This article is mentioned in:

Introduction

Colorectal cancer (CRC) is the third most common cancer in males, the second in females and the fourth leading cause of cancer-related death worldwide. In 2008, more than 1.2 million new CRC cases and 608,700 deaths were caused by CRC; notably, the highest incidence rates are found in The Occident, especially in males (1). However, CRC incidence rates are rapidly increasing in the Asian countries (2).

Among all the risk factors that may cause CRC directly, molecular and genic effects are the dominant causes. Since there are multiple unknown carcinogens and varying genetic backgrounds, it is difficult to determine which factor is the most important during the development of CRC (3). Moreover, lack of efficient early diagnostic biomarkers, such as an actual molecule involved in the progression of CRC, urges path-breaking studies on these aspects.

Increasing evidence shows that non-coding genes may be strictly accountable for the gene expression complexity in humans (4–6). Non-coding RNAs (ncRNAs) are proven to be the key regulators in the process of transcription and expression, which could be divided into small ncRNAs (<200 nt) and long ncRNAs (lncRNAs) (7,8). An increasing number of studies have focused on the functional hot spot of lncRNAs in their roles as regulators of biological processes, such as genomic imprinting, chromatin modification and post-transcriptional processing (9–11), displaying more complex regulatory mechanisms than microRNAs (12,13). In cis- and trans-regulatory mechanisms (14,15), lncRNAs have been shown to play important roles in various human diseases, especially in malignant tumors (16–19).

Prostate cancer non-coding RNA 1 (PRNCR1), also known as PCAT8 and CARLo-3, is a ~13 kb intronless lncRNA, which is transcribed from the 'gene-desert' region of 8q24 (20). It has been reported that PRNCR1 is associated with prostate cancer susceptibility and PRNCR1 could be involved in prostate carcinogenesis by modulating androgen receptor (AR) activity. This mechanism was further described by Yang et al. Binding of PRNCR1 to the acetylated AR and its association with DOT1L appear to be required for recruitment of a second lncRNA, PCGEM1, to the DOT1L-mediated methylated of AR at the N-terminus. The interactions of these overexpressed lncRNAs may potentially serve as important regulators in prostate cancer (21). For CRC, many researchers have reported that the crucial locus of 8q24 may contribute to susceptibility to CRC (22,23). It has been gradually recognized that aberration of PRNCR1 might be a biological signature of CRC, but its specific expression pattern related to CRC remains unknown.

In this study, we identified the PRNCR1 expression profile in CRC patients and assessed the association between PRNCR1 expression and clinicopathological features. By using antisense oligonucleotide (ASO)-mediated inhibition, we evaluated the impact of PRNCR1 on cancer cell proliferation, apoptosis, migration and invasion in vitro.

Materials and methods

Patients and tissue samples

This study was approved by the Ethics Committee of the Cancer Institute of Jiangsu, and written informed consent was obtained from all patients. A total of 63 pairs of primary CRC tissues and adjacent normal tissues (5 cm or more from tumor tissues) were collected from patients who had undergone surgery at the Colorectal Cancer Center, Cancer Institute of Jiangsu, between 2013 and 2014. Patients who received chemotherapy or radiotherapy before surgery were excluded. All tumor specimens were collected immediately after removal from the resected colorectum, frozen and stored at −80°C. All tumors and paired normal tissues were ascertained by experienced pathologists. The clinical and pathological characteristics for each patient were also collected.

RNA extraction and qRT-PCR analyses

Total RNA was extracted from tissues or cultured cells with TRIzol reagent (Life Technologies, Gaithersburg, Scotland, UK) according to the manufacturer's protocol. Total RNA (1.5 µg) was reverse transcribed in a final volume of 20 µl using random primers under standard conditions using the PrimeScript R™ Master Mix (cat. no. RR036A; Takara Bio, Inc., Dalian, China).

After the RT reaction, qRT-PCR was performed using the SYBR Select Master Mix (cat. no. 4472908; Applied Biosystems, Foster City, CA, USA) with 0.5 µl complementary DNA (cDNA) on an ABI 7300 system (Applied Biosystems) according to the manufacturer's instructions. By using β-actin as an internal control, the PRNCR1 expression level was determined by qRT-PCR using the following primer sequences: forward, 5′-CCAGATTCCAAGGGCTGATA-3′ and reverse, 5′-GATGTTTGGAGGCATCTGGT-3′. The forward primer sequences for the β-actin primer were 5′-CCAGATTCCAAGGGCTGATA-3′ and reverse for β-actin were 5′-GATGTTTGGAGGCATCTGGT-3′. The qRT-PCR reaction included an initial denaturation step at 95°C for 10 min, followed by 40 cycles of 92°C for 15 sec and 60°C for 1 min. The Ct value for each sample was calculated with the ΔΔCt method, and fold-changes in expression (tumor vs. normal) were calculated using 2−ΔΔCt methods (24).

