Lysosomal-Associated Protein Transmembrane 4 Beta-35 Overexpression Is a Novel Independent Prognostic Marker for Gastric Carcinoma

Objective The purpose of this work was to analyze the relationships between the expression status of Lysosomal-associated protein transmembrane-4 beta 35 (LAPTM4B-35) in cancerous tissues and clinicopathological characteristics and prognosis of the patients with gastric carcinoma (GC). Methods The GC samples from 157 patients in a discovery cohort and 148 patients in a testing cohort with follow-up data were used to validate the feasibility of expression of LAPTM4B-35 protein in predicting GC prognosis. Immunohistochemical staining was used to determine the expression of LAPTM4B-35 protein in precancerous gastric lesions and gastric carcinomas. The correlation between the expression of LAPTM4B-35 and clinicopathologic characteristics of patients with gastric carcinoma was analyzed using chi-square test. Univariate and multivariate analyses were performed to determine the association between LAPTM4B-35 expression and prognosis. Results LAPTM4B-35 expression was increased steadily in sequential stages of precancerous gastric lesions. Positive LAPTM4B-35 expression was more frequently detected in patients with distant metastasis (P = 0.023) and III+IV TNM stages (P = 0.042) in the discovery cohort. Kaplan-Meier survival curves and univariate analysis showed that expression of LAPTM4B-35 had a significant impact on overall survival of patients with gastric carcinoma in discovery cohort (P<0.001) and testing cohort (P = 0.001). LAPTM4B-35 expression was an independent prognostic indicator for the overall survival of patients with gastric carcinoma in both cohorts. Conclusions The present research demonstrated that LAPTM4B-35 over-expression was an independent factor in gastric carcinoma prognosis. LAPTM4B gene may be a useful target of interventions slowing the progression of precancerous gastric lesions and a new therapy method to improve the prognosis of gastric carcinoma.


Introduction
Gastric carcinoma (GC) was a very common cancer worldwide with high mortality rate. Over 70% of new GC cases and deaths occurred in developing countries, especially in East Asia. Diagnosed at later stages and accepted inappropriate therapy were main causes of the high mortality rate of GC [1]. Molecular and genetic alterations underlying the initiation, progression and metastasis of GC made it possible to find effective markers to predict the progression and prognosis of precancerous gastric lesions and GC [2,3]. According to these researches, interventions to slow the progression of precancerous gastric lesions and appropriate therapeutic facilities and drugs applied according to these researches might reduce the incidence of GC and improve the prognosis of GC. But the exact molecular mechanisms underlying gastric carcinogenesis and GC progression were not fully understood until now.
In the present research, we detected LAPTM4B-35 expression status in precancerous gastric lesions and gastric carcinomas by immunohistochemical staining. The purpose of our study was to investigate the relationships between expression of LAPTM4B-35 and the clinicopathological characteristics and prognosis of the patients with GC. We hypothesize LAPTM4B may be a useful marker to predict the progression of precancerous gastric lesions and the prognosis of patients with GC.

Materials and Methods Patients
We collected a discovery cohort including 157 patients from the Affiliated Hospital of Binzhou Medical University between 2004 and 2007, and a testing cohort including 148 patients from the Yantai Affiliated Hospital of Binzhou Medical University between 2003 and 2007. All patients were diagnosed with gastric adenocarcinoma and received radical gastrectomy in the Surgical Department. There were 119 males and 38 females with a mean age of 57.8 years (range, 31-78 years) in discovery cohort, and there were 98 males and 50 females with a mean age of 57.6 years (range, 25-82 years) in testing cohort. The clinicopathological features of patients in two cohorts, including age, sex, tumor size, histopathological differentiation, TNM staging, Lauren type, vessel permeation, lymph node metastasis and distant metastasis were summarized in Table 1 and Table 2. None of the patients received systemic chemotherapy, immunotherapy or radiotherapy before and after surgery. Normal gastric mucosa (10 cases), mucosa diagnosed as chronic atrophic gastritis without intestinal metaplasia and dysplasia (10 cases), chronic gastritis with intestinal metaplasia (10 cases) and chronic gastritis with dysplasia (10 cases) were obtained from gastroscope biopsy samples. All patients consented in writing to participate in this study. This project was approved by the Ethics Committee of Binzhou Medical University before performance.

