Background
Renal cell carcinoma (RCC) is the most prevalent malignancy of the kidney, and it accounts for 2.4 % of all adult malignancies [
1]. Clear cell renal cell carcinoma (ccRCC) represents the predominant histologic subtype of RCC and constitutes approximately 80-90 % of all cases [
1,
2]. Surgery is the most effective treatment of early and local ccRCCs, but after the resection for local disease, 20–40 % patients will develop recurrence [
3], mainly due to the tumour’s high resistance to both chemotherapy and radiotherapy [
2,
4]. Therefore, it is of paramount importance to understand the molecular mechanisms underlying the tumourigenesis of ccRCC. The identification of novel genes that are functionally involved in the initiation and progression of ccRCC may provide more sophisticated early diagnostic and further therapeutic strategies.
The human
chemokine-like factor (
CKLF)
-like MARVEL transmembrane domain-containing family (
CMTM) is a gene family consisting of nine members,
CKLF and
CMTM1-8 [
5,
6]. Their encoded products are structurally and functionally intermediate between classical chemokines and the transmembrane-4 superfamily (TM4SF), playing important roles in the immune system [
7‐
11], the male reproductive system [
12‐
14] and tumourigenesis [
15‐
25]. Several members, such as
CMTM3,
5,
7 and
8, have been reported to exhibit tumour suppressor functions in many types of malignancies, including gastric, pancreatic, liver, lung, cervical, oral, ovarian and oesophageal cancers [
15‐
25].
CMTM4 is the most conserved member of this family and forms a gene cluster with
CKLF and
CMTM1-3 on chromosome 16q22.1, a locus that is frequently deleted or modified in multiple tumours and that harbours a number of tumour suppressor genes [
26‐
33].
CMTM4 encodes three transcript variants, CMTM4-v1, −v2 and -v3. Among them, CMTM4-v2 is the full length cDNA product and is highly conserved in most vertebrate animals [
34]. In HeLa cells, knockdown of CMTM4 can lead to cell cleavage defects and binucleated cells after mitosis [
35], while overexpression of CMTM4-v1 and -v2 can inhibit cell growth by causing G2/M phase arrest without inducing apoptosis [
34]. These findings suggest that
CMTM4 might be an important gene involved in cell growth and cell cycle regulation. However, the function of CMTM4 in tumourigenesis remains poorly defined. In this study, we analysed the expression pattern of CMTM4 using a bioinformatics strategy and focused on its expression and function in ccRCC.
Materials and methods
All of the array data related to cancers from the Affymetrix human genome U133 plus 2.0 platform were downloaded from the GEO database (
http://www.ncbi.nlm.nih.gov/geo/), and a TumourProfile database (
http://tumour.bjmu.edu.cn/, unpublished) has been developed to analyse the differentially expressed genes in tumours using previously described data processing and microarray analysis methods [
36,
37]. The expression profile of CMTM4 in a variety of cancers and the corresponding control (normal or non-tumour) tissues was searched in this database, and the expression levels were represented as average rank scores (ARS). Rank-based gene expression (RBE) curves, which visually reflected the gene expression profile (GEP) across multiple tissues, were generated using the TumourProfile data set.
Patient samples
A total of 61 patients with ccRCC (aged 22 to 78 years, median age of 60 years) who underwent surgery between January 2013 and April 2014 at the Department of Urology, Peking University People’s Hospital (Beijing, China) were enrolled in the present study. Paired tumour and adjacent non-tumour tissues were collected and tested for CMTM4 expression. All of the specimens were pathologically confirmed. The paraffin-embedded blocks of tumour tissues from each patient were assembled from the archival collections at the Department of Pathology. All participants gave informed consent according to the Helsinki Declaration, and the protocol for the present study was approved by the Ethics Committee of Peking University People’s Hospital (Beijing, China).
