miR-146a Polymorphism (rs2910164) Predicts Colorectal Cancer Patients’ Susceptibility to Liver Metastasis

miR-146a plays important roles in cancer as it directly targets NUMB, an inhibitor of Notch signaling. miR-146a is reportedly regulated by a G>C polymorphism (SNP; rs2910164). This polymorphism affects various cancers, including colorectal cancer (CRC). However, the clinical significance of miR-146a polymorphism in CRC remains unclear. A total of 59 patients with CRC were divided into 2 groups: a CC/CG genotype (n = 32) and a GG genotype (n = 27), based on the miR-146a polymorphism. cDNA microarray analysis was performed using 59 clinical samples. Significantly enriched gene sets in each genotype were extracted using GSEA. We also investigated the association between miR-146a polymorphism and miR-146a, NUMB expression or migratory response in CRC cell lines. The CC/CG genotype was associated with significantly more synchronous liver metastasis (p = 0.007). A heat map of the two genotypes showed that the expression profiles were clearly stratified. GSEA indicated that Notch signaling and JAK/STAT3 signaling were significantly associated with the CC/CG genotype (p = 0.004 and p = 0.023, respectively). CRC cell lines with the pre-miR-146a/C revealed significantly higher miR-146a expression (p = 0.034) and higher NUMB expression and chemotactic activity. In CRC, miR-146a polymorphism is involved in liver metastasis. Identification of this polymorphism could be useful to identify patients with a high risk of liver metastasis in CRC.


Introduction
Colorectal cancer (CRC) is the third most common neoplasm worldwide and is the second leading cause of cancer-related deaths in developed countries, with the majority of deaths attributable to distant metastasis [1,2]. Despite major advancements in diagnostic and therapeutic approaches to CRC, the prognosis of patients with distant metastasis commencing with liver metastasis is still unfavorable [3,4]. Therefore, there is an urgent need to establish a novel biomarker for cancer progression and metastasis in CRC.
Single nucleotide polymorphisms (SNPs) in miRNA have attracted attention because the SNPs may affect the expression and function of the miRNAs and be involved with the initiation and progression of cancer [17][18][19]. A common genetic variant rs2910164 located within the miR-146a precursor sequence was reported to change the stability of pri-miR by mispairing within the hairpin, followed by a reduction in the predicted ΔG from -43.1 kcal/mol to -40.3 kcal/mol [20]. The association between miR-146a polymorphism and the susceptibility to CRC has been well studied [21][22][23][24].With regard to the progression of CRC, Chae et al. demonstrated that patients with pre-miR-146a/C have a poor outcome [24], however, little is known about the clinicopathological relevance of miR-146a polymorphism in CRC. Here, we show that the association between the miR-146a polymorphism and liver metastasis in CRC occurs through Notch signaling and JAK/STAT3 signaling.

Collection of samples from CRC patients
Between December 2005 and July 2006, 59 CRC patients who underwent surgery at the National Cancer Center Hospital were enrolled in this study. This project was approved by Kyushu University Institutional Review Board for Human Genome/Gene Research and written informed consent was obtained from each patient. We collected peripheral blood mononuclear cell samples to determine the genotype of the miR-146a polymorphism. Resected tumor samples were immediately cut and stored in RNAlater (Ambion) or embedded in Tissue-Tek OCT (optimum cutting temperature) medium (Sakura, Tokyo, Japan), frozen in liquid nitrogen and kept at -80°C until RNA extraction.

CRC cell lines
Human CRC cells (RKO, HT29, WiDr, CaR1, LoVo, COLO320 and DLD1) were provided by the Japanese Cancer Research Bank (Tokyo, Japan). Cell lines were maintained in Dulbecco's Modified Eagle's medium supplemented with 10% fetal bovine serum and antibiotics. All cells were cultured as monolayers at 37°C in a humidified atmosphere containing 5% CO 2 .

