Abstract
Lysosome-associated membrane protein 3 belongs to the lysosome-associated membrane glycoprotein family, which is associated with lymph node, metastasis, poor overall survival, and resistance to chemotherapy and radiotherapy. Epithelial ovarian cancer is one of the most deadly global female gynecologic malignant tumors. Its clinical outcome is poor and most epithelial ovarian cancer patients tend to relapse because of drug resistance. However, lysosome-associated membrane protein 3 expression in epithelial ovarian cancer and its relationship between clinicopathologic factors remain poorly understood. To clarify the prognostic implications of lysosome-associated membrane protein 3 in epithelial ovarian cancer, we analyzed both messenger RNA and protein levels of lysosome-associated membrane protein 3 in ovarian carcinomas. Polymerase chain reaction results showed higher expression of lysosome-associated membrane protein 3 messenger RNA in epithelial ovarian cancer than in noncancerous tissues. Immunohistochemical results showed that high lysosome-associated membrane protein 3 cytoplasmic expression was significantly related to tumor grade (p = 0.038), lymph node metastasis (p = 0.049), metastasis (p < 0.001), level of CA125 (p = 0.030), and International Federation of Gynecology and Obstetrics (FIGO) (p < 0.001). High lysosome-associated membrane protein 3 nuclear expression was significantly associated with tumor grade (p = 0.046), tumor single or double (representative whether the tumor involving one or both ovaries) (p = 0.016), metastasis (p < 0.001), and FIGO stage (p < 0.001). Survival analysis indicated that high lysosome-associated membrane protein 3 cytoplasmic expression (hazard ratio: 4.632, 95% confidence interval: 2.421–8.864; p < 0.001), patients’ age (hazard ratio: 1.729, 95% confidence interval: 1.027–2.911; p = 0.039), and FIGO stage (hazard ratio: 2.049, 95% confidence interval: 1.113–3.774; p = 0.021) were significantly correlated with poor survival outcome of epithelial ovarian cancer patients.
Keywords
Introduction
The incidence of ovarian cancer, the deadliest malignancy among women, has been increasing persistently in the world, including China.1,2 More than 200,000 new cases of epithelial ovarian cancer (EOC) occur worldwide every year. 3 The majority of EOC patients are diagnosed at advanced stages because of asymptomatic characteristics and lack of sensitive detection at early stage. 4 Normally, platinum–taxane combination chemotherapy is an essential strategy in treating ovarian cancer before or after surgery. 5 However, most patients relapse because of drug resistance.6,7 So, nowadays it is remarkably important to find novel and specific biomarkers of prognostic significance for EOC patients.
Lysosome-associated membrane protein 3 (LAMP3), one of the members of lysosome-associated membrane glycoprotein family, mainly residing in lysosomal membranes, is an integral membrane protein. LAMP3 (DC-LAMP, TSC403, CD208), originally regarded as a lung-specific gene and clarified as a molecular marker for mature dendritic cells as well, was observed to be overexpressed in several kinds of malignancies including ovarian cancer. 8 Previous studies have observed that higher LAMP3 protein expression in both colorectal and gastric cancers is significantly associated with advanced tumor stage and poorer overall survival. 9 Other studies have shown that LAMP3 protein expression was extremely high in esophageal squamous cell carcinoma (ESCC) and is an independent prognostic factor for ESCC. 10 Mujcic et al. 11 found in cervix cancer that LAMP3 protein expression has been related with metastasis, hypoxia, and poor prognosis partly by PERK/eIF2a signaling pathway. Some researches suggest that PERK/elfa has involved in hypoxia resistance and tumor growth. Nagelkerke et al. 12 found in breast cancer that LAMP3 protein expression was established to be of prognostic relevance and was related with locoregional recurrence, tumor grade, size, lymph node, resistance to therapy, and steroid hormone receptor status. What’s more, previous information suggested that LAMP3 expression was related with resistance to chemotherapy and radiotherapy.13,14 Therefore, recent studies suggest that LAMP3 may serve as an important marker for tumor metastasis and prognostic factor.
However, the role of LAMP3 in EOC patients remains unclear. In this study, we focused on the expression of LAMP3 protein in both normal ovarian and fallopian tube and malignant samples in tissue microarrays (TMAs) by using immunohistochemistry (IHC) and LAMP3 messenger RNA (mRNA) expression by reverse transcription polymerase chain reaction (RT-PCR). What’s more, we analyzed the potential contribution of this molecule expression level in EOC in the prediction of prognosis.