Cell lines, adenovirus and siRNAs
The ccRCC cell lines A498 and 786-O and the normal renal tubular epithelial cell line HK-2 were routinely cultured in MEM (Invitrogen, Carlsbad, CA, USA), RPMI-1640 (HyClone, Logan, UT), and K-SFM medium (Gibco™ Life Technologies, Grand Island, NY) containing 10 % FBS (HyClone) supplemented with 1 % penicillin/streptomycin, respectively. All cells were grown at 37 °C in a humidified incubator containing 5 % CO2. Adenoviruses carrying the CMTM4 gene (Ad-CMTM4) and the empty adenovirus (Ad-null) were packaged by AGTC Gene Technology Company, Ltd. (Beijing, China). The 786-O cells were infected with the adenoviruses at an MOI of 100. Small interfering RNAs (siRNAs) targeting CMTM4 were designed and chemically synthesised by GenePharma Co., Ltd. (Suzhou, China). The following sequences were used: si-CMTM4-3, 5′-GAAAUUGCUGCCGUGAUAUTT-3′ (sense), 5′-AUAUCACGGCAGCAAUUUCTT-3′ (antisense); si-CMTM4-6, 5′-GCAUAUGCAGUGAACACAUTT-3′ (sense), 5′-AUGUGUUCACUGCAUAUGCTT-3′ (antisense); and negative control (si-NC), 5′-UUCUCCGAACGUGUCACGUTT-3′ (sense), 5′-ACGUGACACGUUCGGAGAATT-3′ (antisense). 786-O cells were transfected with the siRNAs using Lipofectamine™ 3000 (Life Technologies, Grand Island, NY) according to the manufacturer’s instructions.
Protein extraction and western blotting
The cells were lysed in RIPA buffer (Sigma-Aldrich, St. Louis, MO, USA) supplemented with a 1 % protease inhibitor cocktail (Roche, Basel, Switzerland). The protein concentrations were determined using BCA protein assays (Pierce, Rockford, IL, USA). The whole cell lysates were then fractionated using 12.5 % or 15 % SDS–PAGE gels and electrotransferred onto polyvinylidene difluoride membranes (Hybond; GE Healthcare, Buckinghamshire, United Kingdom). Western blotting was performed as previously described [
18]. The rabbit anti-CMTM4 pAb was prepared in our lab [
38]. The anti-cyclin B1, −cyclin E, −cyclin-D1, −p21 and -p27 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). β-actin blotting was used as a lysate loading control. The density of the bands was analysed by ImageJ software (National Institutes of Health, Bethesda, Maryland, U.S.). The absolute intensity of the target protein was normalised to the absolute intensity of β-actin.
PCR and qPCR
The total RNAs were isolated from ccRCC tissues and cell lines using TRIzol reagent (Invitrogen). Reverse transcription was performed according to standard protocols using a RevertAid™ II First Strand cDNA synthesis Kit (Thermo Fisher Scientific Inc., Waltham, MA USA). Semiquantitative and quantitative PCR (qPCR) were performed as previously described [
18]. GAPDH was amplified as an internal standard. The primers for PCR of CMTM4 were as follows: CMTM4V2-F: 5′-CAGAAATTGCTGCCGTGAT-3′, CMTM4V2-R: 5′-TGACTGAGAGACAGGCACG-3′, and the 72# probe (Roche) was used for qRT-PCR of CMTM4. The primers for PCR of p21 were p21-F: 5′-CTCAGAGGAGGCGCCATGTC-3′ and P21-R: 5′-TTAGGGCTTCCTCTTGGAGAAG-3′.
Immunohistochemistry (IHC)
Immunohistochemical analysis was performed on formalin-fixed, paraffin-embedded clinical tissues as previously described [
18]. A rabbit anti-CMTM4 pAb (4 mg/L) was used as the primary antibody.
Cell proliferation assay
Cell proliferation was analysed using the Cell Counting Kit-8 (CCK-8, Dojindo Laboratories, Kumamoto, Japan) and viable cell counting assays. For the CCK-8 assays, the cells were seeded in 96-well plates at a density of 3000 cells per well and then incubated at 37 °C in a 5 % CO2 humidified atmosphere. At the indicated time points, 10 μL CCK-8 solution was added into each well and incubated for 2 h. The absorbance at 450 nm was measured to assess the number of viable cells. The results were obtained from three independent experiments in triplicate. For the viable cell counting assays, the cells were seeded in 24-well plates at a density of 20,000 cells per well. The viable cells marked by trypan blue exclusion were counted using a Vi-CELL TM_XR Cell Viability Analyzer (Beckman Coulter, Inc., Brea, CA, USA).
