The Expression Of LIN28B And Its Relationship With Prognosis In Colon Carcinoma

Background and objectivesDespite great achievements in surgery, chemotherapy and the development of novel molecular-targeted drugs, such as bevacizumab (Avastin), the incidence of colon cancer continues to increase. Each year, tumours of the colon are responsible for655,000deaths globally.LIN28B is a newly recognized gene related with neoplasms development. It is believed LIN28is expressed in developing tumor and promotor tumor maglinance, e.g. liver cancer and colon cancer. This gene acts as a suppressor of microRNA (miRNA) biogenesis by specifically binding the precursor let-7(pre-let-7), a miRNA precursor.LIN28was initially identified in C. elegans and was shown to be responsible for the timing of development. Lin28induces pluripotency when expressed in somatic fibroblasts along with OCT4, SOX2and KLF4. LIN28B, as a homologue of LIN28, was first cloned from and shown to be overexpressed in human hepatocellular carcinoma cells and clinical samples in2006. LIN28B is a developmentally regulated RNA-binding protein that promotes tumourigenesis by blocking the post-transcriptional processing of the tumour-suppressive pri-/pre-let-7microRNA. The ability of LIN28B to regulate let-7, a bona fide tumour-suppressor, is consistent with its developmental role in regulating cell proliferation and differentiation. Interestingly, several studies revealed that LIN28B itself is also a target of let-7family members, including miR-125b. Besides let-7, LIN28B may also bind the mRNA of intestinal stem cell markers, such as LGR5and PROM1. In summary, LIN28B promotes transformation primarily by suppressing let-7. Competition between LIN28B and let-7may be critical for normal cell biology and may accelerate tumour development once this balance is perturbed. Because let-7functions as a potential growth suppressor in human colon cancer cells, we examined LIN28B expression in colon cancer tissues to determine the balance between LIN28B and let-7expression.LIN28B is hypermethylated in somatic tissues, and its aberrant reactivation may promote tumourigenesis. LIN28B is frequently overexpressed in multiple cancers, especially advanced types such as progressive hepatocellular carcinoma, epithelial ovarian cancer, Wilm’s tumour and germ cell tumours. In addition to being targeted by antecedent microRNAs, LIN28B is activated by several oncogenic pathways. C-/N-Myc induces LIN28B expression in multiple human and mouse tumour models, resulting in let-7repression and cell proliferation. MYCN amplification also results in LIN28B overexpression, and it has been reported that c-Myc is related to sporadic large bowel cancer and familial polyposis coli, implying a role for Lin28in colon cancer.At present, the relationship between LIN28expression and its clinical pathological characters is not clear. Our experiment objective is to check the clinical significance of LIN28B expression in colon cancer, and to investigate the relationship between LIN28B expression and sensitivity of colon cancer cells to oxaliplatin.In order to investigate impact of LIN28B expression to cancer cell sentitivity to oxaliplatin chemotherapy,3colon cell lines SW480、HCT116and Caco2were cultured under provider’s instruction.Basic LIN28B expression in mRNA and protein level in HCT116and SW480colon cancer cell lines was examined by RT-PCR and Western blot respectively. HCT116and SW480cells were selected for use in these experiments due to their low and high level of LIN28B expression, respectively. Oxaliplatin induced concentration-dependent cytotoxicity in these cells. Oxaliplatin half maximal inhibitory concentration (IC50) for the2cell lines was tested with CCK-8method.Further, sh-LIN28B (hampering LIN28B expression) and NC (control plasmid) plasmids were constructed and transfected into the both cells. RT-PCR, Western blot and CCK8were realized again to check LIN28B level and Oxaliplatin IC50for the2type cells respectively. Transwell test was also finished to check immigration of SW480cells after transfection.Methods1. Tumour tissue analysis:Samples of human colon carcinomas and normal colons were obtained from a human colon carcinoma tissue array (CO2161; US Biomax), which contained208adenocarcinomas and8normal colon mucosa samples. Detailed clinical information, including differentiation status and TNM grading, were also provided for each sample.2. An additional cohort of149colon carcinoma samples was collected from specimens that had been surgically resected at Nanfang Hospital from Jan.2003to Dec.2004, according to an Institutional Review Board-approved protocol. Each of these cases was followed, in detail, through June2009.3. Clinical pathological features (TNM grade, differentation, lymph node and distant metastasis, survival time) for every sample were recorded clearly. We confirmed the TNM grade for samples from both cohorts. In total,357colon