Cell culture and ASO transfection

Three CRC cell lines, SW620, HCT116 and SW480, were obtained from the Shanghai Institutes for Biological Science, Shanghai, China. LoVo and HT-29 cells were donated by Dr Zhicheng Chen (Department of Anorectal Clinic, The Medical School of Southeast University). A normal human colorectal epithelial cell line (FHC) was purchased from ScienCell Research Laboratories.

All cell lines were grown in RPMI-1640 medium (Kaiji, Nanjing, China) at 37°C in a humidified 5% CO2 atmosphere. The CRC cell lines at 50% confluency were transfected with 100 nM of either the ASO targeting PRNCR1 or scrambled negative controls (GenePharma, Shanghai, China) using Lipofectamine RNAimax reagent (Invitrogen Inc., Carlsbad, CA, USA) according to the instructions provided by the manufacturer. The ASO sequences were as follows: ASO-1 for PRNCR1, 5′-ACUCUCCTTCTCCACCUCCA-3′; ASO-2 for PRNCR1, 5′-ACUCCCACACCACCACCACC-3′ and scrambled ASO, 5′-AAGCGCGCACCAGCGCCUCC-3′, which were designed by a professional website (http://www.idtdna.com/Scitools/Applications/AntiSense/Antisense.aspx?source=menu).

Cell proliferation assay

Cell proliferation was assayed by the Cell Counting Kit-8 (CCK-8) assay (Promega, Madison, WI, USA) according to the manufacturer's protocol. The transfected cells were plated in 96-well plates (2,000 cells/well). Following the manufacturer's protocol, cell proliferation was detected every 24 h. In brief, 10 µl of CCK-8 solution was added to each well and incubation was carried out for 2 h at 37°C. Then, each solution was measured spectrophotometrically at 450 nm.

In vitro cell migration and invasion assays

HT-29 cells transfected with 100 nM ASO-PRNCR1 or scramble were harvested after 24 h. For the migration assays, the transfected cells (2.5×105) were plated in the upper chamber of Transwell assay inserts (Millipore, Billerica, MA, USA) containing 200 µl of serum-free DMEM with a membrane (8-mm pores). Then the inserts were placed into the wells of the bottom chamber of a 24-well plate filled with conditioned medium. After 24 h of incubation, the cells on the filter surface were fixed with methanol, stained with crystal violet, and photographed with a digital microscope. Cell numbers were calculated in five random fields for each chamber.

For the invasion assays, the transfected cells (4×105) were plated in the top chamber with a Matrigel-coated membrane (BD Biosciences) in 500 µl serum-free RPMI-1640 accompanied by 750 µl 10% FBS-RPMI-1640 in the bottom chamber. After a 48-h incubation period, the invasive ability was assessed as mentioned previously for the migration assay.

Flow cytometric analysis

Transfected cells were harvested after transfection. HT-29 cells were stained with Annexin V and propidium iodide (PI) using Annexin V-FITC/PI apoptosis detection kits (BD Biosciences) and then examined by flow cytometry (FACScan; BD Biosciences). Cells were discriminated into viable cells, early apoptotic cells, apoptotic cells and dead cells. Cells for cell cycle analysis were stained with PI by the Cycletest™ Plus DNA Reagent kit (BD Biosciences) and analyzed by FACScan.

Statistical analysis

The Student's t-test, Spearman's test, and one-way AVONA, the receiver operating characteristic (ROC) curve, binary correlation analysis and logistic regression were performed to analyze the data. The ROC curve was calculated to estimate the diagnostic efficiency of PNRCR1, and the cut-off value of best diagnostic efficiency was also determined. All statistical analyses were performed with SPSS software package (version 19.0; SPSS Inc.). All P-values were two-sided, and P<0.05 was considered to indicate a statistically significant difference.

Results

Correlation between PRNCR1 expression and clinical characteristics

The PRNCR1 expression levels were analyzed in 63 paired primary CRC and adjacent non-cancerous tissues. It was found that PRNCR1 was significantly overexpressed in CRC, with an average fold increase of 10.55 (P=0.006) (Fig. 1A). Then, the association between PRNCR1 expression and clinicopathological parameters was explored. As shown in Table I, tumor volume was significantly associated with the expression level of PRNCR1 after univariate and multivariate analyses. Specifically, the PRNCR1 expression level was higher in patients with a large volume tumor [Fig. 1B and Table I; large vs. small, 19.17 vs. 10.55; Punivariate=0.012, Pmultivariate=0.018 (OR=5.227; CI, 1.328–20.581)]. PRNCR1 expression was not correlated with gender, age, tumor site, differentiation, family history or TNM stage.

