM5, HNRNPM, and AQR genes were significantly expressed in TGCT, with SF1 and RBM5 significantly down-regulated in tumor tissues and HNRNPM and AQR significantly up-regulated in tumor tissues. Figure 5B shows the expression levels of four hub genes in multiple tumor types, with HNRNPM and AQR being the most prominent in TGCT. Subsequently, the Spearman rank correlation test was used to perform correlation analysis on SF1, RBM5, HNRNPM, and AQR, with P-value less than 0.05 being statistically significant (Figure 5C, D). Correlation analysis indicated that most of the four genes were significantly correlated, and the Spearman correlation coefficient between the genes SF1 and HNRNPM was the highest (0.68). GO and KEGG enrichment analysis of these four genes (SF1, RBM5, HNRNPM, and AQR) showed that GO analysis was mainly enriched in spliceosomal complex, mRNA binding, mRNA splicing, and RNA binding (Figure 5E); KEGG analysis was mainly enriched in Spliceosome (Figure 5F). The enrichment results are detailed in Table 5 and Table 6.

The protein expression levels of SF1, RBM5, HNRNPM and AQR

To further verify the expression of SF1, RBM5, HNRNPM and AQR in TGCT, we reviewed the protein expression levels of these four genes in normal testis tissues and TGCT tissues using the HPA database (Figure 6A). According to the results of immunohistochemical (IHC) staining, SF1 and RBM5 were down-regulated in tumor tissues, while HNRNPM was up-regulated. The relevant IHC results of AQR were not included in the HPA database, so the relevant images are not shown. Meanwhile, we explored the localization information of the four genes in human cells at the subcellular level. Figure 6B shows the immunofluorescence (IF) profiles of the four genes, and all four genes were found to be localized in the nucleoplasm.

The association of SF1, RBM5, HNRNPM and AQR expression with immune cell infiltration

Based on the gene expression data of TGCT patients, we utilized the ssGSEA algorithm to explore the characteristics between the infiltration levels of 24 immune cells and the expression levels of four genes, as shown in Figure 7. Correlation coefficients with absolute values greater than 0.3 were considered as highly correlated. The results showed that both genes SF1 and HNRNPM were mainly negatively correlated with immune cell infiltration (Figure 7A, C). The gene RBM5 had both positive and negative correlations with 24 immune cells, among which 3 immune cells were highly positively correlated, including T helper cells, Tem, and Tcm; 6 immune cells were highly negatively correlated, including DCs, neutrophils, macrophages, Th1 cells, iDCs, and NK CD56+ cells (Figure 7B). The gene AQR was also positively and negatively correlated with 24 immune cells, among which 2 immune cells were highly positively correlated, including T helper cells and Tcm, and 6 immune cells were highly negatively correlated, including Tgd, DCs, pDCs, mast cells, macrophages, iDCs, NK CD56+ cells, and NK cells (Figure 7D).

Clinical significance of genes SF1, RBM5, HNRNPM and AQR in TGCT

First, the clinical correlations between the four genes and TGCT were analyzed using RNAseq data and clinical data of TGCT (Figure 8A). The results indicated that at the pathologic T stage, the genes SF1, HNRNPM, and AQR had no significant clinical correlation with TGCT, while the expression level of gene RBM5 at the T1 stage was significantly higher than that in T2 and T3 stages in TGCT. Secondly, we further validated the prognostic value of these four genes in terms of both OS and RFS in the Kaplan-Meier plotter. Survival analysis demonstrated that all four genes had no significance in OS (Figure 8B), while high expression of the genes SF1 and HNRNPM predicted a shorter RFS in TGCT patients (Figure 8C). The ROC curve analysis was conducted on the four genes (Figure 8D), and the results showed that all four genes had good diagnostic values (AUC>0.7), especially SF1 and RBM5 (AUC>0.9).

Epidemiological evidence shows that the prevalence of cryptorchidism has increased significantly in recent years, but the detailed mechanism has not been completely elucidated. In addition, the most worrying complication of cryptorchidism is malignant transformation, the risk of which is significantly higher than that in normal people. It has been suggested that it may be the gradual malignant transformation of immature gonocytes into precancerous cells, forming intraductal germ cell tumors, and then developing into TGCT after puberty (15). Nonetheless, the underlying causes for the malignant transformation of cryptorchidism remain a major difficulty in the field. The present study was intended to use bioinformatics technology to further explore the molecular mechanism of cryptorchidism and its malignant transformation into TGCT by reviewing the microdata.