Methods and materials
Patients’ samples and TMA analysis
We used 205 formalin-fixed paraffin-embedded (FFPE) tissue blocks consisting of 15 normal ovarian tissues, 20 normal fallopian tube tissues, 20 benign ovarian tumors, 15 borderline ovarian tumors, and 135 EOC tissues. As for 20 benign ovarian tumors, there were 14 serous and 6 mucinous cystadenoma. As for 15 borderline ovarian tumors, there were 10 serous and 5 mucinous borderline tumors. A total of 135 specimens from primary EOC patients who underwent surgery between 2005 and 2010 were obtained from Clinical Bio-Bank of Affiliated Hospital of Nantong University and included in the present retrospective study. Of the 135 cases of EOC, there were 104 serous carcinoma and 31 other types (14 of endometrial carcinoma, 4 clear-cell carcinoma, 6 mucinous carcinoma, 2 transitional cell carcinoma, and 5 adeno-squamous carcinoma). As for FIGO stage, there were 67 stage I and 68 stage II–IV cases. In view of histological grading, 105 cases were of high grade and 30 cases were of low grade. Detailed clinicopathological data are shown in Table 1.
Association of LAMP3 expression with clinical characteristics of EOC.
LAMP3: lysosome-associated membrane protein 3; EOC: epithelial ovarian cancer.
Representative whether the tumor involving one or both ovaries.
p < 0.05.
The significance of bold values was to illustrate the LAMP3 expression in EOC was related to metastasis and FIGO stage.
Clinical data of the patients (including age, tumor classification, tumor grade, follow-up including 5-year survival and other information) were recorded in detail, which was obtained from the medical records of patients. None of the patients received adjuvant chemotherapy, radiation therapy, or immunotherapy. All patients underwent standard surgery, and none of the patients received any therapy such as chemotherapy, radiation, or immunotherapy prior to the operation. After surgery, a platinum-based chemotherapy was adopted for at least six cycles per patient. All patients were diagnosed by at least two pathologists and followed up from the time of diagnosis through 31 December 2015. Ethical approval was obtained from the Ethics Committee of Nantong University Affiliated Hospital.
LAMP3 mRNA isolation and real-time RT-PCR assay
We used 60 fresh frozen tissue samples in the study, which included 15 cases of normal ovarian tissue, 15 cases of normal fallopian tube tissue, 10 cases of benign ovarian tumors, 10 cases of borderline ovarian tumors, and 10 cases of EOC. Total mRNA was isolated from EOC and normal samples by using TRIzol reagent (Life Technologies New York, USA). The expression of LAMP3 was determined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using the LightCycler FastStart DNA Master SYBR Green I Kit (Roche Diagnostics, Tokyo, Japan). According to the manufacturer’s instructions, we used the following primers: F, 5′-TGAAAACAAC CGATGTCCAA-3′ and R, 5′-TCAGACGAGCACTCA TCCAC-3′, which were described specifically by Anika Nagelkerke et al. 14 The data were obtained by normalizing LAMP3 gene Ct (cycle threshold) values with corresponding β-actin Ct and then analyzed with 2−ΔΔCt Ct method. The primer sequences of β-actin was as follows: F, 5′-CCTCTATGCCAACACAGTGC-3′ and R, 5′-CCTGCT TGCTGATCCACATC-3′.
TMA construction and IHC analysis
The TMA system (Quick-Ray, UT06, UNITMA, Korea) was used for IHC analysis. Tissue sections were deparaffinized and rehydrated in graded ethanol solutions. The sections were incubated in 3% H2O2 to block the activity of endogenous peroxidase. Tissue sections were then boiled under pressure in citrate buffer (pH 6.0) for 5 min and incubated for 1 h with rabbit polyclonal anti-human LAMP3 antibody (dilution 1:100) (ab111090, Abcam, Cambridge, UK). Not knowing about the clinical background of the cases, two independent pathologists evaluated and scored the staining intensity of LAMP3 expression of each slide. Staining intensity was graded on a scale as follows: 0 (negative), 1(weak), 2 (moderate), and 3 (strong). Positivity of cell staining was scored as a percentage (0%–100%). The final LAMP3 staining score was a product of the percentage and intensity score. The cutoff point for the LAMP3 expression score (both cytoplasm and nucleus staining) that was statistically significant in terms of overall survival was performed using the X-tile software program (The Rimm Lab at Yale University; http://www.tissuearray.org/rimmlab) as described previously. 15 We defined the staining scores using a two-level grading system for both cytoplasm and nucleus staining: 0–100: low expression; 101–300: high expression.
Statistical analysis
Statistical calculations were carried out using SPSS 20 statistics software (SPSS Inc., Chicago, IL, USA) and STATA 12.0 (StataCorp, College Station, TX, USA). χ2 tests were performed to evaluate relationships between clinicopathological factors and LAMP3 expression. Statistical calculations of LAMP3 mRNA expression were performed using t-test. Patient survival curves were plotted using the Kaplan–Meier method. Univariate and multivariate Cox regression models were adopted to evaluate prognostic significance. For all analyses, we use p < 0.05 of statistical significance.