Total RNA extraction
Tumor samples were used as pure cancer cells separated by laser microdissection as described in our previous report [25]. Total RNA from clinical samples and CRC cell lines were extracted using the modified acid-guanidine-phenol-chloroform method [26].

Quantitative reverse transcription PCR (RT-qPCR)
Quantitative analysis of miR-146a and RNU6B (internal control) was performed using specific cDNAs derived from total RNA extracted from CRC cell lines using gene-specific primers, according to the TaqMan MicroRNA Assay protocol (Assay IDs: 0000449 for has-miR-146a and 001093 for RNU6B; Applied Biosystems, Carlsbad, California, USA), as previously described [27]. The raw miR expression levels were normalized to RNU6B expression for calculation of the relative miR expression values.
RT was performed using an M-MLV reverse transcriptase kit according to the manufacturer's protocol (Invitrogen). cDNA was generated from 8 μg total RNA in a 30 μL reaction mixture and was diluted up to 100 μL with TE. To determine the relative expression levels of NUMB, qPCR was performed with a LightCycler 480 instrument (Roche Applied Science, Basel, Switzerland) using a LightCycler 480 Probes Master kit (Roche Applied Science), according to the manufacturer's protocols. The sequences of the primers for NUMB were as follows: sense, 5'-GTTGTCATGGGGGAGGTG-3' and antisense, 5'-TTGCTTAAGCCTCAAAT CTGC-3' . Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers were as follows: sense, 5'-TTGGTATCGTGGAAGGAC TCTCA-3' and antisense, 5'-TGTCATATTTGG CAGGTT-3'; GAPDH served as the internal control to normalize the expression level of NUMB. The amplification conditions were as follows: 10 min at 95°C followed by 45 cycles of 10 s at 95°C and 30 s at 60°C. The expression levels were expressed as the values relative to the expression levels of Human Universal Reference Total RNA (Clontech, Palo Alto, CA, USA).

Immunoblotting analysis
Total protein was extracted from CRC cell lines with RIPA buffer. Immunoblotting was performed as described in our previous report [28]. In brief, NUMB protein was detected using rabbit polycolonal antibodies (ab14140; Abcam)?. Protein levels were normalized to the level of β-actin protein, which was detected using mouse monoclonal antibodies (Cytoskeleton, Denver, CO, USA). Blots were developed with horseradish peroxidase-linked anti-rabbit or antimouse immunoglobulin (Promega, Madison, WI, USA). Enhanced chemiluminescence detection reagents (Amersham Biosciences, Piscataway, NJ, USA) were used to detect antigen-antibody reactions.

Migration assay
Cell migration was assessed using the BD Falcon FluoroBlok 24 Multiwell Insert System (BD Bioscience, San Jose, CA). The cells (1.0x10 5 cells/500 μl/well) were then placed in the upper chamber of the 24-well plate with serum-free medium. The lower chamber was filled with 10% fetal bovine serum, which acts as a chemoattractant, and incubated in a humidified atmosphere (37°C and 5% CO 2 ). After a 48 h incubation, invasive cells that migrated through the membrane were evaluated in a fluorescence plate reader at excition /emission wavelengths of 485/ 530 nm. Invasiveness was measured as the percentage of fluorescence of an invasive fibrosarcoma cell line (HT-1080) that served as a control. Each independent experiment was performed with at least three replicates.

PCR Amplification of Markers and DNA direct sequencing
The TaqMan probes and primers for rs2910164 were purchased from Applied Biosystems (assay ID C_15946974_10). Genotyping of clinical samples was performed with the ABI 7900HT Sequence Detection System and SDS 2.0 software (Applied Biosystems). In addition, miR-146a polymorphism of CRC cell lines was analyzed by PCR direct sequencing as described in a previous report [29,30]. The sequences of primers for a 192 bp fragment containing miR-146a polymorphism site (rs2910164) were as follows: 5'-CCGATGTGTATCCT CAGCTTTG-3' and 5'-GCCTGAGACTCTGCCTTCTG-3' .