Through the analysis of datasets GSE16191 and GSE25518, we identified a total of 438 common DEGs associated with cryptorchidism. Further GO enrichment analysis showed that these common DEGs were mainly reflected in the insulin-like growth factor receptor signaling pathway, PcG protein complex, and insulin receptor binding. These biomolecular processes play an important role in the development and growth of the reproductive system. Existing studies have shown that insulin-like growth factor receptor-1 (IGFR-1) exists abundantly in the cremaster muscle of normal males, and its binding to IGF-1, IGF-2 and insulin in humans contributes to the remodeling and development of the cremaster muscle, which has the potential to promote the descent of the testis(16). Epigenetics is a hot topic of molecular genetics research in recent years, and it also plays a crucial role in the development of the reproductive system(17, 18). As an epigenetic regulatory factor, PcG protein has the function of maintaining the transcriptional inhibition of related target genes(19), but its relationship with cryptorchidism is uncertain. Meanwhile, KEGG pathway analysis indicated that these common DEGs were mainly enriched in the axon guidance, spliceosome, nicotinate and nicotinamide metabolism, and mTOR signaling pathway. It has been documented that mutation in genes that mediate the embryonic axon guidance and branching in the nervous system can lead to pituitary stalk interruption syndrome (PSIS) with phenotypes including cryptorchidism(20), which gives us new insights. The mTOR signaling pathway can maintain the spermatogenesis in the spermatocyte stage of the pachytene by regulating the distribution of gap junction α-1 protein in sertoli cells, and it is also an important signaling pathway in cryptorchidism(21, 22). To the best of our knowledge, there is no study reporting the correlation between other pathways and cryptorchidism, and the significance of abnormal expression of related genes in cryptorchidism remains to be verified.

Of 438 overlapping DEGs, 9 hub genes were identified, including HNRNPM, SF1, U2SURP, SNRPA1, AQR, RBM39, PCBP2, RBM5, and HNRNPU. KEGG enrichment analysis manifested that these 9 genes were significantly enriched in spliceosome. The co-expression network also indicated that most gene functions were associated with the spliceosome. The spliceosome is a ribonucleoprotein complex composed of small nuclear ribonucleoprotein(snRNP) complexes (U1, U2, U4, U5, and U6) and multiple proteins, which can mediate the processing of precursor messenger RNA (pre-mRNA) into mRNA. Hao Wu et al (23) demonstrated that Small Nuclear Ribonucleoprotein Polypeptide A' (SNRPA1), a component of the spliceosome, played a crucial role in male fertility, and that mutations in the gene SNRPA1 inhibited the assembly of the spliceosome and then affected the differentiation of spermatogonia, which in turn would induce non-obstructive azoospermia (NOA). Therefore, it is reasonable to assume that the down-regulation of SNRPA1 expression is closely related to the occurrence of azoospermia in patients with cryptorchidism. Heterogeneous Nuclear Ribonucleoprotein U (HNRNPU), as a potent regulator of nuclear ribonucleoprotein particles, has been noted to be related to purified spliceosomes. Yichun Zheng et al [24] demonstrated that the gene HNRNPU was significantly down-regulated in testicular tissues of patients with spermatogenesis, and was involved in the pathological process of spermatogenic dysfunction. The gene HNRNPU is likely to be involved in the process of spermatogenesis disorders in patients with cryptorchidism. Moreover, existing studies have shown that the effects of environmental pressure can cause lasting epigenetic changes in individuals, which can be transmitted over multiple generations(24). Dennis et al [26] reported that Heterogeneous Nuclear Ribonucleoprotein M (HNRNPM) family proteins were able to show a certain degree of resistance to environmental stress, and had a preventive effect against paternal-mediated intergenerational infertility. The abnormal expression of HNRNPM may affect the epigenetic mechanism of infertility in patients with cryptorchidism. As a key pre-mRNA splicing regulator, RNA Binding Motif Protein 5 (RBM5) plays an important role in spermatogenesis and is necessary for sperm differentiation and male fertility(25). The analysis in our study demonstrated that the gene RBM5 was considered a key gene in the miRNA-mRNA network. We speculated that the down-regulation of this gene may greatly affect the spermatogenic dysfunction in patients with cryptorchidism. The association of other genes (SF1, U2SURP, AQR, RBM39, and PCBP2) with cryptorchidism has not been clearly reported in the literature, but these genes are significantly down-expressed in patients with cryptorchidism and are widely involved in the PPI network, suggesting that they may become important new biomarkers in the future.

