INTRODUCTION

Colorectal cancer (CRC) ranks as the third most common cancer and the second leading cause of cancer-related mortality globally (1,2). Notably, there is an increasing trend in early-onset CRC (patients younger than 50 years), which currently accounts for approximately 10% of all new diagnoses (3). Patients with early-onset CRC are more likely to present with hereditary syndromes compared with their late-onset counterparts (1,4). High-frequency microsatellite instability caused by germline mutations occurs more frequently in early-onset CRC, while sporadic high-frequency microsatellite instability tumors caused by epigenetic inactivation of MLH1 occur more frequently in late-onset CRC (1,5–7). Sporadic CRC happens more frequently than hereditary CRC. The association of epigenetic changes in other genes was not observed in early-onset/late-onset CRC.

Nuclear receptors (NRs) usually play important roles in regulating gene expression, cell growth, and development by functioning as transcription factors (8). NR-interacting protein 1 (NRIP1) has been documented to act as either an activator or inhibitor in various cancer types (8–10). NR-interacting protein 3 (NRIP3) is a novel transcriptional coregulator of NR, which is located on chromosome 11p15.4, a fragile site associated with cancer. A recent study conducted by Wolfe et al (11) reported that defects of this region were related to embryonal tumors. However, the precise regulatory mechanisms and expression patterns of NRIP3 in CRC remain elusive. Therefore, the objective of this investigation was to analyze the epigenetic changes and elucidate the underlying mechanisms of NRIP3 in human CRC, with the ultimate aim of uncovering innovative therapeutic approaches.

METHODS Human tissue samples and cell lines

To assess the possibility of gene expression regulation, we used the TCGA and GTEx databases (http://gepia.cancer-pku.cn/index.html). To ensure the exclusion of tissue-specific methylation, 6 samples of colorectal mucosa were collected from noncancerous individuals. To further investigate the methylation status of NRIP3 in CRC, a total of 187 resected margin samples from CRC tissue, 146 cases of colorectal adenomatous polyps, and 308 CRC tissue samples were examined by methylation-specific PCR (MSP). All samples were obtained from the Chinese PLA General Hospital. Patient medical records were reviewed to collect following information: age, sex, polyp characteristics (number, size, pathology, and location), colorectal cancer characteristics (stage, size, location, differentiation, and lymph node metastasis), and treatment variables (chemotherapy and radiotherapy). All tumors were classified according to TNM staging system (eighth). All CRC cells were previously derived from primary colorectal cancer, including DKO, SW480, LOVO, HT29, HCT116, DLD1, RKO, and SW620. The institutional review board of the Chinese PLA General Hospital granted approval for the procedure.

Lentiviral expression vectors and stable expression cells for NRIP3

For functional study, human NRIP3 CDS (NM_020645.3) was constructed using the expression vector pCDH-CMV-MCS-puro and amplificated using the specified primers: 5′-CCG​GAA​TTC​ATG​TTT​TAC​TCA​GGG​CTC​CTC​ACT​G-3′ (F) and 5′-GCG​GGA​TCC​TTA​TGC​TTC​TGA​AGT​GTT​GTC​TTC​ATT​C-3′ (R). NRIP3-expressing lentiviral or empty vectors were packaged according to previously described methods (12). The lentivirus was introduced into DLD1, RKO, and HCT116 cells, by adding it to the growth medium. NRIP3 stably expressed cells were selected with 2.0 μg/mL (DLD1), 1.0 μg/mL (RKO), and 1.5 μg/mL (HCT116) of puromycin for 3 days.

SiRNA knockdown technique

Specific sequences for NRIP3 siRNA and scrambled control duplex are listed in Supplementary Table 2 (see Supplementary Digital Content 3, https://links.lww.com/CTG/B68). SiRNA oligonucleotide and SiRNA scrambled control duplex (Gene Pharma, Shanghai, China) were transfected into NRIP3 highly expressed cells. The efficiency of transfection was assessed, and siRNA#3 was found to be the most effective.

Xenograft mouse model

NRIP3 stably expressed and unexpressed DLD1 (4 × 106 cells) and RKO cells (6 × 106 cells) were suspended in 0.15 mL of phosphate-buffered saline and subcutaneously injected into the dorsal right side of 4-week-old BABL/c nude mice. The tumor volume was measured every 3 days for a total of 18 days, using the formula V = L × W2/2, where V represents volume (mm3), L represents the biggest diameter (mm), and W represents the smallest diameter (mm). All experimental procedures were conducted in accordance with the guidelines by the Animal Ethics Committee of the Chinese PLA General Hospital.

