Cadmium is a common toxicant that is detrimental to many tissues. Although a number of transcriptional signatures have been revealed in different tissues after cadmium treatment, the genes involved in the cadmium caused male reproductive toxicity, and the underlying molecular mechanism remains unclear. Here we observed that the mice treated with different amount of cadmium in their rodent chow for six months exhibited reduced serum testosterone. We then performed RNA-seq to comprehensively investigate the mice testicular transcriptome to further elucidate the mechanism. Our results showed that hundreds of genes expression altered significantly in response to cadmium treatment. In particular, we found several transcriptional signatures closely related to the biological processes of regulation of hormone, gamete generation, and sexual reproduction, respectively. The expression of several testosterone synthetic key enzyme genes, such as Star, Cyp11a1, and Cyp17a1, were inhibited by the cadmium exposure. For better understanding of the cadmium-mediated transcriptional regulatory mechanism of the genes, we computationally analyzed the transcription factors binding sites and the mircoRNAs targets of the differentially expressed genes. Our findings suggest that the reproductive toxicity by cadmium exposure is implicated in multiple layers of deregulation of several biological processes and transcriptional regulation in mice.
Cadmium is an environmental and occupational toxic heavy metal that is widely used in industrial process and consumer products. The usual pattern of the nonoccupational cadmium intake is mainly from food, drinking water, and smoking [
It has been well documented that cadmium exposure leads to the impairment of male and female reproductive system both in human and animals. The high level cadmium in the serum and seminal fluid positively correlated to the azoospermia in the infertile Nigerian males [
In this study, we performed the RNA-seq to profile the alterations of gene expression in response to chronic cadmium exposure. By analyzing the transcriptome between the cadmium treated and untreated mice, we identified a number of transcriptional signatures, which provided mechanistic insight into the mechanism of how the male reproductive system is affected by chronic cadmium exposure.
Cadmium chloride (CdCl2) was purchased from Sigma Chemical Co. (St. Louis, MO).
Thirty six-week-old male Chinese Kun Ming (KM) mouse weighing about 30–32 g were used in the experiment. The animals were obtained from Wuhan University Center for Animal Experiments/A3-Lab. All animals were housed in a laboratory-controlled environment (25°C, 50% humidity, and light : dark cycle 12 h : 12 h). The animals were permitted free access to food and drinking water
After acclimatization and one-week observation, we found the daily food consumption per mouse was about 6–8 g. Then all animals were randomly divided equally into three groups and every five mice were housed in a cage. To calculate the consumption of food containing cadmium, we make high-cadmium food as 0.3 mg CdCl2/g and low-cadmium food as 0.15 mg CdCl2/g. Then, each cage of high level cadmium-exposed group was supplied with 50 g high-cadmium food daily, and low level cadmium-exposed group was supplied with 50 g low-cadmium food as well. On the following day, we collected and removed the remaining food and residue and new 50 g food was given. By deducting the weights of remaining food and residue, we calculated that the food intake for cadmium treated mice was
A total of 9 testis samples (three samples from each group) were selected for RNA isolation. Total RNA was isolated using Trizol Reagent (Invitrogen) according to the manufacturer’s instructions. Then these RNA samples were sent to Analytical & Testing Center at Institute of Hydrobiology, Chinese Academy of Sciences (
Raw data were mapped to the mouse reference genome (mm10 downloaded from UCSC) using TopHat (version 2.0.3) and RNASEQR (version 1.0.2) software, respectively [
Aligned reads from TopHat and RNASEQR were assembled by Cufflinks (version 2.0.2), an
In order to determine the differentially expressed transcripts within the dataset, we used Cuffdiff, a separate program included in Cufflinks, to calculate expression in two or more samples and test the statistical significance of each observed change in expression between them. Cuffdiff reports numerous output files containing the results of the differential analysis of the samples, including genes and transcripts expression level changes, familiar statistics such as log2-fold change,
qPCR analyses were performed to validate the results of RNA-seq. The reverse transcription is synthesized using RevertAidTM First Strand cDNA Synthesis Kit from Fermentas according to the manufacturer’s instructions. The PCR primers were designed with Primer Premier 5.0 software and
Primers used for qPCR validation.
