ABSTRACT
Colorectal cancer (CRC) is one of the leading causes of cancer-linked deaths globally. The determination of biomarkers is important in the prognosis and treatment of CRC. Previous studies emphasized the relationship between hypoxia and CRC in humans, and there is strong evidence that this process is strongly related to HIF-1. KDM3A is a histone demethylase that could directly bind to HIF-1α, a subunit of HIF-1. This study aimed to reveal whether the expression level of the KDM3A gene could be used as a predictor of CRC. The expression levels of HIF-1α, KDM3A, and Epithelial-Mesenchymal Transition (EMT) genes were evaluated by qRT-PCR in leukocyte samples of 50 CRC patients in different stages and 50 healthy controls. HIF-1α and KDM3A expression levels were significantly higher in the CRC group, compared to the controls. Slug and ZEB-1 genes, the mesenchymal markers, showed the same significance pattern between groups. We acquired 0.664 AUC with 54% sensitivity and 85.4% specificity for separating controls from CRC patients by using the KDM3A expression levels in ROC analysis. This data support that KDM3A could be a novel supplementary biomarker in diagnosis of CRC, which could be noninvasively detected in circulation.
Keywords: biomarker, colorectal cancer, KDM3A, qRT-PCR
INTRODUCTION
Colorectal cancer (CRC) is a leading cause of morbidity and mortality throughout the world. It is the third most frequent cancer among worldwide and the second most common cause of death (1). Although the high incidence and mortality have increased interest to better understand the pathogenesis of CRC, the molecular mechanism which triggers CRC progression is not clearly identified (2). Also, determination of blood-based biomarkers for diagnosis and treatment in CRC is one of the most studied subjects.
The HIF-1 gene is the key regulator of hypoxic cell response and development of many cancer types. Hypoxia-inducible factor-1 (HIF-1) consists of two subunits: HIF-1α and HIF-1β. Its construction is O2-dependent in the nucleus. Under normoxia with adequate O2 conditions, the HIF-1α subunit is degraded in the proteasome (3). But under hypoxic conditions, this degradation pathway is disrupted, and HIF-1α accumulates in the nucleus (4). The HIF-1α function has been shown to influence particular Jumonji C-domain-containing histone demethylases (JHDM) (5, 6). These histone demethylases form a broad family of enzymes, each of which has a specific ability to influence transcriptional repression on specific histone residues (7). KDM3A regulates the expressions of some pro-angiogenic hypoxia dependent genes by reducing histone methylation in promoters (8-10). Uemura et al. demonstrated the association between KDM3A and CRC in human colon cancer cell lines and tissue samples and this gene was indicated to be a useful biomarker for hypoxic tumor cells and a prognostic marker that could be a therapeutic target against CRC (11). These data suggest a link between tumorigenesis in the colon and KDM3A gene expression.
Early detection of CRC influences the survival rate of patients (12). Colonoscopy is the gold standard for the diagnosis of CRC; however, it is an invasive, time-consuming and expensive method (13). Therefore, it is important to find an easily detectable, and non-invasive blood-based biomarker for the early detection of CRC. We aimed to evaluate whether the expression levels of the KDM3A gene could be useful predictors of colorectal carcinogenesis in circulation.
Epithelial-Mesenchymal Transition (EMT) ensures a mechanism for cancer cells to obtain a more aggressive phenotype which is controlled by several transcription regulators, including E-cadherin, Claudin-1, Slug, and ZEB-1. In the process of EMT, expression levels of epithelial markers such as E-cadherin and Claudin-1 are decreased, whereas those of mesenchymal markers such as Slug, and ZEB-1 are increased (14). We investigated the expression levels E-cadherin, Claudin-1, Slug, and ZEB-1 to determine this association.
In this study, it was determined that KDM3A is related to CRC, and in our opinion, these results are very valuable as this is the first expression study based on human peripheral leukocyte samples that showed this association. Based on this data, KDM3A is an important target in the diagnosis in CRC.