Flow cytometry
Cellular apoptosis was evaluated by FITC-conjugated Annexin V/propidium iodide (PI) staining followed by flow cytometry analysis, as previously described [
18]. For the cell cycle analysis, the cells were harvested 48 h after infection with adenoviruses or transfection with siRNAs. After washing with PBS, the cells were fixed in ice-cold 70 % ethanol overnight at −20 °C. The fixed cells were then pelleted by centrifugation, washed twice in PBS, and incubated in PBS containing 500 mg/mL RNase A (Sigma-Aldrich) at 37 °C for 30 min. After staining with 10 mg/mL PI (Sigma-Aldrich) in 0.1 % Triton X-100, the cells were collected on a BD FACSCalibur (BD Bioscience, San Jose, CA, USA). The cell cycle distribution was analysed with the ModFit LT software (Verity Software House, Topsham, ME).
Wound healing assay
The 786-O cells infected with Ad-CMTM4 or Ad-null were cultured in 24-well plates until confluent. The cell layer was then scratched using a sterile 10 μL micropipette tip and washed twice with and subsequently maintained in serum-free media. The cells were photographed 0, 24 and 48 h after wounding.
Cell migration assay
Forty-eight hours after infection or transfection, the 786-O cells were serum-starved for 6 h. Then, 3 × 104 cells in 250 μL serum-free media were seeded into the upper chamber of a transwell with a fibronectin-coated filter (8-mm pore size, Corning Life Sciences, NY, USA). The bottom chamber contained medium supplemented with 10 % FBS. After a 14-h (for the siRNA-transfected cells) or 16-h (for the adenovirus-infected cells) incubation at 37 °C in a 5 % CO2 humidified atmosphere, the nonmigrated cells were scraped off of the filter using a cotton swab and the migrated cells were stained with crystal violet following fixation with 4 % paraformaldehyde. The number of cells was counted in 8 randomly chosen fields (magnification, ×200). Triplicate wells were performed in each assay, and the assay was repeated at least three times.
Xenograft model in nude mice
All protocols for the animal studies were reviewed and approved by the institutional Animal Research Ethics Board. Female BALB/c nude mice (4–6 weeks old, weighing 18–22 g) were maintained in a germ-free environment in the animal facility. The tumourigenesis assay was performed as previously described, with some modifications [
39]. Briefly, 5 × 10
6 Ad-CMTM4- or Ad-null-infected 786-O cells in 100 μL PBS were injected subcutaneously into the right and left flanks of nude mice, respectively. The tumour diameter was measured with a calliper every 3 days, and the tumour volume was calculated by length × width
2 × 0.5. The mice were sacrificed at day 27, when the tumours were dissected, weighed and lysed for western blotting analysis.
Statistical analysis
The bioinformatics analysis of the differences in CMTM4 expression between the cancers and control tissues were evaluated using the Wilcoxon rank-sum test in the R (
http://www.r-project.org/) software environment. Bonferroni’s correction of the R function “p.adjust” was used to adjust the
P-values. The experimental data were analysed using SPSS software 17.0 (SPSS, Inc., Chicago, IL, USA). CMTM4 expression was correlated with the clinical characteristics using one-way ANOVA (for the classification variables, such as gender, stage and grade) or Pearson’s correlation analysis for two variables (for the continuous variables, such as age). The differences between two independent groups were analysed using Student’s t test. A
P-value < 0.05 was considered to represent a statistically significant difference.