Specimen characteristics
Tumor specimens were obtained at the time of surgery. A small sample of carcinoma tissue was collected and frozen in liquid nitrogen. All the tissue blocks got from specimens were fixed in 10% neutral buffered formalin solution, embedded in paraffin and stored at room temperature. 4μm thick sections were prepared for immunohistochemical staining. Four frozen fresh GC tissue samples were used in the Western blot analysis.

Assay Methods
Immunohistochemical staining. Immunohistochemistry was used to detect the presence of LAPTM4B-35 proteins in specimens. The prepared 4μm sections were deparaffinized in xylene, rehydrated in a graded ethanol (100%, 95%, 80% and 75%) and washed in phosphate buffered saline (PBS). Endogenous peroxidase activity was blocked by incubation for 10 minutes in 3% H 2 O 2 . After antigen retrieval performed in an autoclave for 3 minutes (10 mM citrate buffer, pH 6.0), the sections were incubated overnight at 4°C with the primary antibody at a dilution of 1:50. The rabbit anti-human polyclonal antibody (named LAPTM4B-N 1-99 -pAb, against the N-terminus of LAPTM4B-35) [27], was kindly provided by Prof. Rouli Zhou. Washed with PBS for three times, the polyperoxidase-anti-rabbit IgG (Polymer Detection System for Immuno-Histological Staining, GBI, Mukilteo, WA, USA) was linked to the primary antibody. Colour development was performed by presenting peroxidase activity with 3,3'-daiminobenzidine tetrahydrochloride (DAB). The slides were counterstained with haematoxylin and viewed under light microscope. Positive and negative controls were routinely used. Normal nonimmunone rabbit serum was used to replace primary antibody to confirm that the brown signal was presented for the immunoreactivity between special antibody and antigen.
Immunohistochemical staining assessment and scoring. The brown signal of LAPTM4B-35 staining was mainly located in the cytoplasm or on the cell membrane. The extent of LAPTM4B-35 staining was divided to three grades. 0: no brown signal or the brown signal was found in less than 10% epithelial cells or cancer cells on the slides; 1: 11*50% epithelial cells or cancer cells were observed with brown signal; 2: brown signal was found in more than 50% epithelial cells or cancer cells. The intensity of LAPTM4B-35 staining was divided into three grades too: no staining = 0; weak = 1; strong = 2. The expression of LAPTM4B-35 for each slide was finally separated into two groups (negative group and positive group) according to the extent grade plus the intensity grade. Samples with staining score of 0 or 1 were defined as "negative" and samples with staining score of 2, 3 or 4 were defined as "positive". All the slides were examined by three experienced pathologists, blinded to clinical data.
Western blotting analysis. Frozen tumor specimens were thawed and homogenized in lysis buffer containing 1% Triton X-100 with protease inhibitor. The mixtures were centrifuged at 12,000g for 20 min at 4 C, and the supernatant was obtained. The resulting proteins (50μg per lane) were separated in a 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel (PAGE) and transferred to a PVDF membrane (Millipore, Bedford, MA, USA). Membranes were blocked with 5% nonfat dry milk in Tris-buffered saline (TBS) and then incubated overnight with rabbit anti-human LAPTM4B-35 polyclonal antibodies (LAPTM4B-N 1-99 -pAb, 1:300 dilution) at 4°C. After washing with Tris-buffered saline with Tween (TBST), the PVDF membrane was incubated with diluted horseradish peroxidase-conjugated goat anti-rabbit IgG for 1h at room temperature. The blots were detected using the Amersham ECL (Enhanced Chemiluminescence) Western Blotting System. Membranes were then washed with stripping solution for 1h and treated as described above but with rabbit anti-human β tubulin polyclonal antibodies (1:3,000 dilution; Santa Cruz biotechnology, Santa Cruz, CA, USA) as an internal control.

Study design
We retrospectively analysed tumor samples from patients with primary invasive gastric carcinoma. All patients were followed up until December 2013 with a minimum 5 years. Survival time was defined to be the period of time in months from the date of diagnosis to the date of death from GC.