Evaluating the sensitivity of CRC cells to VE-822 and AZD0156

The IC50 value was evaluated using an MTT assay. CRC cells, both with and without NRIP3 expression, were seeded into 96-well plates at a density of 2000 cells/well. The cells were treated by 0.1 μM cisplatin (#S1166; Selleck, Houston, TX) combination with 0, 0.2, 0.4, 0.8, 1.6, 3.2, 6.4, 12.8, and 25.6 μM VE-822 (#S7102; Selleck) or 0, 1, 2, 4, 8, 16, 32, 64, and 128 μM AZD0156 (#S8375; Selleck) for 48 hours. For synthetic lethality analysis, NRIP3 unexpressed DLD1, RKO, and HCT116 cells were treated with 0.2 μM VE-822/combined with 50 nM NVP-BEZ235 (#S1009; Selleck) or treated with 0.5 μM AZD0156/combined with 50 nM NVP-BEZ235 under 0.1 μM cisplatin treatment for DNA damage. Each experiment was repeated 3 times.

Additional materials and methods

RNA isolation, RT-PCR, methylation detection, immunohistochemistry, MTT assay, colony formation, cell cycle, transwell assay, gene expression array analysis, and Western blot approaches were described in Supplementary Materials and Methods (see Supplementary Digital Content 1, https://links.lww.com/CTG/B66). Primer sequences are listed in Supplementary Table 1 (see Supplementary Digital Content 2, https://links.lww.com/CTG/B67).

Statistical analysis

SPSS software (version 22.0; IBM, Armonk, NY) was used in this study, and the χ2 test was used for independent dichotomous variables. The Student t test was applied for mean ± SD analysis. Kaplan-Meier plots and log-rank tests were performed to evaluate overall survival (OS). The association with 5-year OS was analyzed using univariate and multivariate Cox proportional hazards regression models. P < 0.05 is considered as significant difference.

RESULTS The expression of NRIP3 is regulated by promoter region methylation

The expression of NRIP3 was reduced in CRC samples compared with that in normal colonic mucosa from noncancerous patients (P < 0.05, Figure 1a). Furthermore, we observed an inverse association between the mRNA levels of NRIP3 and the methylation status around the transcription start site in 307 cases of available CRC data (cg03963327, cg03323636, cg03666741, all P < 0.05, Figure 1b and c).

Figure 1.:

The expression and methylation status of NRIP3 in CRC. (a) The levels of NRIP3 in CRC tumor tissue and normal tissue samples (TCGA + GTEx datasets). num, number; T, tumor; N, normal; COAD, colorectal adenocarcinoma cancer; READ, rectal adenocarcinoma cancer. (b) The association of NRIP3 expression and methylation status of CpG sites close to TSS site in the promoter region. TSS, transcription start site. (c) Scatter plots show representative CpG sites methylation status and NRIP3 expression association (cg03963327, cg03323636 and cg03666741). (d) Semiquantitative RT-PCR shows the expression of NRIP3 in CRC cells. 5-Aza, 5-aza-2′-deoxycytidine; GAPDH, internal control of RT-PCR; H2O, double distilled water; (−), absence of 5-Aza; (+), presence of 5-Aza. (e) MSP results of NRIP3 in CRC cells. U, unmethylated alleles; M, methylated alleles; IVD, in vitro methylated DNA, serves as methylation control; NL, normal lymphocytes DNA, serves as unmethylation control; H2O, double distilled water. (f) BSSQ results of NRIP3 in DLD1, RKO, HCT116, and DKO cells. The size of unmethylated MSP products is 183 bp. Bisulfite sequencing was performed in a 261-bp region around the NRIP3 transcription start site (from −354 bp to −94 bp). Filled circles, methylated CpG sites; open circles, unmethylated CpG sites; TSS, transcription start site.