Results
LAMP3 mRNA expression in EOC by PCR
To evaluate the LAMP3 mRNA expression between EOC and normal tissue, RNA was isolated from 10 cases of ovarian carcinoma, 15 normal ovary and 15 normal fallopian tube tissue, 10 benign ovarian tumor, and 10 borderline ovarian tumor using real-time polymerase chain reaction (PCR). As shown in Figure 1, the mean of LAMP3 mRNA in carcinoma and in noncancerous tissues (ovary and fallopian tube) was 0.354 ± 0.023, 0.135 ± 0.018, 0.135 ± 0.044, 0.134 ± 0.078, and 0.136 ± 0.085, respectively. PCR results showed higher expression of LAMP3 mRNA in ovarian cancer samples than in noncancerous tissues (all p < 0.05).
LAMP3 mRNA expression in EOC and in noncancerous tissues. PCR results showed that the expression of LAMP3 mRNA in EOC and in noncancerous tissues (ovary and fallopian tube) were 0.354 ± 0.023, 0.135 ± 0.018, 0.135 ± 0.044, 0.134 ± 0.078, and 0.136 ± 0.085, respectively. The expression of LAMP3 mRNA in ovarian cancer samples was significantly higher than in noncancerous tissues (all p < 0.05).
LAMP3 protein level in EOC tissues and normal and benign ovarian tissues by IHC
We determined LAMP3 protein expression in 205 archived ovarian tissue blocks, including 15 normal ovarian tissues, 20 normal fallopian tube tissues, 20 benign ovarian tumors, 15 borderline ovarian tumors, and 135 EOC tissues. Positive staining was localized mainly in the cytoplasm and nucleus of EOC cells. High LAMP3 cytoplasmic expression was detected in 72 (53.33%) cases of EOC. High LAMP3 nuclear expression was seen in 35 (25.93%) cases. Typical IHC staining patterns for LAMP3 protein expression are manifested in Figure 2.
LAMP3 protein level in EOC tissues and normal and benign ovarian tissues by IHC. (a) Strong IHC staining of LAMP3 in EOC samples; (b) weak IHC staining of LAMP3 in a borderline ovarian tumor sample; (c) negative IHC staining of LAMP3 in a benign ovarian tumor; (d) negative IHC staining of LAMP3 in normal ovarian tissue; (e) negative IHC staining of LAMP3 in normal fallopian tube tissue. Original magnification ×40 in a, b, c, d, e (scale bars 500 µm); ×400 in a1, b1, c1, d1, e1 (scale bars 50 µm).
LAMP3 expression was not detected in normal or benign ovarian tumor tissues (Table 2). The level of LAMP3 expression (both cytoplasmic and nuclear) in EOC was significantly higher than in normal and benign ovarian tissues (χ2 = 57.544, p = 0.001 and χ2 = 15.541, p = 0.004, respectively).
LAMP3 immunohistochemical staining of protein in normal ovarian, normal fallopian tube, benign ovarian tumor, borderline ovarian tumor, and EOC tissues.
p < 0.05.
The significance of bold values was to illustrate the LAMP3 expression in EOC was related to metastasis and FIGO stage.
Association between LAMP3 expression and survival
Next, we correlated LAMP3 protein expression with clinicopathological features of 135 EOC (Table 1). High LAMP3 cytoplasmic staining was related to grade (p = 0.038), lymph node metastasis (p = 0.049), metastasis (p = 0.001), CA125 (p = 0.030), and FIGO stage (p = 0.001). High LAMP3 nuclear expression was related to grade (p = 0.046), single or double (representative whether the tumor involving one or both ovaries) (p = 0.016), metastasis (p = 0.001), and FIGO stage (p = 0.001). Finally, we analyzed several known predictive factors of poor outcome in EOC using both univariate and multivariate analysis (Table 3).
Univariate and multivariable analysis of prognostic factors in cancer for 5-year survival.
HR: hazard ratio; CI: confidence interval.
Representative whether the tumor involving one or both ovaries.
p < 0.05.
In univariate analysis, LAMP3 protein cytoplasmic overexpression (hazard ratio (HR): 0.367, 95% confidence interval (CI): 0.186–0.723; p = 0.004) and nuclear overexpression (HR: 5.662, 95% CI: 3.098–10.350; p = 0.001) were significantly associated with 5-year survival. Other prognostic factors such as age (HR: 2.032, 95% CI: 1.255–3.289; p = 0.004), grade (HR: 2.230, 95% CI: 1.104–4.506; p = 0.025), single or double (HR: 1.772, 95% CI: 1.093–2.872; p = 0.02), lymph node metastasis (HR: 2.005, 95% CI: 1.162–3.458; p = 0.012), metastasis (HR: 4.399, 95% CI: 2.606–7.427; p = 0.001), ascites cell (HR: 1.887, 95% CI: 1.103–3.226; p = 0.020), and FIGO stage (HR: 3.748, 95% CI: 2.199–6.388; p = 0.001) were also statistically significant.