Discussion
The tumour suppressor functions of members of the CMTM family, particularly CMTM3, 5, 7 and 8, have been extensively studied in multiple types of malignancies. In contrast, CMTM4 remains less investigated. A comprehensive analysis of CMTM4 expression across multiple cancers using bioinformatics indicated that CMTM4 is most significantly downregulated in brain cancers and ccRCC, which implies a tissue-specific function of CMTM4. Currently, omic data analysis has become a major trend in numerous fields, among which gene expression profile (GEP) analysis is generally an essential step in functional gene studies. Analyses using other databases, as well as Delic S. and colleagues’ [
25] and our experimental data, demonstrate the viability of our analysis method [
37] in GEP predictions, with high efficiency and accuracy.
Using a total of 61 paired ccRCC tissues and adjacent normal tissues, we show that CMTM4 expression is frequently downregulated in renal cancer tissues. However, the expression levels of CMTM4 were not correlated with the patients’ gender and age. Because surgical resection is restricted to early and local ccRCCs, most patients are at stage one and histologically exhibit high and moderate differentiation (grade I and II). Therefore, this correlation was not available due to the limitation of the clinical samples. Moreover, the survival data are still being collected, because most of the patients had undergone surgical resection only a short time ago.
CMTM4 is tightly linked with
CMTM1-3 on chromosome 16q22.1, a genomic region prone to both genetic and epigenetic modifications in various cancers. Chromosomal aberrations, such as deletions, amplifications [
26‐
29], single nucleotide polymorphisms (SNPs) [
30], loss of heterozygosity (LOH) and microsatellite instability (MSI) [
31,
32], as well as aberrant methylations [
29], occur frequently in this region in different types of malignancies. Our previous studies have also shown that
CMTM3 is frequently inactivated by promoter CpG methylation [
18]. It remains to be clarified whether these mechanisms are also involved in the downregulation of CMTM4 in ccRCC.
Regular cell cycle progression is a key factor in cell proliferation, and alterations of the cell cycle may influence cell growth. CMTM4 has been suggested to be an important regulator of cell cycle progression and division in HeLa cells [
34,
35]. Here, we also observed that overexpression of CMTM4 inhibited 786-O cell growth by inducing G2/M phase accumulation. p21 was increased in the process, which plays complex roles in tumourigenesis by regulating the cell cycle, senescence, apoptosis and migration [
41]. Through its interaction with the Cdk1/CyclinB complex, the p21 protein interferes with the transition of cells from the G2 phase of the cell cycle into mitosis; moreover, by inhibiting the Rho cascade, p21 can also influence cytoskeletal factors and cell motility [
41]. Therefore, the upregulation of p21 may be responsible for the tumour suppressor functions of CMTM4 in 786-O cells. However, increased p21 expression is not necessarily linked to growth arrest; thus, the sophisticated mechanism underlying the inhibitory activities of CMTM4 is still to be explored. On the other hand, p21 is well known to be induced by p53. In addition, several p53-independent pathways have also been identified [
42]. Overexpression of CMTM4 increased p21 not only at the protein level but also at the mRNA level, whereas knockdown of CMTM4 decreased both. However, because p53 is inactive in 786-O cells [
43], the mechanism by which CMTM4 regulates p21 and whether it influences the transcription or the degradation of the p21 mRNA requires further investigation.
Conclusions
In summary, CMTM4 is frequently reduced in ccRCC tissues and cell lines, according to omic data analysis as well as our experimental data. Restoration of CMTM4 suppresses the tumourigenicity of 786-O cells both in vitro and ex vivo, whereas knockdown of CMTM4 led to promoting effects. These observations highlight the potential of CMTM4 as a tumour suppressor in ccRCC. A better understanding of the roles of CMTM4 in tumourigenesis may allow researchers to develop novel diagnostics and more effective treatment strategies for this malignancy.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
WH and TX designed the project, supervised the study and revised the manuscript. TL performed the western blotting, RT-PCR, IHC, cell growth and migration assays, flow cytometry, and animal experiments and drafted the manuscript. YC participated in the western blotting and IHC assays. PW performed the bioinformatics analysis and related statistical analysis and revised the manuscript. WW assisted in the animal experiments and statistical analyses. FH contributed to sample handling, storage and collection of the clinical data. XM performed the quantitative PCR. HL assisted in the semiquantitative PCR and plasmid construction. All authors read and approved the final manuscript.