Notably, NRIP3 exhibited high expression in DKO cells, no expression in LOVO, HT29, HCT116, DLD1, and RKO cells, and reduced expression in SW480 and SW620 cells (Figure 1d). NRIP3 was unmethylated in DKO cells, completely methylated in LOVO, HT29, HCT116, DLD1, and RKO cells, and partially methylated in SW480 and SW620 cells (Figure 1e). The regulation of NRIP3 expression by methylation was further validated by 5-aza-2′-deoxycytidine (5-Aza) treatment. The expression of NRIP3 was restored in LOVO, HT29, HCT116, DLD1, and RKO cells, increased in SW620 and SW480 cells, but remained unchanged in DKO cells (Figure 1d). These findings suggested that the expression of NRIP3 is regulated by promoter methylation. NRIP3 was densely methylated in DLD1, RKO, and HCT116 cells and unmethylated in DKO cells (Figure 1f). Consistent results were observed in bisulfite sequencing and MSP.

NRIP3 is frequently methylated in human primary CRC

As shown in Figure 2a, tissue-specific methylation was excluded by detection of 6 cases of normal colorectal mucosa. Among the margin samples obtained from resected CRC tissues, NRIP3 methylation was observed in 5 of 187 cases (2.7%). Notably, all the methylated cases were older than 69 years. Although no statistical significance was observed, the findings imply a potential relationship between NRIP3 methylation and old age. NRIP3 was methylated in 32.2% (47/146) of colonic adenomatous polyps and 50.6% (156/308) of CRC (P < 0.01, Figure 2a and b), suggesting that NRIP3 may be an early detection marker for CRC. The median age of adenomatous polyp patients was 60 years (28–88 years), including 93 cases of male and 53 cases of female individuals. The number of cases were 64 in proximal colon (28 in ascending colon and 36 transverse colon) and 82 in distal colon (17 in descending colon, 41 in sigmoid colon and 24 rectum). A significant association was observed between NRIP3 methylation and older age (age 50 years or older, P < 0.05, see Supplementary Table 3, Supplementary Digital Content 4, https://links.lww.com/CTG/B69), while no significant associations were found with gender, polyp size, or location (all P > 0.05, see Supplementary Table 3, Supplementary Digital Content 4, https://links.lww.com/CTG/B69).

Figure 2.:

The methylation and expression status of NRIP3 in normal colorectal mucosa, colonic polyps, and primary CRC. (a) Representative MSP results of NRIP3 in normal colorectal mucosa, colonic adenomatous polyps, and primary CRC samples. N, normal colorectal mucosa; CRP, colonic adenomatous polyp samples; CRC, primary CRC samples. IVD, in vitro methylated DNA, serves as methylation control; NL, normal lymphocytes DNA, serves as unmethylation control; H2O, double distilled water. (b) NRIP3 methylation frequency in colonic adenomatous polyps and primary CRC samples. **P < 0.01. (c) The association of NRIP3 methylation and 5-year OS of CRC. (d) NRIP3 staining in CRC and adjacent tissue samples (top: 200×; bottom: 400×). (e) Box plots for NRIP3 expression, horizontal lines represent the median score; the bottom and top of the boxes represent the 25th and 75th percentiles, respectively; vertical bars represent expression levels. (f) Bar diagram: NRIP3 expression and case numbers; white: unmethylation; gray: methylation.

Among patients with CRC, the median age was 60 years (29–86 years), including 192 cases of male and 116 cases of female individuals. The number of cases were 80 in proximal colon (68 in ascending colon and 12 in transverse colon) and 228 in distal colon (19 in descending colon, 59 in sigmoid colon, and 150 in rectum). No chemotherapy or radiotherapy was performed before surgery for all patients. NRIP3 methylation was significantly associated with older age (age 50 years or older), tumor differentiation, and lymph node metastasis (all P < 0.05, Table 1). No significant associations were observed between NRIP3 methylation and gender, tumor size, TNM stage, and tumor location (all P > 0.05, Table 1). In 190 cases of CRC with available OS data, NRIP3 methylation was found significantly associated with poor 5-year OS (P < 0.05, Figure 2c) and was an independent prognostic factor for poor 5-year OS (P < 0.05, Table 2).

Table 1. - Clinical factors and NRIP3 methylation status in patients with colorectal cancer Clinical parameter n = 308 Methylation status P value Unmethylated (n = 152, 49.4%) Methylated (n = 156, 50.6%) Gender  Male 192 101 91 0.1417  Female 116 51 65 Age  <50 56 35 21 0.0296a  ≥50 252 117 135 Differentiation  Well/moderately 239 130 109 0.0010a  Poorly 69 22 47 TNM stage  I/II 188 99 89 0.1460  III/IV 120 53 67 Lymph node metastasis  Negative 203 110 93 0.0182a  Positive 105 42 63 Tumor size  <5 cm 163 81 82 0.8985  ≥5 cm 145 71 74 Tumor location  Proximal 80 35 45 0.2442  Distal 228 117 111

aP values are obtained from the χ2 test, significant difference (P < 0.05).