Univariate prognostic factors were included in the multivariable analysis together, but FIGO stage already included single or double, ascites, ascites cell, lymph node metastasis, and metastasis; after the removal of these factors, the rest were included in the multivariable analysis. High LAMP3 cytoplasmic expression (HR: 4.632, 95% CI: 2.421–8.864; p = 0.001), patients’ age (HR: 1.729, 95% CI: 1.027–2.911; p = 0.039), and FIGO stage (HR: 2.049, 95% CI: 1.113–3.774; p = 0.021) were identified as independent predictive factors for poor outcome. Kaplan–Meier survival curve analysis also demonstrated that LAMP3 cytoplasmic expression, patients’ age, and FIGO stage were significantly related to 5-year survival (log rank, p < 0.001, Figure 3).
Kaplan–Meier plots using the log rank survival test. (a) Overall survival rate in patients with high LAMP3 cytoplasmic expression was significantly lower than that in patients with low and no LAMP3 cytoplasmic expression (p < 0.001). (b) Overall survival rate in patients age at diagnosis >60 years was significantly lower than that in patients age at diagnosis ≤60 years. (c) Overall survival rate in patients with high FIGO stage was significantly lower than that in patients with low FIGO stage (p < 0.001).
Discussion
Human ovarian cancer is one of the most common gynecologic malignancies and led to the first case of cancer death among women. 16 Currently, the conventional therapy to EOC was before or after surgery and total six to eight cycles of regular platinum–taxane-based chemotherapy. 17 For the past 30 years, 5-year survival has increased slightly but still remains below 35%. Therefore, assessment of factors affecting behavior of EOC patients is indispensable to improve postoperative prognosis.
Recently, some studies reported that LAMP3 is a newly identified tumor-specific protein. Kanao et al. 18 reported that the overexpression of LAMP3 protein in cervical cancer is related to an advanced metastatic potential and may serve as a noble prognostic factor. What’s more, a growing number of studies have suggested that LAMP3 expression has been associated with hypoxia-induced therapy resistance and metastasis and poor overall survival in certain types of cancers.12,18 Nagelkerke et al. 12 analyzed LAMP3 expression in breast cancer cell lines and found LAMP3 expression induced in an oxygen concentration-dependent manner. So far, only one study has reported that LAMP3 gene was upregulated in ovarian cancer cell line selected by complementary DNA (cDNA) microarray analysis. 19 However, LAMP3 protein expression in EOC and its prognostic value are not clearly.
In our present investigation, we determined mRNA and protein expression levels of LAMP3 in both malignant and normal ovarian tissues. LAMP3 mRNA level was significantly higher in EOC than in normal ovarian tissues, normal fallopian tube tissues, benign ovarian tumors, and borderline ovarian tumors. High cytoplasm staining of LAMP3 protein level was associated with high grade, presence of lymph node metastasis, presence of metastasis, higher serum CA125 level, and higher FIGO stages. High Nucleus staining of LAMP3 protein level was associated with high grade, presence of single or double, presence of metastasis, and higher FIGO stages. Finally, high LAMP3 cytoplasm protein expression is an independent prognostic marker for poor overall survival in EOC patients. Our data clearly showed that high cytoplasmic expression of LAMP3 and higher age at diagnosis were associated with significantly poorer survival.
Although there are some limitations such as the small patient number and relatively short follow-up time, to our knowledge, this is the first report of overexpression of LAMP3 in EOC. Our data indicated that LAMP3 may be nominated as a novel prognostic marker for EOC. In our findings, there was a higher expression of LAMP3 in EOC patient specimens, which is associated with a poorer prognosis. Both mechanistic and prospective experiments are required to determine whether LAMP3 acts as a molecular marker for EOC.
In conclusion, our study demonstrates the expression of LAMP3 in EOC patient and its prognostic value for EOC. Our data supported that LAMP3 may serve as a novel diagnostic and prognostic marker and be used as a therapeutic target for EOC.
Footnotes
Acknowledgements
D.W. and X.C. contributed equally to this work. Tissue samples used in this study were obtained from Clinical Bio-Bank of Affiliated Hospital of Nantong University. Y.X. designed the study; D.W. collected the tissue samples; C.Y. and Y.L. performed the IHC analysis; Y.X. and Y.Z. collected clinical data; X.C. participated in the evaluation of the IHC; X.C. and D.W. drafted the manuscript; Y.X. supervised the study. W.G. revised some part of the manuscript. All authors read and approved the final manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Technological Innovation and Demonstration of Social Undertakings Projects (MS32016023) of Nantong, China.