PRSS3 is a prognostic marker in invasive ductal carcinoma of the breast

Objective Serine protease 3 (PRSS3) is an isoform of trypsinogen, and plays an important role in the development of many malignancies. The objective of this study was to determine PRSS3 mRNA and protein expression levels in invasive ductal carcinoma of the breast and normal surrounding tissue samples. Results Both PRSS3 mRNA and protein levels were significantly higher in invasive ductal carcinoma of the breast tissues than in normal or benign tissues (all P < 0.05). High PRSS3 protein levels were associated with patients’ age, histological grade, Her-2 expression level, ki-67 expression, and the 5.0-year survival rate. These high protein levels are independent prognostic markers in invasive ductal carcinoma of the breast. Materials and Methods We used real-time quantitative polymerase chain reactions (N = 40) and tissue microarray immunohistochemistry analysis (N = 286) to determine PRSS3 mRNA and protein expression, respectively. PRSS3 protein levels in invasive ductal carcinoma of the breast tissues were correlated with the clinical characteristics of patients with invasive ductal carcinoma of the breast and their 5.0-year survival rate. Conclusions PRSS3 acts as an oncogene in invasive ductal carcinoma of the breast development and progression. This finding implies that detection of PRSS3 expression can be a useful prognosis marker and the targeting of PRSS3 can potentially represent a new strategy for invasive ductal carcinoma of the breast treatment.


INTRODUCTION
Breast cancer is one of the most common malignancies that threatens women's health and is now the number one malignant tumor among women. Studies have shown that [1] malignant tumors are caused by a variety of genes that are involved in multistage synergistic results. Over the past two decades, several studies have been conducted on the early diagnosis of breast cancer. Our study was focused on the tertiary prevention of cancer to determine the best adjuvant therapy to prevent post-surgery recurrence and metastasis [2][3]. Multidisciplinary scientists have collaborated to promote breast cancer treatment and have made great progress in mortality reduction with the use of endocrine therapy. However, further studies are needed to identify effective biomarkers, particularly among patients (estimated to account for 30-40% of general breast cancer population) who are insensitive to endocrine therapy [3][4][5]. Therefore, identifying useful biomarkers is essential for precision treatment of breast cancer.
The serine protease super family of chymotrypsin is expressed in intestinal digestion of food proteins in mammals [6]. These key enzymes are involved in digestion, fibrinolysis, reproduction, blood clotting, immune response, and signal transduction from the extracellular environment to the cell; however, there are many characteristics and functions that are not yet clear.
Serine proteases share a common chemical mechanism of having the same reactive serine residue catalytic site, but they differ greatly in degrees of native properties and substrate selectivity [7]. Chemokines,

Research Paper
Oncotarget 21445 www.impactjournals.com/oncotarget growth factors, growth factor receptors, and other signaling receptors are activated and transmitted through an extensive array of membranes and secreted serine proteases, and then enhance the transduction of signals for tumorigenesis, tumor proliferation, and metastasis [8][9][10].
Serine protease 3 (PRSS3) is one of the isoforms of trypsinogen, which plays an important role in the development and progression of many types of malignant tumors [17]. In 2010, Hockla et al. [7]. created an HMT-3522 breast cancer-growth model under 3-dimentional organ culture conditions and found that the expression of PRSS3 in breast cancer cell line T4-2 was enhanced. In our study, real-time quantitative polymerase chain reaction (qPCR) and tissue microarray immunohistochemistry analysis (TMA-IHC) were used to determine the mRNA and PRSS3 expression, respectively, in invasive ductal carcinoma of the breast from tissue samples. We sought to determine both the mRNA and protein expression of PRSS3 in invasive ductal carcinoma of the breast (IDC) tissue samples and correlate this with the patient's clinical characteristics and 5.0-year survival rate.