According to the validation of the GEPIA database, we identified that 4 genes (SF1, HNRNPM, RBM5, and AQR) were significantly expressed in TGCT from 9 hub genes in cryptorchidism. The GO and KEGG enrichment analysis of these four genes was mainly associated with the spliceosome. Splicing Factor 1 (SF1), as an RNA-binding protein, is involved in the assembly of the early spliceosome during pre-mRNA splicing, which is found in Sertoli cells and germ cells of the seminiferous tubules. Matin et al [28] found that the SF1 expression level was correlated with germ cell tumor development. SF1 deficiency inhibited testicular tumorigenesis of TGCT. This partly explains that the high expression of SF1 is mainly negatively correlated with immune cell infiltration and poor prognosis in terms of RFS. Like the gene SF1, HNRNPM is also an important splicing factor. The expression of the gene HNRNPM is associated with the proliferation, invasion, and metastasis of tumor cells. Several studies have shown that high expression of HNRNPM promotes the development of cancer (such as breast cancer, liver cancer, or colon cancer)(26–28). Although no studies have been reported on the correlation between HNRNPM and TGCT, based on the negative correlation between HNRNPM and immune cell infiltration and its prediction of shortened RFS in TGCT patients, it is reasonable to assume that high expression of HNRNPM is also likely to promote the proliferation and development of TGCT. RBM5 is an RNA-binding protein that can modulate apoptosis by regulating the alternative splicing of the genes involved in the control of apoptosis. It has been documented that RBM5 is a tumor suppressor gene, of which the expression is down-regulated in multiple cancers (such as lung cancer, breast cancer, bladder cancer, or prostate cancer)(29–32). No study has reported the correlation of the RBM5 expression with TGCT. However, through the clinical correlation analysis in our study (the RBM5 expression at the T1 stage was significantly higher than that in T2 and T3 stages in TGCT) and the ROC curve showing good diagnostic value, we consider that RBM5 may be also a putative tumor suppressor in TGCT. Aquarius Intron-Binding Spliceosomal Factor (AQR) is involved in pre-mRNA splicing as a component of the spliceosome. The role of AQR in tumor development is currently unclear, and no relevant literature has pointed out its relationship with TGCT. An important paralog of the gene AQR is HELZ2, which can induce tumorigenesis of endometrial cancer and retinoblastoma(33, 34). The role of AQR in TGCT remains to be proven in future. Notably, the expression of SF1 and RBM5 was down-regulated in TGCT, consistent with their expression in cryptorchidism, whereas HNRNPM and AQR were significantly up-regulated in TGCT patients. Meanwhile, KEGG enrichment analysis showed that AQR and HNRNPM were enriched in the spliceosome. Given that the high expression of AQR and HNRNPM both play a role in promoting tumorigenesis, it is reasonable to speculate that AQR and HNRNPM may induce malignant transformation from cryptorchidism to TGCT through the spliceosome pathway.

In summary, we identified nine hub genes that were involved in the pathogenesis of cryptorchidism, of which four genes may mediate the malignant transformation from cryptorchidism to TGCT. This study provides new ideas for further study of the molecular mechanism of cryptorchidism and its relationship with TGCT. Nevertheless, it should be noted that in the two datasets of cryptorchidism, the controls were the contralateral descending normal testicular tissues from patients with testicular hypoplasia, and not from men with normal testes bilaterally, which may miss the comparison of DEGs between patients and normal males. The main reason for the current lack of relevant datasets may be that it is difficult to obtain the testicular tissue from normal males, especially normal adolescents. Moreover, the results of this study are based on bioinformatic analysis, and further experimental validation and clinical practice are necessary. We look forward to more studies and larger sample sizes in the future to expand and refine our conclusions.

Data availability statement

The datasets generated and/or analysed during the current study are available in the Gene Expression Omnibus (GEO) database under the numbers GSE16191 and GSE25518.

