Background
Globally, breast cancer is the most commonly diagnosed cancer in women, while metastatic disease is the leading cause of cancer-related deaths in this group [
1]. Epithelial integrity and intact extracellular matrix (ECM), which includes basement membrane and interstitial connectivity tissue, are essential for normal cell behavior and tissue homeostasis [
2]. Remodeling and degradation of the ECM, along with defects in structural cell-adhesion molecules, play a significant role in breast cancer progression [
3]. Type II transmembrane serine proteases (TTSPs) are a relatively new subfamily of S1 class serine proteases in humans comprised of 17 proteolytic enzymes [
4,
5]. In addition to their roles in normal tissue development and homeostasis, TTSPs are also involved in several human diseases, including cancer, and many show potential as biomarkers of tumor progression and represent prospective therapeutic targets [
6,
7]. TTSPs localize to the cell membrane and are able to degrade the ECM and remodel intercellular and cell-ECM junctions. Accordingly, dysregulation of TTSPs is thought to be involved in the early stages of tumorigenesis, tumor growth, and cancer cell invasiveness that lead to metastasis [
8,
9]. In this study, we looked at the expression of two members of the TTSP family, hepsin (also called TMPRSS1), which is encoded by the
TMPRSS1 gene, and TMPRSS3, encoded by the
TMPRSS3 gene.
Hepsin upregulation in malignant tumors has been demonstrated in prostate and ovarian cancers as well as in renal cell carcinoma [
10‐
13]. A recent study used immunohistochemistry to show that hepsin protein levels were upregulated in human breast cancer tumor samples [
14].
TMPRSS1 mRNA overexpression is associated with ER(α)-positive human breast tumors [
15], while
TMPRSS3 overexpression has been implicated in pancreatic and epithelial ovarian cancers [
16,
17]. Missense mutations in the
TMPRSS3 gene that lead to structural TMPRSS3 defects are associated with hereditary deafness [
18]. Both hepsin and TMPRSS3 belong to the hepsin/TMPRSS subfamily of TTSPs and share structural features [
5,
6]. TTSPs are anchored to the cell membrane via an N-terminal transmembrane domain. At the C-terminus, TTSPs have an extracellular serine protease domain that is required for their catalytic activity [
4,
7]. Notably, several soluble forms of TTSPs that retain catalytic activity have also been detected [
4,
9]. Hepsin and TMPRSS3 appear to be capable of autocatalytic activation, suggesting that they play roles as initiators of proteolytic cascades that lead to ECM remodeling [
19,
20]. Overexpressed hepsin activates proteolytic pathways and also directly interferes with cell-cell and cell-ECM adhesion molecules. Hepsin can activate hepatocyte growth factor (HGF) and urokinase-type plasminogen activator- (uPA) mediated proteolytic pathways, which results in ECM degradation [
21‐
23]. Hepsin plays a physiological role as it directly and specifically cleaves laminin-332 (ln-332, previously termed laminin-5), an important ECM protein involved in maintaining the structural integrity of the basement membrane [
24]. It was shown recently that hepsin becomes mislocalized when liver kinase B1 (lkb1) expression is lost and that overexpressed hepsin induces basement membrane degradation in breast cancer [
25].
This is the first study to examine
TMPRSS3 gene expression in a set of clinical breast cancer samples and to investigate whether altered
TMPRSS1 and
TMPRSS3 gene expression has an impact on the clinical outcome of breast cancer patients. Here, we analyzed the associations of mRNA and protein expression of these genes with clinicopathological parameters and breast cancer-specific survival. Recently, we reported that
TMPRSS3 SNP rs3814903 associated with both breast cancer risk and survival and SNP rs11203200 associated with breast cancer survival [
26]. Furthermore,
TMPRSS1 SNPs rs12151195 and rs12461158 remained independent prognostic factors of breast cancer survival [
26]. Our previous study showed that another member of the TTSP family, matriptase (encoded by the
ST14 gene), is associated with breast cancer survival [
27]. We also showed that several
TMPRSS6 (encoding matriptase-2) variants are related to breast cancer prognosis and matriptase-2 expression levels decrease with tumor progression [
28]. These previous findings prompted us to investigate whether altered expression of hepsin and TMPRSS3 might also have a role in the molecular pathology of breast cancer. Although the physiological substrates for TMPRSS3 remain unclear, it is possible that the biological mechanisms that lead to ECM degradation are similar to those of hepsin. When overexpressed in breast cancer, hepsin and TMPRSS3 could promote cancer cell invasiveness via dysregulated proteolytic activity. This results in defects in the basement membrane and in uncontrolled ECM degradation. However, the expression levels seem to decrease as tumor malignancy increases, and low expression levels of these proteins are associated with poor breast cancer survival as well as with the adjuvant treatments the patients received.