PRSS3 mRNA level was significantly higher in IDC tissues than in adjacent normal tissues
To determine the PRSS3 mRNA level in IDC patients, total RNA was isolated from the cancerous tissue of 40 IDC patients. Concurrently, 40 tissue samples of matched adjacent normal breast were collected to determine mRNA levels in noncancerous tissues. Realtime qPCR revealed that PRSS3 mRNA expression was upregulated in IDC samples compared with that in corresponding adjacent normal breast tissue samples. The expression of PRSS3 mRNA in IDC tissue samples (7.04 ± 2.04) was significantly higher than that in noncancerous tissues (2.39 ± 0.50) (P < 0.001) ( Figure 1).

PRSS3 expression was significantly higher in IDC than surrounding normal and benign tissues
The overexpression of PRSS3 and its cellular distribution in IDC or surrounding tissues were analyzed by TMA-IHC using rabbit polyclonal antihuman PRSS3 antibodies ( Figure 2). We determined PRSS3 protein expression in 386 archived breast tissue blocks comprising 286 IDC tissues and 100 normal breast tissues. Positive PRSS3 staining, identified as brown particles, was distributed in the cytosol and cell nuclei of the tissue samples. The incidence of PRSS3 expression in IDC tissue (66.43%, 190 of 286) was significantly higher than that in normal breast tissue (22%, 22 of 100) (χ2 = 14.773, P < 0.001). The frequency of positive PRSS3 expression in IDC tissue and in normal tissue is shown in Table 1.

Association of PRSS3 expression with IDC clinical characteristics
We next investigated the relationship between PRSS3 protein levels and the clinical pathologic parameters of the patient ( Table 2).
Among the 286 IDC patients, we got 26 cases of triplenegative breast cancer (TNBC) patients; we correlated PRSS3 protein expression with TNBC patents' clinical characteristics (Table 3). There were no significant associations between high PRSS3 protein expression and the clinicopathological features.

High PRSS3 protein expression predicts poor 5-year survival in IDC patients
We analyzed the association between PRSS3 expression and the survival status of 286 patients using