MATERIALS AND METHODS
Study design
The study consisted of two groups; 50 healthy controls without CRC, inflammatory bowel disease, and positive family history, and 50 pathologically confirmed CRC patients in different stages. All blood samples collected from the patient and control groups were taken early in the morning. Furthermore, blood samples were collected from the patient groups at the time of diagnosis before treatment was started. Usage of drugs, hormones, immune suppressors, cytotoxins, or free radical scavengers were exclusion criteria for all groups. Depending on the severity of the CRC, patients were divided into four groups (stage I-IV). The study was approved by the Ethics Committee of SANKO University (2018/05-03). Written informed consent was acquired from all individuals who agreed to participate in the study.
Total RNA Isolation and cDNA Synthesis
PureLink® (Thermo Fisher) RNA Mini Kit was used to extract the total RNA from peripheral blood leukocytes, according to the protocol recommended by the manufacturer. The RNA concentration of each sample was measured, and purities were evaluated by the NanoDrop spectrophotometer. Then, cDNA transcriptions from RNA samples (1µg) were done using High Capacity cDNA Reverse Transcription Kit (Thermo Fisher).
Quantitative Real Time-PCR
The gene expression levels of HIF-1α, KDM3A, E-cadherin, Claudin-1, Slug, and ZEB-1 were determined using quantitative real-time polymerase chain reaction (qRT-PCR) from total leukocyte RNA of peripheral blood samples. StepOnePlus QRT-PCR (Qiagen, Germany) was used for cDNA amplifications. The PCR mixture was composed of SYBR Green PCR Master Mix (Qiagen), 20 pmol of forward/reverse primers, RNase-free water. The cDNAs were constructed in a total volume of 20 μL. β-actin (ACTB), the housekeeping gene, expression level was used as an internal control to evaluate the integrity of each sample. The PCR primers of each gene are described in Table 1. Cycling conditions were 95°C for 15 min, 40 cycles; at 95°C for 15 sec; at 60°C for 1 min, and at 72°C for 30 sec. Data were examined by Rotor-Gene Q Series software v2.1.0 (Qiagen), and expression levels were calculated by using the standard curve method. Each gene was analyzed separately and ran by duplicate. The mean CT threshold values for each sample were used to calculate the relative gene expression levels using the 2−ΔΔCT method expressed as Fold-Change (FC) (15).
Table 1.
The detailed PCR primers of each gene
| Gene | Forward Primer | Reverse Primer |
|---|---|---|
| HIF-1α | 5′-TCCATGTGACCATGAGGAAA-3′ | 5′-CC AAGCAGGTCATAGGTGGT-3′ |
| KDM3A | 5′-GGCGGACTTTAGACGTTCCA-3′ | 5′-AGATGAGCCTTCCACTTGGC-3′ |
| E-cadherin | 5′-CGGGAATGCAGTTGAGGATC-3′ | 5′-AGGATGGTGTAAGCGATGGC-3′ |
| Claudin-1 | 5′-GATAGCAATCTTTGTGGCCACCGT-3′ | 5′-TTCGTACCTGGCATTGACTGGG-3′ |
| Slug | 5′-CATGCCTGTCATACCACAAC-3′ | 5′-GGTGTCAGATGGAGGAGGG-3′ |
| ZEB-1 | 5′- GCCAATAAGCAAACGATTCTG-3′ | 5′- TTTGGCTGGATCACTTTCAAG-3′ |
| ACTB | 5′-GCCGGGACCTGACTGACTAC-3′ | 5′-TTCTCCTTAATGTCACGCACGAT-3′ |
Statistical analysis
Statistical analysis was performed using SPSS software (standard version 22.0; SPSS). The normal distributions of expressions in groups were analyzed using the Shapiro-Wilk method. As descriptive statistics, median values were used when expression levels were not normally distributed, and mean values were used when expression levels were normally distributed. A Non-parametric Mann-Whitney U test was used for non-normally distributed genes and the student-t test was used for normally distributed Claudin-1 gene. Differences between patient subgroups were tested using Kruskal-Wallis with Dunn’s multiple comparison test. The KDM3A diagnostic value was assessed by the ROC curve. P <0.05 was considered statistically significant.
RESULTS
Characteristics of study subjects
Of 50 CRC patients, 22 (44%) were women, and the mean age was 59.98±15.58. In all, 12.2% of patients were diagnosed as stage I, 28.6% were stage II, 20.4% were stage III, and 38.8% were stage IV. And the mean age of the control group was 52.82±12.1, and 35 of them (70%) were women.