Expression Analysis by cDNA Microarray
We performed microarray assays on 59 CRC tissues with the Human Whole Genome Oligo DNA Microarray Kit (Agilent Technologies, Santa Clara, CA) as described in our previous report [25]. A list of expressed genes on this cDNA microarray is available online (http://www.chem. agilent.com/scripts/generic.asp?1page=5175&indcol=Y&prodcol=Y&prodcol=N&indcol= Y&prodcol=N). A hierarchical cluster analysis of all the samples was performed by Euclidean distance and Ward's linkage algorithms.

Gene Set Enrichment Analysis (GSEA)
The association between miR-146a polymorphism and previously annotated gene expression signatures was analyzed by applying GSEA using expression profiles from cDNA microarray analysis of 59 CRC patients whose miR-146a polymorphism was provided, as previously described [28]. Gene sets extracted from the Broad Institute database and the Uniform Resource Locator of their source are as follows: KEGG_NOTCH_SIGNALING_PATHWAY (http://software.broadinstitute.org/gsea/msigdb/cards/KEGG_NOTCH_SIGNALING_ PATHWAY) and V$STAT3_01 (http://software.broadinstitute.org/gsea/msigdb/cards/V $STAT3_01).
Statistical analysis χ 2 tests or Fisher's exact test was used for comparisons between miR-146a polymorphisms and clinicopathological factors. Comparisons of miR-146a, NUMB expression or migratory response between a dominant model of miR-146a polymorphism were evaluated using Mann-Whitney's U-test. These results were analyzed using JMP 9 software (SAS Institute, Cary, NC, USA) or R version 3.1.1 (R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL: http://www.Rproject.org/). P values less than 0.05 were considered statistically significant.

Correlations between the miR-146a polymorphism and clinicopathological factors
We divided the 59 patients with CRC into 2 groups: those with the CC/CG genotype, having the pre-miR-146a/C (n = 32) and the GG genotype, not having pre-miR-146a/C, according to the dominant model of miR-146a polymorphism and compared the clinicopathological findings between the two genotypes (Table 1). Intriguingly, whereas no significant differences were noted with respect to distant metastasis, peritoneal dissemination or other factors pertaining to tumor progression, liver metastasis occurred more frequently in the CC/CG genotype than the GG genotype (p = 0.007).

Different cDNA expression profiles and gene set signatures found in the miR-146a polymorphism
From the microarray assay, we identified 570 significantly upregulated genes and 443 significantly downregulated genes in the CC/CG genotype compared with the GG genotype. We examined a heat map of the clustering of the expression pattern of those genes in the 59 CRC cases (Fig 1). The characteristics of the expression profiles of the miR-146a polymorphism were clearly stratified. Furthermore, GSEA indicated that a number of Notch signaling pathway signatures and JAK/STAT3 signaling pathway signatures were significantly enriched in the CC/CG genotype (p = 0.004 and p = 0.023, respectively) (Fig 2).

The influence of the miR-146a polymorphism on the expression of miR-146a and NUMB and migratory response
To explore the influence of miR-146a polymorphism on the expression of miR-146a and NUMB, we examined the relationship between the polymorphism and the expression of miR-146a or NUMB in 7 CRC cell lines, whose polymorphism was determined by direct sequencing ( Table 2). miR-146a expression was significantly higher in 3 CRC cell lines with the pre-miR-146a/C than the 4 CRC cell lines without the pre-miR-146a/C (p = 0.034) (Fig 3A). There was no association between the miR-146a polymorphism and NUMB expression in the CRC cell lines (Fig 3B), however, immunoblotting analysis revealed that NUMB expression of CRC cell lines with pre-miR-146a/C (CaR1, Lovo and DLD-1) was lower than WiDr and HT29, CRC cell lines without pre-miR-146a/C (Fig 3C). Furthermore, DLD-1, one CRC cell line with the pre-miR-146a/C had higher chemotactic activity than the CRC cell lines without the pre-miR-146a/C (Fig 3D). Inhibition of miR-146a expression increased NUMB expression in CRC In order to confirm the influence of miR-146a expression on the NUMB expression in CRC, we investigated NUMB expression of CRC cell lines with the pre-miR-146a/C treated by miR-146a inhibitor. Knockdown of miR-146a, confirmed by RT-qPCR ( Fig 4A) enhanced NUMB expression in CRC cell lines with the pre-miR-146a/C (Fig 4B).

