Expression levels of miR-143-3p and -424-5p in colorectal cancer and their clinical significance

Abstract

BACKGROUND:

Colorectal cancer (CRC) is one of the most prevalent cancers and microRNAs are involved in colorectal carcinogenesis and progression. The role of our candidate microRNAs (miR-143-3p, -424-5p, -212-3p and -34a-3p) have been investigated in various cancers.

OBJECTIVE:

The aim of the current study was to evaluate expression levels of microRNAs (miR-143-3p, -424-5p, -212-3p and -34a-3p) in the sera of patients with CRC in order to identify potential non-invasive biomarker for CRC and investigate the relationship between their expression and clinicopathological features of CRC.

METHODS:

The serological expression of candidate microRNAs were measured in 124 participants, including 62 CRC patients and 62 healthy controls and the serum expression levels of candidate miRNAs were quantified by stemloop reverse transcriptionquantitative polymerase chain reaction.

RESULTS:

In the present study, results showed a significant upregulation expression level of miR-424-5p ( 0.001) and decreased expression level of miR143-3p was observed in the sera of patients with CRC ( 0.001). Receiver operating characteristic (ROC) curve analysis demonstrated the area of miR-424-5p and miR-143-3p under the ROC curve for CRC diagnosis were 0.703 and 0.724 respectively ( 0.001). In addition, down expression of miR-143-3p was significantly associated with tumor size ($P=$ 0.005) and lymph node metastasis ($P=$ 0.020) in CRC patients.

CONCLUSIONS:

The present investigation suggested that low expression of miR-143-3p and increasing level of miR-424-5p in CRC patients may play an important role in development of CRC and they could function as potential non-invasive biomarkers for CRC.

1. Introduction

Colorectal cancer (CRC) is the third most prevalent malignancy and the fourth leading cause of cancer-related death worldwide [1] and a total of 1 million new cases of CRC, 660,000 deaths were expected worldwide each year [2]. The development and progression of CRC is a multi-factorial, multi-step and multi-stage process, which is resulted from the incorporated actions of environmental factors and genetic factors [3]. Identification of novel biomarkers is a need in order to choose appropriate therapy for various patients [4].

MicroRNAs (miRNAs/miRs) are small non-coding single-stranded RNA molecules [5], which can be paired to the coding region [6] or 3’ untranslated region (UTR), the 5’ UTR [7] of target messenger RNAs (mRNAs) which can induce either mRNA degradation or translational repression [8]. MicroRNAs are found in serum, plasma and other body fluids [9, 10], and exist in an apparently stable extracellular form [11]. Exosomes which contain miRNA produced by tumor cells may have a substantial role in metastasis by promoting angiogenesis, cell proliferation and/or tumor cell invasion [11, 12, 13].

Growing evidence demonstrates that miRNAs play an important role in cancer development, differentiation and progression, functioning as tumor oncogenes or suppressors [14, 15, 16, 17].

Studies have represented the different expression of miRNAs in the sera of patients with tumor compared with the ones for healthy people. In addition, recent studies are indicative of the fact that miRNAs, released by cancer cells, circulating in the plasma are potential candidates for non-invasive detection of cancer [18, 19].

The role of candidate microRNAs (miR-143-3p, -424-5p, -212-3p and -34a-3p) were investigated in progression of cancer [20, 21, 22, 23] and some of them were studied as prognostic biomarkers in other malignant neoplasms [24, 25].

Therefore, the present study was performed to evaluate the expression of miR-143-3p, -424-5p, -212-3p and -34a-3p in the sera of patients with CRC and to define the relationships between candidate miRNAs expression and clinicopathological features of CRC.

2. Materials and methods

2.1. Patients and samples

A total of 62 patients pathologically diagnosed with CRC, who had been admitted to Shahid Sadoughi Hospital (Yazd, Iran) between November 2016 and December 2017, were enrolled in this study. Sixty two people undergoing health check-up for the same period made up the control group. The blood samples were collected and utilized after receiving informed consent from all participants. Exclusion criteria included genetic CRC related to genetic syndrome, previous chemotherapy and radiotherapy. Some information included age, gender, tumor size, TNM staging, lymph node status, distant metastasis and perineural invasion were extracted from patient’s medical record. The present study was approved by the ethics committee of Shahid Sadoughi University of Medical Sciences (ID: IR.SSU.MEDICINE.REC.1395.200).