Analysis of HIF-1α, KDM3A and EMT marker gene expressions in CRC and control groups
The gene expression levels comparing the groups were shown in Table 2. HIF-1α, KDM3A, Slug, and ZEB-1 expression levels were detected as statistically significant between the CRC and the control groups. HIF-1α expression tended to increase among groups (median -1.32, and 4.25), and the difference was significant (p=0.0045) (Fig. 1A). The expression level of the KDM3A gene was significantly higher in the CRC group compared to the controls (p=0.0049) (Fig. 1B). Furthermore, the expression levels of the Slug and ZEB-1 mesenchymal marker genes was significantly higher in the CRC group compared to the controls (p=0.008 and p=0.029) (Fig. 1C-1D) There was no statistically significant difference between the groups in terms of E-cadherin and Claudin-1 gene expressions (p>0.05).
Table 2.
Comparing the control and CRC groups gene expression levels
| Control | CRC | ||
|---|---|---|---|
| Log2(2-ΔΔCT) | Median [%25-%75] | Median [%25-%75] | p |
| HIF-1α | -1.32 [-5.19- 4.59] | 4.25 [0.72 -7.4] | 0.0045 |
| KDM3A | 0.06 [-3.56-3.12] | 4.29 [-0.74-5.97] | 0.0049 |
| E-cadherin | -1.19 [-4.40-4.03] | 0.46 [-2.89-4.01] | 0.189 |
| Slug | 0.56 [-3.18-2.64] | 2.74 [0.75-3.51] | 0.008 |
| ZEB-1 | 0.65 [-2.75-2.18] | 2.49 [-1.09-3.58] | 0.029 |
| Log2(2-ΔΔCT) | Mean±SD | Mean±SD | p |
| Claudin-1 | 0.00058±3.45 | 0.74±4.03 | 0.42 |
Figure 1.
A-C. Box plot of RT-PCR results. The box border represents the interquartile range, the horizontal line in the box is the median, and circles represent outliers. Values are expressed as log2(2-∆∆CT). Significant differences between control and CRC of HIF-1α (A), KDM3A (B), Slug (C), ZEB-1 (D) genes were shown.
The areas under the receiver operating characteristic (ROC) Curve (AUC), as well as the specificities and sensitivities for the optimal cut-points were computed to determine whether the KDM3A might be a prognostic factor in CRC. ROC analysis produced an AUC of 0.664 for separating CRC patients from controls by using the KDM3A expression level with 54% sensitivity and 85.4% specificity (Fig. 2).
Figure 2.
ROC curve, comparing control and CRC cases using KDM3A expression levels
Gene expressions correlated with clinical features
We next examined the HIF-1α, KDM3A, E-cadherin, Claudin-1, Slug, and ZEB-1 gene expression levels in different stages of CRC patients. No significant difference was observed between these subgroups (p>0.05).
DISCUSSION
Hypoxia is a very common characteristic in solid tumors. While normal cells cannot resist prolonged hypoxia and undergo either apoptosis or necrosis, cancer cells can adapt to hypoxia by changing the expression of genes involved in many cellular processes like proliferation, metabolic reprogramming, and angiogenesis (16). Chronic tumor hypoxia is linked to poor cancer outcomes and activates the transcription HIF-1, which regulates the expression of many cancer-related genes, particularly those involved in angiogenesis (17).
HIF-1α is shown to be related to CRC (18), and experimental in-vitro studies have revealed consistent findings regarding the effect of hypoxia on colorectal cancer. Martinez et al. found that HIF-1α protein increased in HCT116 colorectal cancer cells exposed to O2 changes. This changed gene and protein expressions in pathways regulating hypoxia, glycolysis, and extracellular matrix remodeling (17). In addition, several studies proposed that the expression level of HIF-1α is associated with the prognosis, recurrence, and metastasis of CRC (19). Also, it was reported that high expression levels of HIF-1α in patients with CRC might be a potential biomarker for the progression of CRC. In our study, we demonstrated that HIF-1α expression was significantly increased in leukocytes samples of CRC patients. To the best of our knowledge, this is the first study to show expression changes in peripheral leukocytes.
