Aldo-keto reductase 1B10 as a Carcinogenic but Not a Prognostic Factor in Colorectal Cancer

Abstract

Colorectal cancer (CRC) is the leading cause of cancer death, but little is known about its etiopathology. Aldo-keto reductase 1B10 (AKR1B10) protein is primarily expressed in intestinal epithelial cells, but lost in colorectal cancer tissues. This study revealed that AKR1B10 may not be a prognostic but an etiological factor in colorectal tumorigenesis. Using a tissue microarray, we investigated the expression of AKR1B10 in tumor tissues of 592 colorectal cancer patients with a mean follow-up of 25 years. Results exhibited that AKR1B10 protein was undetectable in 374 (63.13%), weakly positive in 146 (24.66%), and positive 72 (12.16%) of 592 tumor tissues. Kaplan-Meier analysis showed that AKR1B10 expression was not correlated with overall survival or disease-free survival. Similar results were obtained in various survival analyses stratified by clinicopathological parameters. AKR1B10 was not correlated with tumor T-pathology, N-pathology, TNM stages, cell differentiation and lymph node/regional/distant metastasis either. However, AKR1B10 silencing in culture cells enhanced carbonyl induced protein and DNA damage; and in ulcerative colitis tissues, AKR1B10 deficiency was associated acrolein-protein lesions. Together this study suggests that AKR1B10 downregulation may not be a prognostic but a carcinogenic factor of colorectal cancer.

Keywords: AKR1B10; Biomarker; Colorectal cancer; Tissue microarrays; and DNA damage.

Figure 1

AKR1B10 expression in colorectal cancer tissues. Immunohistochemistry and evaluation of AKR1B10 protein expression were conducted as described in Materials and Methods. A) AKR1B10 expression at different levels. Data show the representative images. Arrows denote the weakly positive or positive staining. Scale bar, 50μm. B) Statistical summary of AKR1B10 expression in colorectal tissues. C) Databases: AKR1B10 expression in colorectal cancer. Data were from public microarray data sets, GSE4107 dataset (normal controls=10; CRC tissues=12) and GSE8671 dataset (normal controls=32; CRC tissues=32).

Figure 1

AKR1B10 expression in colorectal cancer tissues. Immunohistochemistry and evaluation of AKR1B10 protein expression were conducted as described in Materials and Methods. A) AKR1B10 expression at different levels. Data show the representative images. Arrows denote the weakly positive or positive staining. Scale bar, 50μm. B) Statistical summary of AKR1B10 expression in colorectal tissues. C) Databases: AKR1B10 expression in colorectal cancer. Data were from public microarray data sets, GSE4107 dataset (normal controls=10; CRC tissues=12) and GSE8671 dataset (normal controls=32; CRC tissues=32).

Figure 2

Correlation between AKR1B10 expression and patient survival. Patients were sorted into three groups: AKR1B10 negative, weakly positive (+1), and strongly positive (+2). Kaplan-Meier curves were used for survival analyses of patients. A) Overall survival; B) Disease-free survival.

Figure 3

Stratified survival analysis. Patients were stratified by key clinicopathological factors and then subjected to Kaplan-Meier survival analyses. A) Stratified by clinical stages, B) Stratified by tumor cell differentiation, and C) Stratified by lymph node metastasis. AKR1B10 expression had not correlation with patient survival in any stratified analyses.

Figure 4

Acrolein-protein lesions induced by AKR1B10 deficiency. AKR1B10 silencing, Western blot, and immunocytochemistry were conducted as described in Materials and Methods. A) siRNA-mediated AKR1B10 silencing. B) Western blot for acrolein-adducted proteins in AKR1B10 silencing cells. C) Immunocytochemistry for acrolein-adducted proteins in AKR1B10 silencing cells. Arrows denote acrolein adducts of proteins in the cells. D) AKR1B10 expression in. Data were from public microarray datasets, GSE48958 dataset (normal controls=8; UC tissues=13) and GSE14580 dataset (normal controls=6; UC tissues=24). E) Western blot for acrolein-adducted proteins in ulcerative colitis tissues. N, normal tissue; UC, ulcerative colitis. †, paired specimens.

Figure 5

DNA damage induced by AKR1B10 deficiency. AKR1B10 silencing, DNA mutation assays, and comet assays were conducted as described in Materials and Methods. A) siRNA-mediated AKR1B10 silencing. B) DNA mutations by HPRT selection in AKR1B10 silencing cells. C) DNA breaks by comet assays in AKR1B10 silencing cells. HPRT, hypoxanthine-guanine phosphoribosyl transferase. Arrows denote comet tails of damaged DNA.