Discussion
This is the first study to link altered TMPRSS3 expression to breast cancer tumor progression and to show that low TMPRSS1 and TMPRSS3 expression, both at the mRNA and protein levels, has prognostic value for poorer survival of breast cancer patients. Importantly, this is also the first cancer study to show that altered TMPRSS3 expression has prognostic value for cancer-related death. In benign breast tumor cells, the expression levels of TMPRSS1 and TMPRSS3 are consistently low, whereas the expression levels are higher in cancer cells. In malignant samples, there was clearly a high degree of intertumor variation in the expression levels of the studied genes. However, our results indicated that despite overexpression in well-differentiated tumors, the expression levels decreased as the tumors acquired more malignant characteristics. Poorly differentiated tumors expressed lower levels of both TMPRSS1 and TMPRSS3. Notably, both mRNA and protein expression levels were associated with the clinical characteristics of breast cancer: Low expression levels predicted poorer survival and increased risk of distant metastasis compared to high expression levels. Low TMPRSS1 and TMPRSS3 expression remained independent factors affecting breast cancer-specific survival in the Cox regression analysis.
These results are consistent with previous studies that reported
TMPRSS1 overexpression in various cancers, especially in prostate cancer [
10‐
14], as well as
TMPRSS3 overexpression in pancreatic and ovarian cancers [
16,
17]. We found a notable difference in
TMPRSS1 and
TMPRSS3 mRNA expression between benign samples and grade I malignant tumors in that grade I breast cancer samples expressed considerably higher levels of
TMPRSS1 and
TMPRSS3 than benign samples. This finding supports the theory that hepsin is related to prostate cancer and suggests that hepsin and TMPRSS3 may also play important roles in the early phases of breast carcinogenesis [
32,
33]. Our hepsin and TMPRSS3 immunohistochemical staining results correlated with the mRNA expression results. Specifically, samples with more intense cytoplasmic staining were associated with lower tumor grade and stage, and samples with low expression levels were linked to grade III and stage III and IV tumors. Low mRNA expression levels were common in tumors that did not express hormone receptors but that were HER2-positive. In addition, hepsin expression was low in samples with positive nodal status. In the current study, many of the clinical variables that are generally related to advanced breast tumor progression and higher breast cancer mortality rate were linked with low expression levels of the studied genes [
34,
35].
Our survival results indicated that low expression of both of the studied genes was an independent prognostic factor in breast cancer. Along with positive nodal status and large tumor size (T3, T4), low mRNA expression remained an independent factor for breast cancer survival. Similarly to our results, Dhanasekaran et al. showed previously that low hepsin protein expression in human prostate cancer samples correlated with poor prostate cancer prognosis [
36]. In their study, absent or low hepsin immunostaining was dominant in benign samples, whereas hepsin staining was strong in cancer samples. The strongest hepsin staining was in the precursor lesions of prostate cancer (HG-PIN). Yet among cancer samples, absent or low hepsin expression was associated with prostate-specific antigen (PSA) elevation after radical prostatectomy and large tumor size, indicating poorer survival. In contrast, regarding tumor malignancy, high
TMPRSS1 mRNA expression correlated with advanced tumor stages in prostate cancer [
37]. Roemer et al. showed that in renal cell carcinoma, decreased
TMPRSS1 mRNA expression was an independent factor that predicted poorer renal cell carcinoma-specific survival [
38]. They suggested that hepsin may be involved in both the early and late development of renal cell carcinoma. However, Betsunoh et al. have observed that hepsin overexpression is associated with poorer renal cell carcinoma survival [
39]. In human hepatocellular carcinoma, Chen et al. found that decreased
TMPRSS1 mRNA expression predicted shorter survival time [
40]. These studies illustrate variations among the different studies; even so, many of these studies are in agreement with our findings.