Gene | Primer sequence (5′-3′) | Target size (bp) | Tm (°C) |
---|---|---|---|
Actin, beta | Forward: CTGTCGAGTCGCGTCCACCC |
128 |
59 |
|
|||
Cyp11a1 | Forward: AGATCCCTTCCCCTGGTGACAATG |
192 |
60 |
|
|||
Cyp17a1 | Forward: CCAGGACCCAAGTGTGTTCT |
250 | 59 |
|
|||
Prm2 | Forward: CAAGAGGCGTCGGTCA |
167 | 59 |
|
|||
Tex15 | Forward: ATTTGAGTGGCACAGAC |
194 | 59 |
|
|||
Adam9 | Forward: CGCTTAGCAAACTACCTG |
147 | 59 |
|
|||
Dazl | Forward: GGAGGCCAGCACTCAGTCTTC |
184 | 60 |
Besides those statistical tools embedded in the bioinformatics software and resources, additional statistical analyses were performed using GraphPad Prism (Version 5.00). Cadmium treated groups were compared with the control groups by unpaired Student’s
After six-month cadmium-exposure treatment, all animals survived with the slight loss of body weight for the treated groups (Figure
Body weights, level of cadmium in serum and testis, and serum testosterone. Mice were treated with different level of CdCl2 in their rodent chow for six months. (a) Body weights were measured per week during cadmium treatment. The level of accumulated cadmium in serum (b) and testis (c) were measured using the graphite furnace atomic absorption spectrometry (GFAAS) method. The levels of serum testosterone (d) were measured by ELISA. Data was shown as mean ± standard deviation (
To determine the molecular events during cadmium exposure, we performed RNA-seq on testis samples of treated and untreated mice. Raw data of 80 million, 36-bp pair-ends reads were obtained and aligned to the mouse reference genome by TopHat and RNASEQR software [
Testicular transcriptome reconstruction of RNA-seq under mouse reference annotation. (a) The genes on chr11:69,594,778–70,305,335 were reconstructed as examples. (b) The intergenic transcription was detected beyond the reference annotation. (c) Read coverage of Adam24 gene on chr8 was shown.
After reads mapping, transcripts assembling, and expression level calculating, we next sought to identify differentially expressed genes between samples with different treatment. By using Cuffdiff, two expression profiles were obtained from different mapping software. We then compared them and combined the overlapped differentially expressed genes. Genes with
Cadmium modulated differentially expressed genes analysis. (a) The number of genes differentially expressed between the high level cadmium treated, low level cadmium treated, and control groups. (b) Hierarchical clustering analysis of gene expression profiles. Each column represents one mouse, and each horizontal line refers to a gene. Color legend is on the top-left of the figure. Yellow indicates genes with a greater expression relative to the geometrical means; blue indicates genes with a lower expression relative to the geometrical means. (c) Biological process Gene Ontology (GO) analysis of differentially expressed genes.
In order to gain a comprehensive impact assessment of cadmium exposure on testicular gene expression, all biochemical pathways that altered in response to cadmium exposure were identified by comparing the ontology of all the genes differentially expressed between samples. Here, 373 differentially expressed genes annotated by UCSC and Ensembl of the 830 genes were performed with gene ontology enrichment analysis and functional classification. These genes were classified into several ontology categories according to their function in various biological processes (Figure
Notably, the most enriched ontology category contains the genes associated with regulation of transcription. Genes that are involved in many classical signaling transduction pathways are modulated, such as Nfat5, E2f2, Fos, Junb, Notch1, and Stat4. In addition, we observed that abnormal epigenetic regulation occurred during cadmium exposure. Some of the differentially expressed genes involved in DNA methylation and histone modification are those with DNA methyltransferase, histone methyltransferase, acetyltransferase, or deacetylase activities, including Crebbp, Dmap1, Prdm9, Setd2, Prmt7, and Hdac2. Thus, both the transcriptional program and epigenetic patterns are supposed to be misregulated and implicated in cadmium caused reproductive toxicity.