Table 2. - Univariate and multivariate analyses of NRIP3 methylation status with 5-year overall survival in patients with CRC Clinical parameter Univariate analysis Multivariate analysis HR (95% CI) P value HR (95% CI) P value Gender (male vs female) 1.345 (0.692–2.615) 0.382 Age (≥50 vs <50 yr) 1.418 (0.730–2.757) 0.303 Tumor size (≥4 vs <4 cm) 1.211 (0.635–2.307) 0.561 Differentiation (low vs high or middle differentiation) 2.420 (1.231–4.757) 0.010a 1.476 (0.730–2.984) 0.278 TNM stage (III/IV vs I/II) 4.831 (2.384–9.791) 0.000b 11.185 (3.008–41.587) 0.000b Lymph node metastasis (positive vs negative) 3.358 (1.727–6.530) 0.000b 0.341 (0.097–1.195) 0.093 NRIP3 (methylation vs unmethylation) 2.396 (1.203–4.773) 0.013a 2.256 (1.069–4.761) 0.033a Tumor location (distal vs proximal) 0.648 (0.330–1.272) 0.208

CI, confidence interval; HR, hazard ratio.

aP < 0.05.

bP < 0.001.

To assess NRIP3 levels, an IHC assay was used in 72 cases of available CRC and matched margin samples. The expression of NRIP3 was higher in margin tissue samples compared with CRC tissues (P < 0.001, Figure 2d and e), with staining observed in both the nucleus and cytoplasm. The reduction in NRIP3 levels was associated with the methylation of its promoter region (P < 0.001, Figure 2f), indicating the epigenetic regulation of NRIP3 expression in CRC.

NRIP3 suppresses CRC cell proliferation

To assess the impact of NRIP3 on cell proliferation, MTT assay and colony formation were used. The OD values before and after reexpression of NRIP3 in DLD1, RKO, and HCT116 cells were 1.500 ± 0.135 vs 1.175 ± 0.094, 1.086 ± 0.086 vs 0.702 ± 0.075, and 1.406 ± 0.038 vs 1.042 ± 0.032, respectively (Figure 3a). The restoration of NRIP3 expression led to a significant decrease in the OD value, suggesting that NRIP3 suppresses CRC cell proliferation (all P < 0.05). The siRNA technique was used to validate the effect of NRIP3 on cell proliferation. The OD value in NRIP3 expression DKO cells and siRNA knocking down cells was 0.987 ± 0.092 vs 1.513 ± 0.111 (P < 0.01, Figure 3a), further demonstrating the inhibitory effect of NRIP3 on CRC cell growth. The clone number before and after reexpression of NRIP3 in DLD1, RKO, and HCT116 cells was 210.0 ± 7.34 vs 146.7 ± 7.41, 241.0 ± 4.55 vs 153.7 ± 6.13, and 150.7 ± 5.25 vs 100.0 ± 5.72, respectively. The clone number exhibited a significant reduction after reexpression of NRIP3 in CRC cells (all P < 0.01, Figure 3b). To further validate the effect of NRIP3 on clonogenicity, the siRNA knockdown technique was used. In DKO cells, the number of clones was 82.0 ± 5.10 vs 128.7 ± 2.62 in the control and siRNA knockdown groups. The clone number was increased significantly by knockdown of NRIP3 in DKO cell (P < 0.01, Figure 3b). These results demonstrated that NRIP3 inhibits CRC cell proliferation.

Figure 3.:

The effect of NRIP3 on cell proliferation and cell cycles. (a) Growth curves represent the effects of NRIP3 in CRC cells. *P < 0.05, **P < 0.01, and ***P < 0.001. (b) Colony formation results. The average number of tumor clones is represented by the bar diagram. **P < 0.01, ***P < 0.001. (c) The effect of NRIP3 on cell phase. The bar diagram represents the percentage. *P < 0.05, **P < 0.01, and ***P < 0.001. (d) Western blots show the effects of NRIP3 knockdown. Scramble: siRNA negative control; siRNA#1, siRNA#2, and siRNA#3: 3 different siRNA for NRIP3. (e) Western blot shows the levels of NRIP3, cyclin E1, cyclin A2, and cyclinD1. Actin: internal control. Scramble: siRNA negative control; siRNA#3: siRNA targeting NRIP3.