2.2. Serum preparation, RNA extraction and reverse transcription reaction

Venous blood (5 ml) of CRC patients and control subjects were withdrawn before surgery and all blood samples were centrifuged at 3500 g for 5 min within 1 hour after collection. Supernatant was removed and stored in RNase/DNase free tubes at $-$80${}^{\circ }$C for further experiment.

Serum samples were lysed with TRIzol LS reagent (Invitrogen Inc, San Diego, CA, USA). The aqueous phase was isolated after the addition of chloroform and centrifugation. Isopropanol was added to precipitate the RNA from the aqueous phase and ethanol was used to wash RNA pellet and The resulting RNA was dissolved in DEPC treated water. DNase treatment was accomplished in order to eliminate any contaminating DNA (DNase I, Invitrogen, Carlsbad, CA, USA) The concentration and quality of RNA were analyzed by the NanoDrop Spectrophotometer (A260/A280 $=$ 1.8 $-$ 2; A260/A230 $>$ 1.7) and preserved at $-$80${}^{\circ }$C until further use.

In order to detect miRNA, total RNA was reverse transcribed subsequently to cDNA with the stem-loop reverse transcription primer. cDNA synthesis was carried out using the NG dART RT kit (Eurex). The manufacturer’s instructions were observed during all experiments. The relation mixture leading to RT reactions included 5X NG cDNA Buffer, primers, NG dART RT mix, RNase-free Water and RNA (The concentration of RNA used for conversion to cDNA was 700 ng/$\mu$l). The reaction occurred at 50${}^{\circ }$C for 60 min and then 85${}^{\circ }$C for 5 min. The obtained cDNA was preserved at $-$80${}^{\circ }$C.

2.3. RT-qPCR for miRNAs

RT-qPCR was conducted using the ABI StepOne system (Qiagen GmbH) and amplifications were performed using the HyperScript TM First strand synthesis kit (Geneall Biotechnology Co., Korea). Real-time PCR conditions were 95${}^{\circ }$C for 15 minutes, followed by 40 cycles of amplification at 95${}^{\circ }$C for 10 sec, 61${}^{\circ }$C for 30 sec, and 72${}^{\circ }$C for 20 sec, with a final extension at 72${}^{\circ }$C for 5 min. All reactions were conducted in triplicate and during the PCR data were analyzed in StepOne Software v2.3 software. Ct values were used to determine the RNA expression levels. The quantity of target gene expression was normalized using hsa-miR-93-5p as appropriate reference gene for sera of CRC samples [26]. The amplification specificity was examined by melting curve and the RT-qPCR data analysis was accomplished by 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method.

2.4. Statistical analysis

Mann-Whitney U test was employed to analyze the difference in the candidate miRNAs expression levels between the CRC patients and healthy control group. The chi-square test was performed to evaluate the association between candidate miRNAs expression and the clinicopathological parameters of CRC. In order to interpret the ability of miRNA in discriminating patients from healthy controls, a receiver operating characteristic (ROC) curve was plotted based on the DCt (Ct[miRNA]-Ct[internal control]) of miRNA expression in samples. The area under the curve (AUC), specificity and sensitivity at the optimal cut-off were calculated in order to evaluate the diagnostic application of these miRNAs as cancer biomarkers. Statistical analyses were performed using GraphPad Prism 6 (GraphPad Software Inc., CA, USA). All statistical tests were two-sided and a value of less than 0.05 was considered as a statistically significant difference.

3. Results

3.1. Patient characteristics

A total of 62 patients pathologically diagnosed with CRC were enrolled in the present study. Of these patients, 32 were males and 30 females, with the age range of 48–75 years and mean age of (60.7 $\pm$ 7.4) years. Sixty-two people undergoing health check-up for the same period, including 35 males and 27 females with the age of range 47–69 years and the mean age of (58.5 $\pm$ 5.3) years made up control subjects. The post-operative stages of patients were I–II and III–IV in 34 and 28 patients, respectively. Forty patients had lymph node metastasis. Also, 12 and 50 patients had distant and non-distant metastasis respectively and 22 patients were found to have perineural invasion. The statistical analysis indicated that the down expression of miR-143-3p was associated with tumor size and lymph node metastasis in CRC ( 0.05).