Importantly, we also identified several novel and specific pathways modulated by cadmium exposure, including homeostasis of hormone (Table
Regulation of hormone level related genes.
Gene symbol | Description | Fold change |
|
---|---|---|---|
Adh1 | Alcohol dehydrogenase 1 (class I) | 3.806345 | 0.005 |
Cyp11a1 | Cytochrome P450, family 11, subfamily a, polypeptide 1 | −7.197441 |
|
Cyp17a1 | Cytochrome P450, family 17, subfamily a, polypeptide 1 | −4.438219 | 0.0001 |
Ren1, Ren2 | Renin 1 structural; similar to renin 2 tandem duplication of Ren1; renin 2 tandem duplication of Ren1 | 7.678866 | 0.016 |
Retsat | Retinol saturase (all trans retinol 13, 14 reductase) | 4.603697 | 0.025 |
Star | Steroidogenic acute regulatory protein | −5.377734 |
|
Regulation of reproductive process related genes.
Gene symbol | Description | Fold change |
|
---|---|---|---|
Adam24 | A disintegrin and metallopeptidase domain 24 (testase 1) | −2.872645 | 0.038 |
Adam25 | A disintegrin and metallopeptidase domain 25 (testase 2) | −2.736529 | 0.048 |
Adam26a | A disintegrin and metallopeptidase domain 26A (testase 3) | −3.429332 | 0.023 |
Dazl | Deleted in azoospermia-like | −3.38705 | 0.01 |
Fndc3a | Fibronectin type III domain containing 3A | −2.636543 | 0.044 |
Kitl | Kit ligand | −4.289398 | 0.044 |
Prm2 | Protamine 2 | 1.169034 | 0 |
Qk | Similar to Quaking protein; quaking | −3.974115 | 0.006 |
Sycp2 | Synaptonemal complex protein 2 | −2.977031 | 0.031 |
Tex15 | Testis expressed gene 15 | −2.926283 | 0.0298951 |
Txndc3 | Thioredoxin domain containing 3 (spermatozoa) | −2.973916 | 0.017 |
Zbtb16 | Zinc finger and BTB domain containing 16 | −6.408031 | 0.011 |
Zfp37 | Zinc finger protein 37 | −3.430625 | 0.004 |
Zfx | Zinc finger protein X-linked; similar to zinc finger protein | −4.58343 | 0.027347 |
Zan | Zonadhesin | −3.605458 | 0.00578172 |
Besides, 9 KEGG signaling pathways were affected by cadmium exposure by mapping the differentially expressed genes to the KEGG database. Those modulated signaling pathways were comprised of ribosome, Alzheimer’s disease, asthma, oxidative phosphorylation, focal adhesion, ECM-receptor interaction, C21-Steroid hormone metabolism, metabolic pathways, and prostate cancer (Table
Modulated KEGG pathways.
Pathway name | Number of genes | Corrected |
---|---|---|
Ribosome | 17 |
|
Alzheimer’s disease | 11 | 0.0246 |
Asthma | 2 | 0.0486 |
Oxidative phosphorylation | 9 | 0.0419 |
Focal adhesion | 10 |
|
ECM-receptor interaction | 7 |
|
C21-steroid hormone metabolism | 2 | 0.0106 |
Prostate cancer | 6 | 0.00868 |
Metabolic pathways | 37 | 0.0059 |
To confirm the changes in gene expression observed by RNA-seq, we performed qPCR analysis on three reproductions (Prm2, Tex15, and Dazl), two hormones (Cyp11a1 and Cyp17a1) associated with functional categories genes, and a randomly selected gene named Adam9. qPCR results showed that these genes are significantly differentially expressed (
qPCR validation of the RNA-seq data. log2-fold change determined from the relative Ct values of six genes were compared to those detected by RNA-seq. Replicates (
In an effort to uncover the potential regulatory mechanism underlying the transcription of the cadmium modulated gene sets, we performed transcription factor binding sites analysis within the promoters and microRNA targets analysis of the cadmium modulated genes. Promoter regions for positions of −1000–+200 of the TSS across the cadmium modulated genes were predicted for the binding sites enrichment of several transcription factors (
Enrichment of transcription factors across the promoter regions of differentially expressed genes.