NRIP3 induces G1/S phase arrest

The effects of NRIP3 on the cell cycle were assessed using flow cytometry. In NRIP3 unexpressed and reexpressed DLD1 cells, the percentages of cell phase were 61.08 ± 0.60% vs 64.67 ± 0.84% in G0/G1 phase (P < 0.01), 28.23 ± 0.05% vs 25.88 ± 0.34% in S phase (P < 0.01), and 10.68 ± 0.57% vs 9.45 ± 0.52% in G2/M phase. For NRIP3 unexpressed and reexpressed RKO cells, the percentages of cell phase were 44.15 ± 0.04% vs 46.97 ± 0.98% in G0/G1 phase (P < 0.01), 35.65 ± 0.42% vs 32.64 ± 0.98% in S phase (P < 0.05), and 20.20 ± 0.44% vs 20.39 ± 1.05% in G2/M phase. Similarly, for NRIP3 unexpressed and reexpressed HCT116 cells, the percentages of cell phase were 42.75 ± 0.48% vs 48.98 ± 1.79% in G0/G1 phase (P < 0.01), 39.22 ± 1.14% vs 32.18 ± 0.99% in S phase (P < 0.01), and 18.03 ± 1.30% vs 18.84 ± 0.82% in G2/M phase. These findings suggest that NRIP3 induces G1/S phase arrest (Figure 3c). The cell phase distribution was 64.96 ± 0.48% vs 55.31 ± 1.17% in G0/G1 phase (P < 0.01), 22.49 ± 0.29% vs 33.78 ± 0.55% in S phase (P < 0.01), and 12.55 ± 0.44% vs 10.92 ± 0.65% in G2/M phase before and after knockdown of NRIP3 in DKO cells (Figure 3c), further validating the abovementioned results. In addition, the reexpression of NRIP3 in DLD1, RKO, and HCT116 cells led to a decrease in the levels of cyclin D1, cyclin A2, and cyclin E1. Conversely, NRIP3 knockdown in DKO cells resulted in an increase in the levels of cyclin D1, cyclin A2, and cyclin E1 (Figure 3d and e). These observations further support that NRIP3 induces G1/S arrest.

NRIP3 suppresses cell migration and invasion

The role of NRIP3 on cell migration and invasion was examined by transwell assay. In NRIP3 unexpressed and reexpressed DLD1, RKO, and HCT116 cells, the number of migratory cells was 224.00 ± 13.74 vs 178.08 ± 7.97, 176.92 ± 6.69 vs 102.5 ± 9.71, and 232.63 ± 11.55 vs 181.40 ± 18.65, respectively. Notably, the reexpression of NRIP3 in CRC cells resulted in a significant reduction in the number of migratory cells (all P < 0.01, Figure 4a). Furthermore, the knockdown of NRIP3 in DKO cells led to an increase in the number of migratory cells from 151.29 ± 14.78 to 173.90 ± 35.39. The knockdown of NRIP3 in DKO cells resulted in a significant increase in the number of migratory cells (P < 0.001, Figure 4a). The abovementioned results demonstrate that NRIP3 suppresses CRC cells migration.

Figure 4.:

The effect of NRIP3 on cell migration and invasion, PI3K-AKT signaling in CRC cells. (a) Migration and invasion results. Bar diagram represent the average number of migration and invasion cells. ***P < 0.001. (b) Western blot shows the levels of NRIP3, MMP2, MMP7, and MMP9. (c) The KEGG enrichment analysis in unexpressed vs reexpressed RKO cells. (d) Western blot shows the levels of NRIP3, PI3K, AKT, p-AKT, mTOR, and p-mTOR in CRC cells. (e) MTT assay shows cell growth curves. Empty vector: NRIP3 unexpressed cells; Empty vector +: NRIP3 unexpressed cells plus NVP-BEZ235 treatment (50 nM); NRIP3: NRIP3 reexpressed cells; NRIP3+: NRIP3 expression plus NVP-BEZ235 treatment. Scramble: siRNA negative control; Scramble+: Scramble siRNA plus NVP-BEZ235 treatment; siRNA#3: the most effective NRIP3 targeting siRNA; siRNA#3+: siNRIP3 plus NVP-BEZ235 treatment. ***P < 0.001. (f) Western blot shows the levels of NRIP3, PI3K, AKT, p-AKT, mTOR, and p-mTOR, before and after NVP-BEZ235 treatment, with or without NRIP3 expression.