cancer tumours and8normal colon samples were used for the immunohistochemistry analysis.4. Immunohistochemistry:LIN28B expression in samples above was checked with immunohistochemistry method. Four-micrometer thick sections of the paraffin-embedded tissue blocks were cut and mounted on poly-lysine coated slides. They were dewaxed in xylene and rehydrated through a graded series of ethanol. After deparaffinization, antigen retrieval treatment was performed at120℃(autoclave) for5min in a10nmol/l sodium citrate buffer (pH6.0). Endogenous peroxidase activity was blocked by using a0.03%hydrogen peroxide solution containing sodium azide at room temperature for30min. Next, tissues were incubated in LIN28B (1:10; Cell Signaling, Boston, MA) primary antibodies diluted in PBT (10X PBS,10%BSA,10%Triton X-100in ddH2O) at4℃overnight. Subsequently, a thorough washing in a0.01M phosphate-buffered saline (PBS) solution was performed. The binding sites of the primary antibody were visualized using a Dako EnVison kit (Dako, Denmark A/S) in accordance with the manufacturer’s instructions. Finally, sections were faintly counterstained with hematoxylin and mounted with glycerol gelatin. Two experienced pathologists blindly analyzed all sections under a light microscope. Based on the estimated percentage of positive cells and immunostaining intensity that was determined by comparison with immunoreactivity of3positive controls in each experiment, staining results were divided into4categories:0was tissue specimens without staining;1:tissue specimens with less than25%of cancer tissue and/or weakly stained;2:tissue specimens with25%to50%of cancer tissue and/or moderately stained; and3:tissue specimens with more than50%of cancer tissue and/or strongly stained.5. The standard chisquare test was used to check correlation between LIN28B expression and its clinical pathological features.6. A Log-rank test was performed to compare the survival and recurrence distributions for the various intensity groups (0-2vs.3). The correlation between staining intensity and survival or recurrence was determined according to a chi-square analysis, where p values less than0.05were considered to be statistically significant. A95%confidence interval was calculated for confirmation of statistical significance.7. Cell culture:In order to investigate impact of LIN28B expression to cancer cell sentitivity to oxaliplatin chemotherapy,3colon cell lines HCT116, SW480and Caco2were cultured under provider’s instruction.8. Cell transfection:A LIN28B-specific shRNA plasmid and a negative control plasmid (NC) were constructed by Genepharma. LIN28B-and GAPDH-specific primers were synthesised by TaKaRa. DNA oligoribonucleotides were transfected using Lipofectamine2000(Invitrogen), according to the manufacturer’s instructions. Between50and100ng of the DNA oligoribonucleotides were used for each transfection.9. Reverse-transcription polymerase chain reaction (RT-PCR):Total RNA was extracted using TRIzol reagent and was then treated with DNase Ⅰ. The PCR amplification procedure for GAPDH was as follows:94℃for3min;30cycles of94℃for30s,60℃for30s and72℃for30s; and a terminal elongation at72℃for5min. The amplification of LIN28B was performed as follows:95℃for2min;30cycles of95℃for30s,58℃for30s and72℃for30s; and a final step at72℃for5min.10. Western blot:Basic LIN28B in mRNA and protein expression level was examined in colon cancer lines by Western blot.11. Migration analysis:SW480cells (1×104cells in100μl serum-free medium), which had been transfected with the NC or sh-LIN28B plasmids, were placed in the top chamber of Transwell culture dishes. The lower chamber was filled with600μL of conditioned medium. After24hours, the cells that had not migrated to the lower chamber were removed from the upper surface of the Transwell membrane with a cotton swab. Migrated cells on the lower membrane surface were fixed, stained with0.1%crystal violet and imaged.12. Cell viability analysis:SW480or HCT116cells, which had been transfected with either the NC or sh-LIN28B plasmid, were seeded in triplicate in96-well plates at a concentration of5000cells per well. After the cells had adhered to the plates, they were incubated with different concentrations of oxaliplatin (e.g.100,10,1,0.1,0.01and0.001μg/ml) for4hours and were then transferred to complete medium for an additional20hours. At the indicated time point,10μL of CCK-8solution was added to each well, and the cells were incubated for3h at37℃. The absorbance was recorded at450nm using a plate reader. The experiment was performed in triplicate, and the cell inhibition ratio was determined using the following equation:(1-test group A/control group A)× 100%. The IC50(50%inhibiting concentration) value was calculated using specific software.Results1. To examine the potential roles for LIN28B in colon cancer, we first compared LIN28B protein expression between208tumour samples and8non-matched normal mucosa