Statistical Analysis Methods
Statistical analyses were performed with SPSS 19.0 software. Pearson chi-square test was used to compare the distributions of clinicopathologic characteristics of cases with positive expression of LAPTM4B-35 and cases with negative expression of LAPTM4B-35. Linear-by-Linear Association chi-square test was used to assess trends in LAPTM4B-35 expression by gastric disorders. Overall survival was analyzed by the Kaplan-Meier method. Log-rank test in univariate analysis was used to estimate different impacts on survival between cases with positive LAPTM4B-35 expression and those with negative LAPTM4B-35 expression. The independent influence of each variable on overall survival was analyzed by the multivariate analysis of Cox proportional hazard model. The analyses were adjusted simultaneously for the age, sex, tumour size, differentiation, TNM stage, vessel permeation and lymph nodes metastasis. We used a stepwise variable selection procedure to identify independent factors prognostic for survival, and variables were removed using backward elimination according to a selection stay criterion of 0.05.

Results
Expression of LAPTM4B-35 in normal gastric mucosa, chronic atrophic gastritis, intestinal metaplasia, dysplasia and gastric carcinoma by immunohistochemical staining  Fig. 1. The expression of LAPTM4B-35 increased steadily from 0 in normal gastric mucosa and chronic atrophic gastritis (without intestinal metaplasia or dysplasia) to 30% in chronic gastritis with intestinal metaplasia, 55% in intestinal metaplasia beside cancer tissues, 60% in dysplasia, peaking at 90.4% in GC ( Table 3).

Expression of LAPTM4B-35 in gastric carcinoma tissues by Western blot analysis
Western blot analysis was performed to investigate LAPTM4B-35 protein expression in four frozen GC tissue samples. In all samples, LAPTM4B-35 protein was detected as an approximately 35-kD band (Fig. 2). The expression intensity of LAPTM4B-35 protein observed in Western blot analysis using fresh tissue was consistent with the staining score in immunohistochemical analysis using paraffin section from the same patient.

Expression of LAPTM4B-35 was in relation to clinicopathologic features of gastric carcinoma
The negative and positive expression categories and the clinicopathologic characteristics of GC in two cohorts were summarized in Table 1 and Table 2. Positive LAPTM4B-35 expression was more frequent detected in cases with distant metastasis (P = 0.023) and III+IV TNM stages (P = 0.042) in discovery cohort. However, such association was not observed in the testing

Univariate analysis of prognostic impact of LAPTM4B-35 expression on overall survival in gastric carcinoma
Patients with LAPTM4B-35 positive expression had significantly lower 5-year overall survival rate than those with negative expression in both the discovery cohort (32/142 vs. 12/15, P<0.001) and the testing cohort (40/132 vs. 12/16, P<0.001). Kaplan-Meier survival curves showed that LAPTM4B-35 was the significant variable in the univariate analysis for the impact of LAPTM4B-35 expression on overall survival in discovery cohort and testing cohort (twosided log-rank test, P < 0.001 and P = 0.001 respectively; Fig. 3 and Fig. 4).

Multivariate analysis of prognostic impact of LAPTM4B-35 expression on overall survival in gastric carcinoma
To analyze the independent impact of LAPTM4B-35 expression on overall survival, multivariate analysis by Cox proportional hazards regression model was performed (backward, stepwise). The results revealed that LAPTM4B-35 expression was independent prognostic  indicator for the overall survival in patients with GC in both cohorts. The results were presented in Table 4 and Table 5.