The number of invasive cells was 176.50 ± 9.40 vs 129.58 ± 4.84, 177.08 ± 10.63 vs 115.17 ± 9.61, and 193.63 ± 6.89 vs 137.33 ± 27.33 in NRIP3 silencing and forcing–expressed DLD1, RKO, and HCT116 cells, respectively, demonstrating that NRIP3 inhibits CRC cell invasion (all P < 0.01, Figure 4a). The number of invasive cells was 80.14 ± 7.92 vs 103.40 ± 12.82 in NRIP3 highly expressed and siRNA knocking down DKO cells. The knockdown of NRIP3 resulted in a significant increase in the number of invasive cells (P < 0.01, Figure 4a). The abovementioned data demonstrate that NRIP3 suppresses cell invasion. Figure 4b illustrates that the levels of MMP2, MMP7, and MMP9 decreased after reexpression of NRIP3 in DLD1, RKO, and HCT116 cells and increased after NRIP3 knockdown in DKO cells, further validating the results at the molecular level.

NRIP3 inhibits PI3K-AKT signaling

To understand the mechanism of NRIP3 in the development of CRC, a microarray analysis was conducted to compare the gene expression profiles of RKO cells with and without NRIP3 expression. The results were analyzed using KEGG pathway enrichment, revealing significant alterations in the PI3K-AKT signaling (Figure 4c). NRIP3 was reported to be involved in PI3K-AKT signaling in gastric mucosa in obese patients and patients with obesity-related diabetes (13). Increased expression of NRIP3 was reported in PIK3CA gene mutated breast cancer (14). Thus, we focused on the role of NRIP3 in PI3K-AKT signaling. PI3K, p-AKT, and p-mTOR levels were reduced after reexpression of NRIP3 in DLD1, RKO, and HCT116 cells, while no changes were observed in the total levels of AKT and mTOR. In DKO cells, the knockdown of NRIP3 resulted in an increase in PI3K, p-AKT, and p-mTOR levels, while no obvious changes of total AKT and mTOR were noticed (Figure 4d). This observation suggests that NRIP3 plays an inhibitory role in the PI3K-AKT signaling pathway.

To further validate the involvement of NRIP3 in the PI3K-AKT-mTOR signaling pathway, NVP-BEZ235, a PI3K-mTOR inhibitor, was used. The OD values before and after NVP-BEZ235 treatment were 1.306 ± 0.088 vs 0.857 ± 0.055, 1.154 ± 0.013 vs 0.642 ± 0.018, and 0.888 ± 0.034 vs 0.740 ± 0.032 in NRIP3 unexpressed DLD1, RKO, and HCT116 cells, respectively. These results demonstrate a significant decrease in OD values on inhibition of the PI3K-mTOR signaling pathway (all P < 0.01, Figure 4e). While after reexpression of NRIP3 in DLD1, RKO, and HCT116 cells, the OD values were 0.915 ± 0.095 vs 0.908 ± 0.147, 0.631 ± 0.030 vs 0.581 ± 0.067, and 0.578 ± 0.017 vs 0.559 ± 0.005 under the conditions of NVP-BEZ235 untreatment and treatment, respectively. No significant differences were observed between groups treated with the PI3K-mTOR inhibitor and the untreated groups (all P > 0.05). Before and after NVP-BEZ235 treatment, the OD values were 0.640 ± 0.024 vs 0.617 ± 0.013 in the siRNA scramble control DKO group (P > 0.05). In siNRIP3 knockdown DKO cells, the OD values were 0.755 ± 0.024 vs 0.501 ± 0.027 before and after NVP-BEZ235 treatment, respectively, indicating a significant decrease in OD value by NVP-BEZ235 treatment (P < 0.01, Figure 4e). These results imply that the defect of NRIP3 activated PI3K-AKT pathway.

NVP-BEZ235 resulted in decreased levels of PI3K, p-AKT, and p-mTOR in NRIP3 unexpressed DLD1, RKO, and HCT116 cells, but no significant changes were observed in NRIP3 forced expression cells (Figure 4f). In DKO cells, NVP-BEZ235 led to decreased levels of PI3K, p-AKT, and p-mTOR by knocking down NRIP3, while no obvious difference was observed in scramble DKO cells (Figure 4f). These data further demonstrate that PI3K-AKT signaling is inhibited by NRIP3 in CRC.