In this study we have shown that altered
TMPRSS1 and
TMPRSS3 expressions are associated with the occurrence of relapses and that low
TMPRSS3 mRNA and protein expression are independent factors affecting distant metastasis occurrence. Aberrant expression of TTSPs is associated with tumor invasion and metastasis in various epithelial cancers [
6,
41]. Supporting our results, Vasioukhin hypothesized that hepsin may promote metastasis in prostate cancer [
32]. This hypothesis suggested that in the initial stages of metastasis, hepsin overexpression might stimulate the invasion of primary tumor cells but, once the cells metastasized, hepsin expression would no longer be essential in distant lesions. We found that distant metastases occurred more frequently during follow-up, in patients with low expression levels of the studied genes in primary tumors. This finding supports the theory that distant metastases are more likely to occur once a certain stage in tumor development is reached, and expression of proteolytic serine proteases is needed from primary tumors to achieve that stage. When local breast cancer relapses and distant metastasis were studied together, low
TMPRSS3 mRNA and protein expression remained independent factors that affected relapse in the Cox regression analyses. On the other hand, based on prostate cancer cell line studies, Srikantan et al. suggested that hepsin overexpression could have antitumorigenic effects and hinted that hepsin might be involved in some sort of positive feedback response [
42]. They suggested that decreased hepsin expression could be linked with poor prostate cancer prognosis as exogenously provided hepsin negatively regulated the growth of metastatic prostate cancer cells. However, the first hepsin expression study in MDA-MB-231 and HER18 breast cancer cell lines showed that low hepsin expression levels reduced cell viability and the colony formation rate [
14]. Wittig-Blaich et al. showed in a prostate cancer cell line study that the consequences of increased hepsin expression at the cellular level depend on the cell’s microenvironment, and the authors suggested that hepsin overexpression must be spatially and temporally restricted for the efficient development of tumors and metastases [
43]. Taken together, these findings support the theory that, depending on the phase of tumorigenesis and metastasis, hepsin expression might either promote or suppress tumors and metastasis.
Based on their proteolytic activity at the cell surface, TTSPs could contribute to tumor progression by affecting initiation of the metastatic process in primary breast cancer tumors. Several substrates for hepsin have been linked to epithelial carcinogenesis, including HGF and uPA. Hepsin and another TTSP, matriptase, efficiently convert inactive pro-HGF to biologically active HGF that, in turn, activates the HGF receptor c-Met [
21,
22]. This leads to basement membrane disorganization. Abnormal activation of the HGF/c-Met signaling pathway by aberrant hepsin overexpression is a possible mechanism for the enhancement of tumor progression. In addition, hepsin converts potently pro-uPA into active uPA, which initiates the degradation of ECM by cleaving plasminogen into plasmin [
23]. Hepsin may also directly contribute to tumor progression and metastasis by causing defects in cell junctions. Miao et al. showed in human and mouse ovarian cancer cell line studies that hepsin overexpression contributes to ovarian cancer progression via cell membrane interactions with desmosomes [
44]. By immunofluorescence they showed that, in addition to cytoplasm hepsin co-localizes with desmosomes at the cell junctions; further, intact desmosomes are required for the membrane localization of hepsin. Supporting these findings, Partanen et al. recently reported that hepsin partially co-localizes with the desmosomal junction protein desmoplakin and, in breast cancer, the two proteins no longer co-localize when lkb1 expression is lost [
25]. Notably, loss of lkb1 causes hepsin to relocalize from desmosomes to cytoplasm. Taken together, these studies indicate that the mislocalization and overexpression of hepsin could potentially initiate basement membrane degradation and lead to tumor cell invasion.