Discussion
In the present research, we found that the expression of LAPTM4B-35 was increased steadily in sequential stages of gastric carcinogenesis. No expression of LAPTM4B-35 was found in normal gastric mucosa and chronic atrophic gastritis without intestinal metaplasia and   Table 3). The precise mechanism underling gastric carcinogenesis was not yet fully understood. Gastric carcinoma especially 'Intestinal-type' carcinoma was developed from sequential stages: chronic gastritis, chronic atrophic gastritis, intestinal metaplasia, dysplasia and gastric carcinoma [28]. Intestinal metaplasia and dysplasia were closely correlated with risks of GC in the multistep process of gastric carcinogenesis [29]. To elucidate molecular and genetic alterations underlying the progression of precancerous gastric lesions might guide us to find markers, such as expression of c-Met [30] and p16 methylation [31], which were valuable in predicting progression of precancerous gastric lesions. Previous reports indicated that LAPTM4B gene over-expression increased cell growth and proliferation [20,21,22,23]. Our present research suggested that LAPTM4B might play important roles in the initial stages of gastric carcinogenesis and was ongoing work continuously. The increased expression of LAPTM4B-35 may predict the progression of gastric lesions and guide doctors to take appropriate interventions slowing the progression of precancerous gastric lesions and reducing the incidence of gastric cancer.
We also found that positive expression of LAPTM4B-35 in GC was more frequent detected in cases with distant metastasis (P = 0.023) and III+IV TNM stages (P = 0.042) in the discovery cohort. The present work was consistent with previous studies in carcinomas derived from breast [7], pancreas [8], ovary [9,10,11], colon [12], liver [13,14,15], extrahepatic bile duct [16], cervix [17], endometrium [18] and gallbladder [19]. Carcinomas with LAPTM4B-35 over-expression may present more invasive characteristics. Several studies performed in cell lines demonstrated that over-expression of LAPTM4B-35 promoted the progression of cancer cells toward highly invasive and metastatic stages via mechanisms involving activation of proto-oncogenes such as c-myc, c-fos and c-jun, upregulation of cell cycle regulators such as cyclin D1 and cyclin E [21,22], resistance to apoptosis, activation of PI3K/AKT signaling pathway [23] and modulating molecules associated with degradation of extracellular matrix [26].
Moreover, our data also demonstrated that the GC patients with positive LAPTM4B-35 expression had significantly lower 5-year overall survival rate than those with negative expression. Multivariate analysis demonstrated that LAPTM4B-35 expression was an independent prognostic factor for overall survival in patients with GC. In previous studies, the same conclusions were got that LAPTM4B-35 was an independent prognostic factor in a wide range of carcinomas [7][8][9][10][11][12][13][14][15][16][17][18][19]. LAPTM4B-35 was over-expressed in most human malignant tumors and precancerous lesions. Hence, we believed that LAPTM4B played important roles in the initiation, progression and metastasis of tumors. In fact, many researchs provide various markers in blood or body tissues for diagnosis, prognosis and therapeutics of GC. Such as CD133 [32], matrix metalloproteinase-9 [33], c-Met [34], et al. We hope to generate accurate diagnoses, prognoses and select the most appropriate therapy based on the markers. However, there are no current excellent markers for GC. Determining the significance of the markers requires an accumulation of practice. Recent studies showed that LAPTM4B had critical roles in multidrug resistance (MDR) via increasing drug efflux by P-glycoprotein (P-gp) to reduce drug concentration in cytoplasm and entry into nucleus, decreasing drug-induced DNA damage and escaping from drug-induced apoptosis [35,36]. It was therefore possible that LAPTM4B-35-knockdown by RNAi may provide a promising novel therapy strategy in LAPTM4B-35 over-expression GC or other cancers.
Recent studies showed that there were two alleles of the LAPTM4B gene, named as LAPTM4BÃ1 and LAPTM4BÃ2. The allele Ã2 of LAPTM4B was found to be the risk factor of the certain tumors, such as breast carcinoma [37], colon carcinoma [38], hepatocellular carcinoma [39] and gastric carcinoma [40], and LAPTM4B Ã2 was associated with poor prognosis of breast carcinoma [37], colon carcinoma [38] and hepatocellular carcinoma [39]. Although the exact molecular mechanisms underlying the function of LAPTM4B in carcinogenesis had not as yet been worked out, LAPTM4BÃ2 may be an oncogene or play an important role in cell cycle control. Therefore, further elucidation of the role of LAPTM4B gene in the GC will require genotyping in additional samples in the future studies.
In conclusion, the present research demonstrated that LAPTM4B-35 was over-expressed in a large proportion of patients with precancerous gastric lesions or GC. There were steadily increasing expression of LAPTM4B-35 in series of precancerous gastric lesions. And a close association between LAPTM4B-35 over-expression and disease progression, poor prognosis in GC was found for the first time. These findings suggested that LAPTM4B may be a useful and critical molecular marker to predict the progression of precancerous gastric lesions and the prognosis of patients with GC. LAPTM4B gene may provide a useful target of interventions slowing the progression of precancerous gastric lesions and a new therapy method to improve the prognosis of patients with GC.