CRC cell xenografts were suppressed by NRIP3

In DLD1 and RKO cell xenografts with or without NRIP3 expression, the tumor volume was 688.07 ± 125.32 mm3 vs 444.08 ± 52.76 mm3 (P < 0.01) and 511.53 ± 70.90 mm3 vs 239.33 ± 55.05 mm3 (P < 0.01), respectively. Tumor volume was reduced in NRIP3 reexpressed xenografts compared with NRIP3 unexpressed xenografts (Figure 5a and b). The xenograft weight was 0.500 ± 0.080 g vs 0.254 ± 0.048 g, and 0.121 ± 0.014 g vs 0.037 ± 0.016 g in NRIP3 silenced and reexpressed DLD1 and RKO cells, respectively. Restoration of NRIP3 expression significantly reduced the tumor weight (all P < 0.01, Figure 5c). The abundance of PI3K, p-AKT, and p-mTOR was assessed using IHC in NRIP3 unexpressed and expressed xenografts. The expression of NRIP3 resulted in a reduction of PI3K, p-AKT, and p-mTOR levels, indicating the inhibitory role of NRIP3 in PI3K-AKT signaling in vivo (Figure 5d).

Figure 5.:

The effect of NRIP3 on PI3K/AKT signaling in CRC cell xenografts. (a) Representative results of NRIP3 unexpressed and reexpressed cell xenografts. (b) Growth curves of NRIP3 unexpressed and reexpressed cell xenografts. **P < 0.01, ***P < 0.001. (c) Tumor weight of NRIP3 unexpressed and reexpressed cell xenografts. Bars represent the tumor weight. ***P < 0.001. (d) Representative immunohistochemistry results for xenografts.

NRIP3 methylation is a sensitive marker of combined PI3K-Akt-mTOR inhibitor with ATR/ATM inhibitors

PI3K has been reported to be involved in DNA damage repair, and combination of its inhibitor with a CHK1 inhibitor enforced antitumor efficiency (15,16). Our study demonstrated that epigenetic silencing of NRIP3 activates the PI3K pathway in CRC. Thus, we assessed the sensitivity of NRIP3 expressed and silenced CRC cells to VE-822 (an ATR inhibitor) and AZD0156 (an ATM inhibitor). Under the low levels of cisplatin treatment, the IC50 of VE-822 was 2.540 ± 0.411 μM vs 0.839 ± 0.179 μM, 3.779 ± 0.473 μM vs 0.479 ± 0.088 μM, and 10.272 ± 0.607 μM vs 0.659 ± 0.083 μM in NRIP3 unexpressed and force-expressed DLD1, RKO, and HCT116 cells, respectively. Reduced IC50 value was observed by expression of NRIP3 (all P < 0.01, Figure 6a). Similarly, under low-level cisplatin induction, the IC50 values of AZD0156 were 9.980 ± 1.137 μM vs 5.211 ± 0.817 μM, 13.270 ± 2.010 μM vs 7.973 ± 0.672 μM, and 18.479 ± 0.305 μM vs 10.640 ± 1.470 μM in DLD1, RKO, and HCT116 cells without expression and forced expression of NRIP3, respectively. The IC50 value of AZD0156 was significantly reduced by NRIP3 (all P < 0.05, Figure 6a). These results demonstrated that NRIP3 enhances the sensitivity of CRC cells to ATR and ATM inhibitors.

Figure 6.:

Loss of NRIP3 expression sensitized cells to NVP-BEZ235/VE-822 and NVP-BEZ235/AZD0156. (a) The IC50 curve of VE-822 and AZD0156 under cisplatin treatment in CRC cells. (b) The levels of ATR, p-ATR, ATM, p-ATM, CHK1, p-CHK1, CHK2, and p-CHK2 in NRIP3 unexpressed and reexpressed CRC cells with or without cisplatin treatment (0.1 μM). (c) MTT assay shows the sensitivity of CRC cells to VE-822, AZD0156, or NVP-BEZ235. VE-822: 0.2 μM; AZD0156: 0.5 μM; NVP-BEZ235: 50 nM. VE-822 + NVP-BEZ235: VE-822 plus NVP-BEZ235 treatment; AZD0156 + NVP-BEZ235: AZD0156 plus NVP-BEZ235 treatment. *