Author contributions

Conception and design: XZ, XX
Administrative support: PL, XZ
Provision of study materials or patients: PL, LL

Collection and assembly of data: XX, PL, LL

Data analysis and interpretation: XX, LL

Manuscript writing: XX

Final approval of manuscript: All authors

Acknowledgements

Funding: This work was supported by the Innovation Program for Chongqing’s Overseas Returnees (CX2019146), High-level Medical Reserved Personnel Training Project of Chongqing (the 4th batch), and Research Program of Basic Science and Frontier Technology in Chongqing (cstc2017jcyjAX0435).

Competing Interests Statement

The authors have no conflicts of interest to declare.

Kim JK, Chua ME, Ming JM, Lee MJ, Kesavan A, Kahn N, et al. Variability among Canadian pediatric surgeons and pediatric urologists in the management of cryptorchidism in boys before the publication of major guidelines: a retrospective review of a single tertiary centre. Can J Surg. 2019;62(3):1-6.
Kolon TF, Herndon CDA, Baker LA, Baskin LS, Baxter CG, Cheng EY, et al. Evaluation and treatment of cryptorchidism: AUA guideline. J Urol. 2014;192(2):337-45.
Petryszak R, Burdett T, Fiorelli B, Fonseca NA, Gonzalez-Porta M, Hastings E, et al. Expression Atlas update--a database of gene and transcript expression from microarray- and sequencing-based functional genomics experiments. Nucleic Acids Res. 2014;42(Database issue):D926-D32.
Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 2013;41(Database issue):D991-D5.
Hadziselimovic F, Hadziselimovic NO, Demougin P, Krey G, Hoecht B, Oakeley EJ. EGR4 is a master gene responsible for fertility in cryptorchidism. Sex Dev. 2009;3(5):253-63.
Hadziselimovic F, Hadziselimovic NO, Demougin P, Oakeley EJ. Testicular gene expression in cryptorchid boys at risk of azoospermia. Sex Dev. 2011;5(2):49-59.
Liao Y, Wang J, Jaehnig EJ, Shi Z, Zhang B. WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs. Nucleic Acids Res. 2019;47(W1):W199-W205.
Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607-D13.
Bu D, Luo H, Huo P, Wang Z, Zhang S, He Z, et al. KOBAS-i: intelligent prioritization and exploratory visualization of biological functions for gene enrichment analysis. Nucleic Acids Res. 2021;49(W1):W317-W25.
Bardou P, Mariette J, Escudié F, Djemiel C, Klopp C. jvenn: an interactive Venn diagram viewer. BMC Bioinformatics. 2014;15:293.
Franz M, Rodriguez H, Lopes C, Zuberi K, Montojo J, Bader GD, et al. GeneMANIA update 2018. Nucleic Acids Res. 2018;46(W1):W60-W4.
Li C, Tang Z, Zhang W, Ye Z, Liu F. GEPIA2021: integrating multiple deconvolution-based analysis into GEPIA. Nucleic Acids Res. 2021;49(W1):W242-W6.
Liu J, Lichtenberg T, Hoadley KA, Poisson LM, Lazar AJ, Cherniack AD, et al. An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell. 2018;173(2).
Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science. 2015;348(6235):648-60.
Batool A, Karimi N, Wu X-N, Chen S-R, Liu Y-X. Testicular germ cell tumor: a comprehensive review. Cell Mol Life Sci. 2019;76(9):1713-27.
Özdamar MY, Şahin S, Zengin K, Seçkin S, Gürdal M. Detection of insulin-like growth factor receptor-1 in the human cremaster muscle and its role in the etiology of the undescended testis. Asian J Surg. 2019;42(1):290-6.
Zhang X-L, Wu J, Wang J, Shen T, Li H, Lu J, et al. Integrative epigenomic analysis reveals unique epigenetic signatures involved in unipotency of mouse female germline stem cells. Genome Biol. 2016;17(1):162.
Shi Y, Qi W, Xu Q, Wang Z, Cao X, Zhou L, et al. The role of epigenetics in the reproductive toxicity of environmental endocrine disruptors. Environ Mol Mutagen. 2021;62(1):78-88.