samples from a tissue microarray using IHC. LIN28B is significantly overexpressed in colon tumour tissues. IHC showed that LIN28B was markedly upregulated in tumour tissues compared with the normal mucosa, which demonstrated very little LIN28B expression. Results from tissue microarrays and clinical samples demonstrate that LIN28B was upregulated in colon cancer tissue compared to normal mucosa.2. Microarray colon tumour data were analysed using chi-square tests to compare the pathological characteristics and distribution of samples with low or high staining intensities (LIN28B staining with scores of0-2was considered low intensity and staining with a score of3represented high intensity). Our results indicated that greater LIN28B expression correlated with advanced tumour characteristics, such as poor differentiation, TNM grades Ⅲ-Ⅳ and the presence of lymph node and distant metastases (p<0.001), whereas lower LIN28B expression was correlated to well-to-moderate levels of differentiation and TNM grades Ⅰ-Ⅱ. These results were also reproducible in the cohort of surgically resected tumour sample.3. LIN28B overexpression correlated with reduced patient survival and an increased likelihood of tumour recurrence. Higher LIN28B staining intensity from stage Ⅰ,Ⅱ and Ⅲ colon cancers correlated with reduced patient survival. High LIN28B expression was related to a higher probability of tumour recurrence (p<0.001). Data from patients who underwent surgery (TNM Ⅰ-Ⅲ) were further analysed via log rank tests to investigate the influence of LIN28B overexpression on patient survival and recurrence. This analysis revealed a correlation between low-intensity LIN28B staining from samples of TNM grade Ⅰ and Ⅱ tumours and increased patient survival (p<0.05) and a lower likelihood of tumour recurrence (p<0.05).4. We sought to determine whether LIN28B could influence chemosensitivity to oxaliplatin. We examined LIN28B expression in three colon cancer cell lines using RT-PCR and Western blotting. HCT116and SW480cells were selected for use in these experiments due to their low and high level of LIN28expression, respectively. Oxaliplatin induced concentration-dependent cytotoxicity in these cells.5. sh-LIN28B efficiently inhibited the expression of LIN28B in HCT116and SW480cells. RT-PCR and Western blot analyses confirmed the efficiency of sh-LIN28B knockdown in HCT116and SW480cells.6. ICso values were calculated after curve fitting using the XLfit software. HCT116and SW480cells were transfected with NC or sh-LIN28B plasmids24h prior to oxaliplatin (IC50value) treatment. The inhibitory rate, normalised to NC, was calculated using CCK-8absorbance at the indicated time points. The IC50value for oxaliplatin was6.09±1.42μg/mL for the HCT116cells, and this was increased by30%to8.18±3.64μg/mL in the SW480cells. This finding suggested that differences in LIN28B expression may serve to modulate chemosensitivity in colon cancer cells. To verify this hypothesis, we constructed a LIN28B-specific shRNA-expressing plasmid and control vector. The efficiency of the sh-LIN28B construct was determined in both cell lines, and then a CCK-8analysis was performed to explore the interactions between the sh-LIN28B plasmid and oxaliplatin in HCT116and SW480cells. The results indicated a synergistic effect between sh-LIN28B and oxaliplatin, which suggests that the targeting of LIN28B may be capable of sensitising colon cancer cells to oxaliplatin therapy.7. Transwell:As LIN28B overexpression was correlated with patient survival, the occurrence of lymph node/distant metastases and tumour recurrence, we next sought to explore whether LIN28B expression influences the migration of colon cancer cells. We knocked down LIN28B expression using shRNA, and the Transwell analysis indicated that suppression of LIN28B significantly inhibited the migration of SW480cells. Silencing of LIN28B suppressed the migration of SW480cells. Conclusions1. LIN28B overexpression correlates with aggressive pathologic characteristics of colon carcinoma, and it may serve as a prognostic factor of colon cancer patients.2. Silencing the expression of LIN28B in colon cancer cells could greatly inhibit its migration and promote its chemosensitivity on oxaliplatin. Therefore, LIN28B could be a novel target for clinical therapy drug.Innovative Points1. We found that LIN28B was upregulated in colon cancer tissue compared to normal mucosa, and its overexpression correlated with reduced patient survival and increased tumour recurrence. LIN28B was also related to tumour clinical staging, pathology grade and the occurrence of lymph node and distant metastases.2. LIN28B suppression inhibited the migration of SW480colon cancer cells and facilitated the cytotoxicity induced by oxaliplatin in SW480and HCT116colon cancer cells.3. LIN28B overexpression contributes to colon tumourigenesis, and LIN28B may serve as a diagnostic tool and therapeutic target for colon cancer.