DISCUSSION
In the current study, we determined the levels of both PRSS3 mRNA and PRSS3 protein expression in both IDC tissues and adjacent normal tissues. The PRSS3 mRNA levels were significantly higher in IDC tissues than in adjacent tissues. Similarly, the PRSS3 protein levels were significantly higher in IDC tissues than in adjacent normal tissues. High PRSS3 protein levels were associated with patients' age, Her-2 expression levels, ki-67 expression levels, and pathology typing. Finally, high PRSS3 protein expression is an independent prognostic marker for a poor 5.0-year survival rate in IDC patients.
PRSS3, one of the three major isoforms of trypsinogen, is a serine protease that is synthesized mainly in pancreatic acinar cells. It can be secreted into the small intestine to promote digestion [15][16]47]. As a member of the serine protease family, PRSS3 is highly homologous to trypsinogen I and II (PRSS1 and PRSS2) at both the gene and protein levels [17]; however, unlike PRSS1 and PRSS2, a small portion of pancreatic exocrine secretions is composed of PRSS3, which accounts for 3.0-10% of the trypsinogen content in normal pancreatic juice [18]. The tumor specificity of trypsinogen has been elucidated in many types of tumors and it is thought to be involved in the development and progression of malignancies [9,11,[19][20][21][22][23][24][25][26]. PRSS3 is expressed in the airway epithelium and its active trypsin is detected in the lung bronchial epithelium, which might be related to its ability to activate various proteases in relation to coagulation, fibrinolysis, and/or inflammation [27]. In non-small-cell lung cancer (NSCLC), the expression of PRSS3 is closely associated with metastasis and with a low prognosis for NSCLC patients. In addition, overexpression of PRSS3 in lung-cancer tissue results in increased migration across endothelial cells, thereby highlighting the potential role of trypsin in tumor metastasis [28].
Promoter methylation of the PRSS3 gene suggests silencing of its expression, which is at ~53% in NSCLC; however, some researchers believe that methylation in NSCLC is a random event [17]. While a study of   Oncotarget 21450 www.impactjournals.com/oncotarget esophageal squamous cell carcinoma found that promoter methylation can silence the expression of PRSS3, [29] some other studies reported that methylation of the PRSS3 promoter in bladder cancer, lung cancer, esophageal cancer, and gastric cancer also silence its expression [17,[29][30]. Some tissue microarray analyses have been conducted to confirm that PRSS3 is overexpressed in metastatic lung cancer [28]. Methylated PRSS3 might be a suitable gene for the detection of malignancies, although more studies are needed to elucidate the specificity of this gene for the detection of serum PRSS3 in cancer patients [17].
PRSS3 was expressed in both ovarian epithelial cancer and benign ovarian tumors, but its expression was significantly higher in ovarian epithelial cancer [31][32]. Hockla et al. [33][34][35][36]. demonstrated using nude mouse models that PRSS3 is highly expressed in prostate cancer and is a key factor in lung metastasis of prostate cancer. By reducing the expression of PRSS3, lung metastasis of prostate cancer would be significantly inhibited [35]. In addition, the researchers also found that PRSS3 is also expressed in the ovarian stroma, suggesting that the expression of PRSS3 might also be associated with the tumor's microenvironment [37][38]. In in vitro experiments using a cell model, Jiang et al. [36]. found that overexpression of PRSS3 increased the invasive ability of pancreatic cancer cells, but that when PRSS3 expression was knocked down, the invasion of pancreatic cancer cells significantly decreased; however, the expression of PRSS3 had no significant effect on the ability of cells to migrate. PRSS3 is expressed in the acinar cells of the pancreas but not in the normal ductal epithelium, and PRSS3 expression was significantly enhanced in cancer tissues with vascular and lymphatic metastasis. In pancreatic cancer, the research data showed that the expression of PRSS3 and lymph node metastasis, as well as distant metastasis, were closely related [36]. Our data suggest that PRSS3 acts as oncogenes in IDC.
Studies have shown that trypsin can activate protease-activated receptors in epithelial cells [39], thereby triggering the development of breast cancer through the signaling pathway [40][41][42][43] ; however, another study questioned this claim [44]. In colorectal cancer, trypsin suppresses the proliferation of cancer cells by blocking downstream gene proliferation-1 (BOP-1) [45]. Hockla et al. [7] found a series of trypsinregulated genes using microarray analysis and real time q-PCR. Among them, EGFR, ITHAV, and CD109 were upregulated by trypsin expression, whereas BOP-1 and CD74 were downregulated. EGFR and its EGF-responsive family members were found to play a central role in the progression of breast and other tumors [46].
Our current study indicated that PRSS3 is an independent prognostic factor in the development and progression of IDC. In subsequent work, in vivo and vitro mechanisms will be necessary to further confirm and expand these conclusions. We look forward to a more comprehensive understanding of the development process of IDC so as to provide new findings that will help prevent and treat the disease.

Study limitations
Our study had several limitations. First, our research samples were limited to the female Chinese population. We have not collected tissues from male patients, which might result in sample bias; therefore, our results cannot be extended to other countries or to males, and we do not have international research collaborations with larger and more diverse populations. Secondly, our study examined only IDC in breast cancer. As we know, a tumor is a heterogeneous disease and different types of tumors will have different prognoses; therefore, we do not know whether our findings can be generalized to all types of invasive breast cancer. Finally, we have not provided the mechanism by which PRSS3 occurs in IDC. In the next phase of the study, we intend to further validate the expression of PRSS3 using a cell model and a nude mouse model, which can aid in exploring its molecular mechanism of action in regulating the growth of IDC tumor cells.

Human tissue specimens and patient clinical information
Human breast cancer and surrounding tissues in the study were obtained from 286 IDC patients, who underwent operations between January 2010 and August 2011 at the Affiliated Hospital of Nantong University, Jiangsu Province, China. The patient cohort inclusion and exclusion criteria included: (1) accurate pathologic diagnosis of primary IDC; (2) complete clinic pathologic and follow-up date; and (3) previously untreated, with surgery as the first treatment (Table 5). Thus, analysis of the data would reflect the actual impact of tumor biology on the clinical outcome. This study was approved by the Ethics Committee permission of the Affiliated Hospital of Nantong University, and all patients signed informed consent. The related important clinical information of each patient was collected from their medical records.