3.2. Serum expression levels of miR-143-3p, -424-5p, -212-3p and -34a-3p

The expression of miR-143-3p, -424-5p, -212-3p and -34a-3p in the serum were analyzed using RT-qPCR. The relative expression levels of candidate miRNAs were illustrated in Fig. 1. Compared with the sera of the control group, the expression of miRNA-143-3p in the serum was significantly decreased in the CRC patients ( 0.001), while the levels of miR-424-5p was upregulated ( 0.001, Table 1, Fig. 1). Receiver operating characteristic (ROC) curves were generated to analyze the capability of predicting CRC patients and the area under curves (AUCs) of miR-424-5p and miR-143-3p were 0.703 (95% CI: 0.605–0.801) and 0.724 (95% CI: 0.635–0.813) (Fig. 2). ROC curve analysis of miR-424-5p and -143-3p represented in Table 2.

Figure 1.

Relative expression levels of (A) miR-143-3p, (B) miR-424-5p, (C) miR-212-3p and (D) miR-34a-3p in the sera of control subjects and colorectal cancer patients. The line was drawn to demonstrate the median value and the Mann-Whitney U test was used to calculate statistical significance. miR, microRNA.

Table 1.

The relative expression of miR-424-3p, -143-3p, -212-3p and -34a-3p in sera of colorectal cancer patients in comparison with healthy controls

MicroRNAs	Median in CRC patients	Median in controls	CRC/healthy (fold)	Change	$P$ value
miR-424-5p	4.26	0.91	2.35	Up-regulated	0.001
miR-143-3p	0.32	1.20	$-$2.69	Down-regulated	0.001
miR-212-3p	0.85	1.11	$-$1.09	Down-regulated	$>$ 0.05
miR-34a-3p	1.06	1.37	$-$1.20	Down-regulated	$>$ 0.05

CRC: Colorectal cancer.

Figure 2.

Receiver operating characteristic curve analysis of (A) miR-424-5p and (B) miR-143-3p. miR, microRNA.

Table 2.

Receiver operating characteristic curve analysis

MicroRNAs	Specificity	Sensitivity	AUC	95% confidence
				interval
miR-424-5p	0.726	0.790	0.703	0.605–0.801
miR-143-3p	0.742	0.613	0.724	0.635–0.813

AUC: Area under curve.

In addition, the serum level of miR-212-3p and miR-34a-3p were lower in patients with CRC than healthy control but the results were not statistically significant ($P>$ 0.05).

The relative expression of miR-143-3p, -424-5p, -212-3p and -34a-3p in sera of CRC patients in comparison with healthy controls were listed in Table 1.

3.3. Correlations of miR-143-3p expression with clinicopathological parameters of patients with CRC

To evaluate the correlation of serum miR-143-3p and miR-424-5p expression with clinicopathological parameters, the 62 CRC patients were classified into high and low expression groups using the optimal cut-off calculated by the ROC curve for miR-143-3p and miR-424-5p. The association of miR-143-3p expression with clinicopathological features of patients with CRC was represented in Table 3. Low expression of miR-143-3p was significantly correlated with tumor size and lymph node metastasis in CRC ($P=$ 0.005 and $P=$ 0.020, respectively). Moreover, the association of expression levels of miR-424-5p, -212-3p as well as -34a-3p with clinicopathological characteristics were evaluated and no significant correlation was found.

Table 3.

Correlation of miR-143-3p and -424-5p expression with clinicopathological features in colorectal cancer

Clinicopathological parameters	Case no.	miR-143-3p expression		$P$ value	miR-424-5p expression		$P$ value
		Low	High		Low	High
Age
60	29	18	11	0.706	8	21	0.764
$⩾$ 60	33	22	11		8	25
Gender
Male	32	22	10	0.472	7	25	0.465
Female	30	18	12		9	21
Tumor size
5 cm	33	16	17	0.005${}^{*}$	9	24	0.778
$⩾$ 5 cm	29	24	5		7	22
TNM stage
I–II	34	24	10	0.271	7	27	0.301
III–IV	28	16	12		9	19
Lymph node metastasis
Yes	40	30	10	0.020${}^{*}$	10	30	0.845
No	22	10	12		6	16
Distant metastasis
Yes	12	6	6	0.242	1	11	0.123
No	50	34	16		15	35
Perineural invasion
Yes	22	16	6	0.316	5	17	0.681
No	40	24	16		11	29
Total cases	62	40	22		16	46

TNM, tumor-node-metastasis.