Transcriptional factors | Number of genes | Corrected |
---|---|---|
ETF | 193 |
|
NKX3A | 178 | 0.007 |
Nrf-1 | 170 |
|
HMGIY | 140 | 0.009 |
SRY | 328 |
|
ZF5 | 199 |
|
FOXJ2 | 153 |
|
OCT-1 | 229 |
|
E2F-1 | 226 |
|
LUN-1 | 35 | 0.043 |
FOXP1 | 319 |
|
AP2 | 169 | 0.014 |
Transcriptional regulation analysis of the differentially expressed genes. (a) Biological process gene ontology analysis of the transcription factors that regulate the gene expression. (b) MicroRNA-target gene network. Red circles represent microRNAs; green circles represent the target genes.
We next performed microRNA targets analysis of the differentially expressed genes for further investigating the posttranscriptional control of them. A total of 10 microRNAs were identified as significantly enriched at 3′-UTR region of the differentially expressed genes (
MicroRNAs enriched at 3′-UTR region of the differentially expressed genes.
MicroRNA | Number of targeted genes | FDR |
---|---|---|
Mir-142-3p | 6 | 0.001 |
Mir-342/342-3p | 5 | 0.0015 |
Mir-196ab | 5 | 0.0025 |
Mir-874 | 4 | 0.003 |
Mir-124/506 | 5 | 0.0135 |
Mir-30a | 3 | 0.0075 |
Mir-124/506 | 11 | 0.0205 |
Mir-153 | 4 | 0.0265 |
Mir-25/32/92/92ab/363/367 | 3 | 0.0195 |
Mir-448 | 4 | 0.006 |
Cadmium has been suggested to be anenvironmental and occupational toxic heavy metal that causes several diseases and toxically targets the lung, the liver, the kidney, the bone, the cardiovascular system, the immune system, and the reproductive system [
We next used RNA-seq to analyze the transcriptome of mouse testis affected by cadmium. We found a total of 830 genes and transcripts that were differentially expressed. Gene Ontology analysis revealed that these genes were enriched in several biological processes, in which the genes related to the reproductive process were paid more attention. For example, Fndc3a was reported to be required for adhesion between spermatids and Sertoli cells during mouse spermatogenesis [
Further, we computationally analyzed the transcriptional and posttranscriptional control of the differentially expressed genes. We found several transcriptional factors were enriched with the binding sites at the promoter regions of some gene sets. These binding events should be verified by further ChIP experiments. While these transcriptional factors were unable to be detected as statistically significantly differentially expressed between the samples, it is likely that the slight change of expression ultimately led to the significant expression change of their targets. We also predicted the microRNAs with the binding possibility of some sets of cadmium modulated genes. We identified 10 microRNAs targeted to the differentially expressed genes, the regulatory roles of which in testis response to cadmium could be explored by their expression patterns and the gain- or loss-of-function studies in the future.
In summary, our study demonstrated that many genes in testis were modulated due to chronic cadmium exposure. In particular, aside from the genes related to the functional categories previously reported, we identified novel pathways and the potential transcriptional regulatory mechanism on the cadmium modulated genes. These findings provide evidence for the elucidation of the molecular mechanism linking the chronic cadmium exposure to the impairment of male reproductive system and the clues for future studies of potential biomarkers and therapeutic targets for cadmium exposure.
The authors declare no conflict of interests.
The authors thank Jie Li at the Department of Pharmacology, Basic Medical School of Wuhan University for technical help. They also thank all members in Dr. Zhong’s laboratory for helpful suggestions and technical assistance. This work was financially supported by the National Natural Science Foundation of China (Grant no. 31172395) and Key Technologies Research and Development Program of China (Grant no. 2013BAI12B01-3).