KDM3A is also one of the hypoxic response genes, and HIF-1α up-regulates its expression. It was shown that loss of KDM3A reduced tumor growth in-vivo, consistent with its role in regulating histone methylation during hypoxia (9). In another study, it was emphasized that KDM3A was mainly overexpressed in human CRC specimens, and knock-down of KDM3A significantly suppressed CRC cell proliferation, colony formation, invasion, migration, and metastasis (2). These studies identify a regulatory mechanism in which the induction of KDM3A by HIF-1 acts as an epigenetic signal enhancer (9). Several types of research have shown that abnormal expression of KDM3A exists in numerous types of cancers (9, 11, 20, 21). In this study, this gene has demonstrated to be a novel prognostic marker and a therapeutic target for CRC. It was detected to be statically significant in CRC groups, and the AUC was calculated as 66.4%. We could not find a study investigating KDM3A in CRC patients from human blood samples in our literature searches.
EMT is a dynamic process in which epithelial cells obtain a mesenchymal phenotype with decreased intercellular adhesion and increased cell mobility. Many transcription factors, including Slug and ZEB-1 have been reported to induce EMT by downregulating E-cadherin and Claudin (22, 23). In our study, E-cadherin and Claudin-1 were not statistically significant between groups. However, the significance determined between the groups at Slug and ZEB-1 expression levels was found to be compatible with KDM3A. These results indicate that mesenchymal markers are more important in CRC.
In conclusion, clinical in-vivo studies have been very limited investigating KDM3A and colorectal carcinogenesis relation. Furthermore, the existing ones are either animal or cell culture studies. Concerning all our data, KDM3A is a novel biomarker in the development of CRC, which deserves more detailed research.
Acknowledgments
Overall funding was provided by a PhD Thesis Research Project Grant (No: FEF.DT.18.17) through the Gaziantep University Scientific Research Projects Coordination Unit.
Footnotes
Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
REFERENCES
- 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries 2021;71(3):209. doi: 10.3322/caac.21660. . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 2.Peng K, Su G, Ji J, Yang X, Miao M, Mo P, et al. Histone demethylase JMJD1A promotes colorectal cancer growth and metastasis by enhancing Wnt/beta-catenin signaling J Biol Chem. 2018;293(27):10606. doi: 10.1074/jbc.RA118.001730. . ; ( ): - . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Toffoli S, Michiels C Intermittent hypoxia is a key regulator of cancer cell and endothelial cell interplay in tumours FEBS J. 2008;275(12):2991. doi: 10.1111/j.1742-4658.2008.06454.x. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 4.Bhaskara VK, Mohanam I, Rao JS, Mohanam S Intermittent hypoxia regulates stem-like characteristics and differentiation of neuroblastoma cells PLoS One. 2012;7(2):e30905. doi: 10.1371/journal.pone.0030905. . . ; ( ): . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pollard PJ, Loenarz C, Mole DR, McDonough MA, Gleadle JM, Schofield CJ, et al. Regulation of Jumonji-domain-containing histone demethylases by hypoxia-inducible factor (HIF)-1alpha Biochem J. 2008;416(3):387. doi: 10.1042/BJ20081238. . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 6.Wellmann S, Bettkober M, Zelmer A, Seeger K, Faigle M, Eltzschig HK, et al. Hypoxia upregulates the histone demethylase JMJD1A via HIF-1 Biochem Biophys Res Commun. 2008;372(4):892. doi: 10.1016/j.bbrc.2008.05.150. . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 7.Shi Y, Whetstine JR Dynamic regulation of histone lysine methylation by demethylases Mol Cell. 2007;25(1):1. doi: 10.1016/j.molcel.2006.12.010. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 8.Beyer S, Kristensen MM, Jensen KS, Johansen JV, Staller P The histone demethylases JMJD1A and JM-JD2B are transcriptional targets of hypoxia-inducible factor HIF J Biol Chem. 2008;283(52):36542. doi: 10.1074/jbc.M804578200. . . ; ( ): - . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Krieg AJ, Rankin EB, Chan D, Razorenova O, Fernandez S, Giaccia AJ Regulation of the histone demethylase JMJD1A by hypoxia-inducible factor 1 alpha enhances hypoxic gene expression and tumor growth Mol Cell Biol. 2010;30(1):344. doi: 10.1128/MCB.00444-09. . . ; ( ): - . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sar A, Ponjevic D, Nguyen M, Box AH, Demetrick DJ Identification and characterization of demethylase JMJD1A as a gene upregulated in the human cellular response to hypoxia Cell Tissue Res. 2009;337(2):223. doi: 10.1007/s00441-009-0805-y. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 11.Uemura M, Yamamoto H, Takemasa I, Mimori K, Hemmi H, Mizushima T, et al. Jumonji domain containing 1A is a novel prognostic marker for colorectal cancer: in vivo identification from hypoxic tumor cells Clin Cancer Res. 2010;16(18):4636. doi: 10.1158/1078-0432.CCR-10-0407. . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 12.Hauptman N, Glavac D Colorectal Cancer Blood-Based Biomarkers Gastroenterol Res Pract. 2017;2017:2195361. doi: 10.1155/2017/2195361. . . ; : . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zygulska AL, Pierzchalski P Novel Diagnostic Biomarkers in Colorectal Cancer. 2022;23(2):852. doi: 10.3390/ijms23020852. . ; ( ): . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Matsumoto R, Tsuda M, Wang L, Maishi N, Abe T, Kimura T, et al. Adaptor protein CRK induces epithelial-mesenchymal transition and metastasis of bladder cancer cells through HGF/c-Met feedback loop Cancer Sci. 2015;106(6):709. doi: 10.1111/cas.12662. . ; ( ): - . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Livak KJ, Schmittgen TD Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method Methods. 2001;25(4):402. doi: 10.1006/meth.2001.1262. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 16.Harris AL Hypoxia--a key regulatory factor in tumour growth Nat Rev Cancer. 2002;2(1):38. doi: 10.1038/nrc704. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 17.Martinez CA, Kerr B, Jin C, Cistulli PA, Cook KM Obstructive Sleep Apnea Activates HIF-1 in a Hypoxia Dose-Dependent Manner in HCT116 Colorectal Carcinoma Cells Int J Mol Sci. 2019;20(2):20220021. doi: 10.3390/ijms20020445. . . ; ( ). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cao D, Hou M, Guan YS, Jiang M, Yang Y, Gou HF Expression of HIF-1alpha and VEGF in colorectal cancer: association with clinical outcomes and prognostic implications BMC Cancer. 2009;9:432. doi: 10.1186/1471-2407-9-432. . . ; : . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhao FL, Qin CF EGF promotes HIF-1alpha expression in colorectal cancer cells and tumor metastasis by regulating phosphorylation of STAT3 Eur Rev Med Pharmacol Sci. 2019;23(3):1055. doi: 10.26355/eurrev_201902_16993. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 20.Suikki HE, Kujala PM, Tammela TL, van Weerden WM, Vessella RL, Visakorpi T Genetic alterations and changes in expression of histone demethylases in prostate cancer Prostate. 2010;70(8):889. doi: 10.1002/pros.21123. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]
- 21.Yang H, Liu Z, Yuan C, Zhao Y, Wang L, Hu J, et al. Elevated JMJD1A is a novel predictor for prognosis and a potential therapeutic target for gastric cancer Int J Clin Exp Pathol. 2015;8(9):11092. . ; ( ): - . [PMC free article] [PubMed] [Google Scholar]
- 22.Merikallio H, T TT, Paakko P, Makitaro R, Kaarteenaho R, Lehtonen S, et al. Slug is associated with poor survival in squamous cell carcinoma of the lung Int J Clin Exp Pathol. 2014;7(9):5846. . ; ( ): - . [PMC free article] [PubMed] [Google Scholar]
- 23.Song IH, Kim KR, Lim S, Kim SH, Sung CO Expression and prognostic significance of epithelial-mesenchymal transition-related markers and phenotype in serous ovarian cancer Pathol Res Pract. 2018;214(10):1564. doi: 10.1016/j.prp.2018.07.016. . . ; ( ): - . [DOI] [PubMed] [Google Scholar]