The limitations of our study include that
TMPRSS3 expression in cancer has not nearly been studied as extensively as the expression of
TMPRSS1. More work needs to be done to study the biological role of TMPRSS3 in cancer. Nonetheless, our study presents in a coherent clinical breast cancer sample set that TMPRSS3 is a credible prognostic biomarker. In contrast to our results, a previous study presented that hepsin overexpression was associated with positive nodal status and tumor stage in breast cancer [
14]. However, no survival analyses were done in that study and the analysis methods were different. In addition, to validate our results and the prognostic value of the studied genes in a large clinical breast cancer microarray database, we used an online Kaplan-Meier survival analysis tool [
31]. Similar to our study, in these analyses the cohorts were divided into two groups according to the median expression of
TMPRSS1 and
TMPRSS3. Based on the survival curves displayed and the logrank
P values both low
TMPRSS1 and
TMPRSS3 expression significantly associated with poorer relapse-free survival, overall survival, and distant metastasis-free survival [
31]. To sum up, the survival trend was exactly alike compared to our results.
When the treatment data was included in the multivariate survival analyses, low
TMPRSS1 mRNA expression remained an independent factor of poor prognosis in patients who were treated with radiotherapy. It must be highlighted that low
TMPRSS1 expression remained the only significant variable regarding prognosis which excludes for example poor differentiation level of breast cancer cells in these analyses. Furthermore, no significant results were found in patients who were not given any adjuvant therapies. When
TMPRSS1 expression level is higher and epithelial integrity is still rather intact it might be that radiation induced cellular lethality is much more aggressive in breast cancer cells. However, when
TMPRSS1 expression is low and epithelial integrity already damaged it appears that the remaining breast cancer cells are radioresistant leading to these cells surviving which has negative impact on the clinical outcome. Our results indicate that low
TMPRSS1 expression may independently reduce the therapeutic function of radiation yet the specific cellular mechanisms remain unclear. Interestingly, Nakamura et al. showed in an endometrial cancer cell line study that hepsin overexpression resulted in significant cell accumulation at the G2/M phase leading to cell cycle arrest [
45]. Cancer cells in general are thought to be the most radiotherapy sensitive exactly at the G2/M phase [
46]. These previous studies comply our results even though our significant results are related to mRNA expression and in breast cancer tumor samples. In a previous study of our group we found several
TMPRSS1 and
TMPRSS3 SNP genotypes that associated with survival in patients treated with radiotherapy [
26]. We have planned to study in the future the potential associations between our current results with the ones from our SNP study.
The combinations of low expression levels of mRNA and protein were independent factors that predicted poor survival. This suggests that TTSPs are prognostic biomarkers for breast cancer. Matriptase (encoded by the
ST14 gene) is a TTSP that, similar to hepsin, can activate pro-HGF. A recent study showed that hepsin and matriptase are direct pericellular activators of pro-HGF and hypothesized that their suggested ability to autoactivate might be the initial step in HGF/c-Met-mediated basement membrane degradation [
47]. Our previous study of matriptase expression in breast cancer resulted in conclusions that were similar to those in the present study in that low matriptase expression was associated with poorer breast cancer survival [
27]. However, others have reported the opposite and other studies of matriptase expression in breast cancer have not given consistent results [
48]. Notably, since matriptase and hepsin have identical substrates and since both have possible tumor progression and metastasis-promoting activities, further studies of TMPRSS3 are needed to better understand its functions and substrates. Structurally, TMPRSS3 is almost identical to hepsin, and here we have shown that their mRNA and protein expression patterns are very similar in different phases of breast carcinogenesis and correlate with breast cancer prognosis. Co-expression of these proteolytic serine proteases could enhance their effects on tumor cell invasion and metastasis.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
MP, VK, V-MK and AM contributed to the conception and design of the study. MT, VK, VM-K and AM were involved in the acquisition of data. MP, KL, MT, HA, BB, YS and AM participated in the analysis and interpretation of data. MP, MT and BB drafted the manuscript. MP, KL, MT, BB, V-MK and AM contributed to the revision of the manuscript. HA and YS provided technical and material support. VK, V-MK and AM provided administrative support. KL, V-MK and AM contributed to the supervision of the study. All authors read and approved the final manuscript.