Figure 1 Analysis of PRNCR1 expression in the CRC tissues. (A) PRNCR1 was detected in 63 pairs of CRC tissues by qRT-PCR. The level of PRNCR1 in the CRC tissues was significantly higher than the level in the non-tumorous tissues (P=0.006). (B) PRNCR1 was upregulated in a group of tumors with a large volume (P=0.012) (*P<0.05).

Table I Correlation between PRNCR1 expression and clinicopathological characteristics of the CRC patients.

Table I

Correlation between PRNCR1 expression and clinicopathological characteristics of the CRC patients.

Characteristics	No. (%)	Fold-change	P-value
Total	63 (100)
Age (years)			0.596a
≥60.22d	34 (53.97)	10.73
<60.22d	29 (46.03)	15.69
Gender			0.888b
Male	37 (58.73)	15.30
Female	26 (41.27)	9.76
Family history			0.253b
Positive	17 (26.98)	6.56
Negative	46 (73.02)	15.40
Tumor site			0.632b
Below decending colon	49 (77.78)	14.85
Above sigmoid colon	14 (22.22)	6.59
Tumor size (cm3)			0.012a
≥23.53e	18 (28.57)	19.17	0.018f
<23.53e	45 (71.43)	10.55
Differentiation			0.491b
Poor	25 (39.68)	13.87
Moderate or high	38 (60.32)	12.45
T stage			0.087b
Entire serosal invasion	52 (82.54)	12.60
Subserous invasion	11 (17.46)	14.94
N stage			0.775b
Positive	28 (44.44)	18.81
Negative	35 (55.56)	15.05
M stage			0.756b
Positive	11 (17.46)	13.32
Negative	52 (82.54)	12.95

a P-value of Mann-Whitney test.

b P-value of Spearman's test.

c Significant correlation after univariate analysis.

d Mean age of the patients.

e Mean volume of the tumors.

F Significant correlation after multivariate analysis.

Expression profile of PRNCR1 in CRC cell lines

Initially, the expression profile in CRC cell lines was first assessed by qRT-PCR. When normalized to the FHC cells, the expression level of PRNCR1 was upregulated in most CRC cell lines (Fig. 2A). Specifically, PRNCR1 was upregulated in the HCT116, SW480, LoVo and HT-29 cells but was downregulated in the SW620 cells.

Figure 2 Analysis of the PRNCR1 expression in CRC cells and the silencing efficiency of ASO transfection. (A) Among the CRC cells, when compared to FHC cells, PRNCR1 was upregulated in the LoVo, SW480 and HT-29 cells but downregulated in the SW620 and HCT 116 cells. The HT-29 cell line showed the highest expression level (~8.4 fold). (B) PRNCR1 expression in the HT-29 cells after transfection of scramble or ASO. ASO-2 showed better inhibition efficiency. (C) Cell proliferation assay by CCK-8 assay. ASO-mediated silencing of PRNCR1 significantly inhibited cell proliferation in the HT-29 cells (**P<0.01).

PRNCR1 promotes the proliferation of CRC cell lines in vitro

Primarily, two ASOs specifically targeting PRNCR1 were designed to knock down PRNCR1 in vitro. Based on the relatively higher expression of PRNCR1, HT-29 cells were transfected with ASO-PRNCR1 or scramble. At 36 h after treatment, PRNCR1 expression was effectively knocked down (Fig. 2B). CCK8 assay showed that knockdown of PRNCR1 significantly inhibited cell proliferation in the HT-29 cell line (Fig. 2C). Then, we evaluated whether PRNCR1 could impact proliferation of CRC cells by altering the rate of apoptosis or cell cycle progression. Flow cytometric analysis was performed, and the results revealed that ASO treatment blocked HT-29 cells at the G0/G1 phase with a concomitant decrease of cells in the S phase (Fig. 3A). However, inhibition of PRNCR1 by ASO did not affect apoptosis (Fig. 3B).

Figure 3 Impacts of PRNCR1 on the cell cycle and apoptosis of HT-29 cells as analyzed by flow cytometry. (A) After transfection, HT-29 cells were prevented from G1 to S phase progression. ASO treatment blocked HT-29 cells at the G1 phase and inhibited DNA synthesis significantly. (B) Silencing of PRNCR1 did not affect apoptosis (*P<0.05; **P<0.01).

Furthermore, Transwell invasion and Matrigel invasion assays revealed that ASO-PRNCR1 treatment did not affect the migration and invasion capacities compared to the scramble control (Fig. 4). These results suggest that PRNCR1 promotes the proliferation of CRC cells.

Figure 4 After transfection, the migration (A) and invasion (B) capabilities did not show significant changes in the HT-29 cells.