Ghotbi E, Ye P, Ervin T, Kum A, Benes J, Jones RS. Polycomb-group recruitment to a target gene is the default state that is inhibited by a transcriptional activator. Sci Adv. 2021;7(29).
Bashamboo A, Bignon-Topalovic J, Moussi N, McElreavey K, Brauner R. Mutations in the Human ROBO1 Gene in Pituitary Stalk Interruption Syndrome. The Journal of Clinical Endocrinology and Metabolism. 2017;102(7):2401-6.
Zhou Y, Zhang D, Liu B, Hu D, Shen L, Long C, et al. Bioinformatic identification of key genes and molecular pathways in the spermatogenic process of cryptorchidism. Genes Dis. 2019;6(4):431-40.
Boyer A, Girard M, Thimmanahalli DS, Levasseur A, Céleste C, Paquet M, et al. mTOR Regulates Gap Junction Alpha-1 Protein Trafficking in Sertoli Cells and Is Required for the Maintenance of Spermatogenesis in Mice. Biol Reprod. 2016;95(1):13.
Wu H, Sun L, Wen Y, Liu Y, Yu J, Mao F, et al. Major spliceosome defects cause male infertility and are associated with nonobstructive azoospermia in humans. Proc Natl Acad Sci U S A. 2016;113(15):4134-9.
Nestler EJ. Transgenerational Epigenetic Contributions to Stress Responses: Fact or Fiction? PLoS Biol. 2016;14(3):e1002426.
O'Bryan MK, Clark BJ, McLaughlin EA, D'Sylva RJ, O'Donnell L, Wilce JA, et al. RBM5 is a male germ cell splicing factor and is required for spermatid differentiation and male fertility. PLoS Genet. 2013;9(7):e1003628.
Sun H, Liu T, Zhu D, Dong X, Liu F, Liang X, et al. HnRNPM and CD44s expression affects tumor aggressiveness and predicts poor prognosis in breast cancer with axillary lymph node metastases. Genes Chromosomes Cancer. 2017;56(8):598-607.
Zhu G-Q, Wang Y, Wang B, Liu W-R, Dong S-S, Chen E-B, et al. Targeting HNRNPM Inhibits Cancer Stemness and Enhances Antitumor Immunity in Wnt-activated Hepatocellular Carcinoma. Cell Mol Gastroenterol Hepatol. 2022;13(5):1413-47.
Chen T-M, Lai M-C, Li Y-H, Chan Y-L, Wu C-H, Wang Y-M, et al. hnRNPM induces translation switch under hypoxia to promote colon cancer development. EBioMedicine. 2019;41:299-309.
Sutherland LC, Wang K, Robinson AG. RBM5 as a putative tumor suppressor gene for lung cancer. J Thorac Oncol. 2010;5(3):294-8.
Zhao L, Li R, Shao C, Li P, Liu J, Wang K. 3p21.3 tumor suppressor gene RBM5 inhibits growth of human prostate cancer PC-3 cells through apoptosis. World J Surg Oncol. 2012;10:247.
Rintala-Maki ND, Abrasonis V, Burd M, Sutherland LC. Genetic instability of RBM5/LUCA-15/H37 in MCF-7 breast carcinoma sublines may affect susceptibility to apoptosis. Cell Biochem Funct. 2004;22(5):307-13.
Zhang Y-P, Liu K-L, Wang Y-X, Yang Z, Han Z-W, Lu B-S, et al. Down-regulated RBM5 inhibits bladder cancer cell apoptosis by initiating an miR-432-5p/β-catenin feedback loop. FASEB J. 2019;33(10):10973-85.
Dai H, Zeng W, Zeng W, Yan M, Jiang P, Li Y, et al. HELZ2 promotes K63-linked polyubiquitination of c-Myc to induce retinoblastoma tumorigenesis. Med Oncol. 2021;39(1):11.
Qiao Z, Jiang Y, Wang L, Wang L, Jiang J, Zhang J. Mutations in , , , , , and Are Associated With the Prognosis in Endometrial Cancer. Front Genet. 2019;10:909.

Table 1 to 6 is available in the Supplementary Files section.

No competing interests reported.

table16.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Bioinformatics analysis of key biomarkers for cryptorchidism and potential risk of carcinogenesis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials And Methods

Results

The protein expression levels of SF1, RBM5, HNRNPM and AQR

The association of SF1, RBM5, HNRNPM and AQR expression with immune cell infiltration

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1