Discussion
Numerous previous studies have described the role of miR-146a polymorphism in patients' susceptibility to CRC. In this study, we demonstrated that miR-146a polymorphism was also involved with liver metastasis in CRC patients. We hypothesized that the molecular mechanism of liver metastasis in CRC is caused by dramatic changes of gene signatures affected by miR-146a polymorphism. GSEA showed that Notch signaling and JAK-STAT3 signaling were significantly activated in the patients with pre-miR-146a/C. Notch signaling may be involved in cell-fate decision and promote cell survival or anti-apoptosis by mediating the expression of some anti-apoptosis or pro-survival proteins [31,32]. Notch signaling also mediates the epithelial-mesenchymal transition process, leading to metastasis [31] and has been traditionally implicated as a key mechanism for the initiation and progression in CRC [33]. Sonoshita et al. demonstrated that inhibition of Notch signaling by amino-terminal enhancer of split (Aes) suppressed the invasiveness and the intravasation of CRC cells in an orthotopic transplantation model [34]. We previously reported that Notch induced CCL2 which in turn promoted distant metastasis in vivo [35] . Another pathway HGF/c-MET/ETS-1 also plays a central role in tumor proliferation, invasion and metastasis [36,37]. However, GSEA showed no significant enrichment for the gene set of HGF/c-MET/ETS-1 pathway (data not shown). Thus, miR-146a polymorphism may promote liver metastasis in CRC via Notch signaling and JAK/STAT3 signaling. It is well known that the common genetic variant rs2910164 is located within the miR-146a precursor sequence and that it affects the expression of mature miR-146a [29,30,[38][39][40][41]. The pre-miR-146a/C genotype may reduce the stability of pri-miR by mispairing events within the hairpin [42] and it has been associated with lower expression of the miR-146a in several cancers [38][39][40][41]. We previously reported that patients with the pre-miR-146a/C genotype showed higher expression of miR-146a than those with the pre-miR-146a/G genotype in gastric cancer    [30]. The effect of miR-146a polymorphism on the expression of miR-146a has been demonstrated in vitro [29]. In our study, the clinical samples whose SNP was examined were unfortunately unavailable. However, CRC cell lines with the pre-miR-146a/C genotype have significantly higher miR-146a expression than those with the pre-miR-146a/G. This discrepancy may be explained by the ethnicity, cancer type, and disease status. NUMB protein is a key negative regulator of Notch signaling [43] and is one of the direct targets of miR-146a [15,41,44,45]. miR-146a induces an oncogenic phenotype and tumorigenesis of oral squamous cell carcinoma by directly targeting the 3'-UTR of NUMB [45]. In CRC, Snail induces miR-146a expression, which in turn targets NUMB and stabilizes β-catenin [15]. Contrary to our result with miR-146a expression, CRC cell lines with the pre-miR-146a/C genotype had the lower expression of NUMB protein and more migratory response than those with a pre-miR-146a/G genotype. NUMB suppression may be affected by alteration of miR-146a expression through this SNP, followed by increasing the migratory response via enhancing Notch signaling; however, further examination is needed to confirm the effect of miR-146a polymorphism on liver metastasis in CRC using genome editing.
In conclusion, we demonstrated that miR-146a polymorphism is associated with the tendency of CRC to metastasize to the liver. This occurs through miR-146a-mediated activation of Notch and JAK/STAT3 signaling via suppression of NUMB. Identification of miR-146a polymorphism could identify patients with high-grade CRC and be useful for follow-up with particular attention to liver metastasis.