Diagnostic value of PRNCR1

To further evaluate the potential diagnostic value of PRNCR1, an ROC curve was generated to assess the potential of PRNCR1 as an early diagnostic biomarker for CRC. As shown in Fig. 5, the area under the curve (AUC) was 0.799>0.5, which was higher than CEA-CA199 (AUC=0.651), indicating that PRNCR1 might have adequate potential as a biomarker. According to -the evaluation of the ROC curve, a cut-off value of 2.934 of a fold-change of PRNCR1 maximized the sensitivity (84.8%) and specificity (70.0%) in predicting the risk of CRC, and showed better efficiency than CEA-CA199.

Figure 5 ROC curves for PRNCR1 and CEA-CA199. A cut-off value of a fold-change of 2.934 of PRNCR1 maximized the sensitivity (84.8%) and specificity (70.0%) in predicting the risk of CRC. PRNCR1 showed better efficiency than CEA-CA199.

Discussion

lncRNAs are involved in every aspect of cancer progression, such as initiation, cancer metastasis, and could function as oncogenes and tumor suppressors (22). PRNCR1 can promote the progression of prostate cancer, yet the function and molecular mechanism in CRC remain unknown (20,21,25,26).

In the present study, we assessed the relationship between PRNCR1 expression and clinical characteristics of CRC patients and the possible function of PRNCR1 was probed by ASO-mediated inhibition in CRC cells. PRNCR1 expression was found to be increased in CRC tissues compared to the level in paired adjacent normal tissues. ASO-mediated silencing of PRNCR1 in HT-29 cells showed that silencing of PRNCR1 inhibited the proliferation and blocked cell cycle progression in the HT-29 cells. These data suggest that PRNCR1 may play an important role in CRC.

In addition, our results also indicated that a high level of PRNCR1 expression was markedly correlated with large tumor volume. In addition, it has been found that patients carrying the rs1456315G polymorphism have a tumor of much larger size (22). This could be explained by the following reasons. Primarily, the predicted secondary structure of PRNCR1 mRNA might be influenced by SNPs in PRNCR1, shifting the stability of the lncRNA PRNCR1 (20). Secondly, in terms of our findings, the silencing of PRNCR1 blocked the cell cycle at the G0/G1 phase indicating that PRNCR1 promotes the proliferation of CRC cells, and eventually may bring about larger tumor volume eventually.

Based on a literature review, five SNPs, rs13252298, rs1456315, rs1456315G, rs7007694C and rs16901946G, located in the lncRNA PRNCR1 exhibit a strong relationship with CRC tumorigenesis and development (22). Moreover, emerging genome-wide association studies (GWAS) show that a CpG site at Chr8: 128167809 in PRNCR1 and CRC susceptibility SNP rs1456315G have been found to be highly correlated with each other (27–29). In addition, correlations of these DNA methylation levels at CpG sites with each other within or nearby the GG genotype of rs6983267 in colon cancer associated transcript 2 (CCAT2) could be linked with c-MYC by enhancing Wnt signaling to upregulate transcription of CCAT2 (30–33), suggesting the homologous impact on PRNCR1. In agreement with our hypothesis, a previous study identified correlations for DNA methylation levels at CpG sites located near one another, especially those within distances equivalent to 1–2 kb apart (34). Moreover, compared with the functional mechanism of PRNCR1 in prostate cancer cells (20,21), a similarity may exist in CRC; however, much more research is still required to scrutinize all feasible targets of the characteristic site.

Disappointedly, no correlation between PRNCR1 and invasion, migration, apoptosis, differentiation and TNM stages of CRC was noted. The likely cause of this could be explained as follows. Poor differentiation of CRC cells furthers high metastatic and malignant potentials (35). Patients with SNPs in PRNCR1 have decreased risks to develop poorly differentiated CRC, while others with SNPs in PRNCR1 have increased risks (22). This may be why increased PRNCR1 expression may not relate to worse clinical characteristic of CRC.

Likewise, we detected the diagnostic value of PRNCR1 by ROC analysis, and PRNCR1 showed better predictive efficiency than serum biomarker CEA-CA199, indicating that PRNCR1 may be a potential biomarker for CRC diagnosis.

To sum up, we found that PRNCR1 is upregulated in CRC tissues and the upregulation of PRNCR1 is associated with large tumor volume. In addition, knockdown of PRNCR1 significantly inhibited proliferation and induced cell cycle arrest at the G0/G1 phase in the CRC cells, Moreover, PRNCR1 potentially could be an efficient predictive biomarker.

Acknowledgments

We thank Dr Zhicheng Chen (Department of Anorectal Clinic, Medical School of Southeast University) for gifting the CRC cells. This study was funded by the Natural Science Foundation of China (81372321 to L.X.; 81201830 and 81472200 to R.Y.), the Natural Science Foundation for High Education of Jiangsu Province (13KJB320010 to R.Y.), and Jiangsu Provincial Special Program of Medical Science (BL2012030 to L.X.).

References