4. Discussion

Colorectal tumorigenesis is a complicated and multi-stage process comprising a set of genetic alterations [27]. Considering dysregulation expression of genes seems vital for having an extensive investigation of initiation and progression of colorectal carcinogenesis [28]. Some tumor-specific expression alterations cause non-invasive biomarkers which is detectable in patients’ serum [29, 30]. MicroRNAs are biomarkers having a prominent role in various cellular processes such as proliferation, invasion, and apoptosis [11, 12, 13]. The function of miRNAs might be different in various tissues or it could be influenced by environmental and population features [31], a thorough understanding of their role in the process of carcinogenesis is a difficult task. Furthermore, different dysregulation expression level of miRNAs may be reported in different cancers. Some miRNAs are promising biomarkers for cancer detection [32]. For instance, it has been illustrated that miR-181a and miR-372 can be utilized as prognostic biomarker for CRC [33, 34]. In the current study expressions alterations of the following four miRNAs, miR-424, -143, -212, miR-34a, and their association with clinicopathological parameters has been examined. The present study represents that the circulating miR-143 downregulated significantly in patients with CRC. In addition, serum miR-143 low expression level was associated with tumor size and lymph node metastasis. The role of miR-143 has been investigated in proliferation, cell migration and invasion of cancer cells and its association with clinicopathological characteristics in various carcinomas [35, 36]. Yang et al. reported that miR-143 regulates proliferation and apoptosis of CRC cells and suggested it could be used as a new target for diagnosis [37]. These results were in agreement with the corresponding findings of Slaby et al. stating that expression level of miR-143 decreased in CRC [38]. Also, the results of the current study showed that miR-424 upregulated significantly in the sera of CRC patients. MiR-424 as the circulating miR has been investigated in age-related macular degeneration [39]. Under hypoxia, mir-424 expression in endothelial cells regulates hypoxia inducible factor 1 (HIF-1$\alpha$) and promotes angiogenesis [40] and HIF-1$\alpha$ adjusts proangiogenic genes expression [41]. Under normoxia condition, the regulation of HIF-1$\alpha$ and HIF-2$\alpha$ in cells, is done using oxygen/VHL-dependent proteasomal degradation [41, 42] and transcription [42, 43]. miR-424 has an effective role in HIF-1$\alpha$ expression adjustment. Under hypoxia conditions, UP.1 transcription factor binds with miR-424 promoter and induces mir-424 expression. miR-424 binds with CUL2 mRNA and leads to decreases of their level. CURL2 reduction results in the instability of VCBCR U3-ligase complex and stability of HIF-1$\alpha$ [40]. Also, the study of Wei et al. showed that Over-expression of miR-424-5p causes gastric cancer development by targeting Smad through TGF-$\beta$ signaling pathway [44].

Several studies investigated dysregulation of miR-212 and miR-34 expression levels in various cancers. Dong et al. reported low expression of miR-212-3p can be used as a prognostic biomarker in colorectal cancer [45] and Nugent et al. found circulating miR-34a level was significantly downregulated in CRC [46]. In the current study, Serum expression downregulation of miR-212 and -34a were observed in patients with CRC but the results were not statistically significant ($P>$ 0.05).

Our study has some limitations. Kirschner et al. represented the impact of hemolysis on alteration of circulating miRNAs content in plasma/serum [47]. Evaluating the level of hemolysis is suggested for future studies, to eliminate this confounding factor in miRNAs analysis in blood derived fluids by considering miR-424 expressed in red blood cells [48]. Also, the selection of internal control is an important step for accurate quantification by RT-PCR. However, there is still no well-recognized internal control to quantify miRNA in serum. In current study, we selected miR-93-5p as the candidate internal control based on a study that evaluated appropriate reference genes in sera of CRC patients [26]. In addition, the present study includes small sample size and the AUC for miR-424-5p and -143-3p were 0.703 and 0.724 ( 0.001) and regarding the fact that using a larger study sample will validate the diagnostic capability of miR-424-5p and -143-3p, a larger sample size is recommended for the following studies.

In conclusion, we have demonstrated that low expression level of miR-143-3p is associated with tumor size and lymph node metastasis in CRC patients. The current investigation indicated that decreased expression of miR-143-3p and increasing level of miR-424-5p in CRC patients may play an important role in progression of CRC and they could act as potential non-invasive biomarkers for CRC.

Conflict of interest

There are no conflicts of interest.