Transgenic expression of human cathepsin B promotes progression and metastasis of polyoma-middle-T-induced breast cancer in mice

Abstract

Elevated expression of the cysteine protease cathepsin B (CTSB) has been correlated with a poor prognosis for cancer patients. In order to model high CTSB expression in mammary cancer, transgenic mice expressing human CTSB were crossed with transgenic polyoma virus middle T oncogene breast cancer mice (mouse mammary tumor virus-PymT), resulting in a 20-fold increase in cathepsin B activity in the tumors of double-transgenic animals. CTSB expression did not affect tumor onset, but CTSB transgenic mice showed accelerated tumor growth with significant increase in weight for end-stage tumors, as well as an overall worsening in their histopathological grades. Notably, the lung metastases in the CTSB transgenic animals were found to be both significantly larger and to occur at a significantly higher frequency. Ex vivo analysis of primary PymT tumor cells revealed no significant effects from elevated CTSB levels on tumor cell characteristics, that is, the formation of tumor cell colonies and the sprouting of invasive strands from PymT cell spheroids. However, tumors from CTSB-overexpressing mice showed increased numbers of tumor-associated B cells and mast cells. In addition, more CD31+ endothelial cells were detected in these tumors, correlating with higher levels of vascular endothelial growth factor (VEGF) being present in the tumor and serum. We conclude that elevated proteolytic CTSB activity facilitates progression and metastasis of PymT-induced mammary carcinomas, and is associated with increased immune cell infiltration, enhanced VEGF levels and the promotion of tumor angiogenesis.

Introduction

The family of lysosomal cysteine cathepsins (clan CA; C1a papain family) is primarily involved in lysosomal bulk proteolysis (Barrett, 1992). However, cysteine cathepsins also exert specific functions in several physiological (Brix et al., 2008) and pathological processes, including cancer (Mohamed and Sloane, 2006; Vasiljeva et al., 2007). Human cathepsins comprise 11 members: cathepsin B, C, F, H, K, L, L2/V, O, S, W and X/Z (Rawlings et al., 2010). Elevated expression of certain cysteine cathepsins, such as cathepsin B, L and X/Z, has been described for several human cancers, and they have been found to accumulate at the invasive front of tumors (Jedeszko and Sloane, 2004). Despite their general localization in the endosomal/lysosomal compartment, their secretion or extracellular localization has been observed during neoplastic formation (Rozhin et al., 1994; Mai et al., 2000).

It is now widely accepted that the tumor stroma and tumor-associated inflammation has an active role in tumor promotion (Mantovani et al., 2008). In fact, the majority of tumor-promoting proteases is produced by cells of the tumor microenvironment, such as fibroblasts (Orimo et al., 2005), or tumor-associated immune cells, for example, macrophages and mast cells (Coussens et al., 1999; Allavena et al., 2008), rather than by the cancer cells themselves. In particular, mouse macrophages show an induction of cysteine protease cathepsin B expression upon their recruitment to metastatic foci (Vasiljeva et al., 2006). Notably, increased cathepsin expression correlates with a poor prognosis for cancer patients, indicating the potential of cathepsins to serve as prognostic markers and therapeutic targets (Berdowska, 2004). However, in addition to the identification of cysteine cathepsins as prognostic markers, the elucidation of in vivo functions in tumor progression and metastasis is crucial for designing therapeutic strategies that target cathepsins. Much progress has been made in understanding how proteases, including metallo-, aspartic-, serine- and cysteine proteases (Del Rosso et al., 2002; Egeblad and Werb, 2002; Turk et al., 2004; Shin et al., 2007), functionally contribute to cancer development. There is accumulating evidence that proteases are not only involved in the processing of extracellular matrix (ECM) proteins (Werb et al., 1999), but that they provide tumor-promoting conditions by their ability to modulate the bioavailability of growth- and angiogenic factors (Du et al., 2008), the regulation of bioactive chemokines and cytokines (Van Damme et al., 2004), and the processing of cell–cell adhesion molecules (Gocheva et al., 2006).

Mouse models with deficiencies in individual cathepsins help to understand mechanistic aspects of cathepsin functions during tumor development (Gocheva et al., 2006, 2010; Dennemarker et al., 2010; Sevenich et al., 2010). Using the polyoma virus middle T oncogene (mouse mammary tumor virus (MMTV)-PymT) induced mouse model for breast cancer (Guy et al., 1992), we recently demonstrated that ablation of mouse cathepsin B leads to diminished tumor and metastatic burden due to reduced tumor cell invasion and proliferation (Vasiljeva et al., 2006, 2008). As elevated human cathepsin B (CTSB) expression is often found in breast cancer and correlates with a poor prognosis, we crossed MMTV-PymT-expressing mice with mice harboring a genomic human CTSB construct under the control of its genuine promoter (Houseweart et al., 2003) to induce CTSB overexpression in order to mimic the clinical situation in a mouse model for breast cancer. The aim of the present study was to analyze the effects of CTSB expression on breast cancer development and metastasis, and to elucidate cancer-related CTSB functions. Here we demonstrate that human CTSB expression in mice facilitates progression of PymT-induced breast cancer and lung metastasis, and is associated with enhanced immune cell infiltration and angiogenesis.

Results

Human CTSB expression in mice was achieved by transgenesis in mice with a CTSB construct comprising the entire intron–exon structure of CTSB, including the 5′ and 3′ untranslated regions, and the genuine human CTSB promoter (Houseweart et al., 2003). To investigate the effects of CTSB expression on cancer development and metastasis, mice harboring the CTSB transgene (Tg(CTSB)+/0) were crossed with mice expressing the PymT in mammary epithelial cells under the control of the MMTV long-terminal repeat promoter (Guy et al., 1992). Female mice hemizygous for the MMTV-PymT transgene were grouped according to their cathepsin B genotypes into mice with endogenous CTSB levels (PymT+/0;wild type (wt)) and mice expressing mouse and human cathepsin B (PymT+/0;Tg(CTSB)+/0).

Histological localization, expression and proteolytic activity of CTSB

To assess expression of the CTSB transgene in mammary tumors, we examined the histological localization, protein levels and proteolytic activity of CTSB (Figure 1). Immunohistochemical detection of CTSB revealed expression in tumor cells (Figure 1a; left panel), and particularly high expression in cells of the tumor stroma (Figure 1a; right panel). Pro-CTSB and the active single- and double-chain forms were detected by western blotting in breast cancer homogenates, proving the correct processing of CTSB in mice (Figure 1b). Human and mouse cathepsin B show very similar enzymatic properties (Caglic et al., 2009). Therefore, we compared the cathepsin B activities in cancers from PymT+/0;wt and CTSB expressing PymT+/0;Tg(CTSB)+/0 mice. The latter showed about 20-fold higher total cathepsin B activity compared with tumors from PymT+/0;wt mice at 10 and 14 weeks of age. Notably, the cathepsin B activity in liver and lungs of PymT+/0;Tg(CTSB)+/0 was increased by only 3- and 10-fold, respectively (data not shown). Furthermore, we analyzed intracellular CTSB expression and CTSB secretion by in vitro cultured primary PymT tumor cells. By using an antibody selective for human CTSB, we could detect the active 25-kDa form of CTSB in cell lysates (Figure 1d), and the 42-kDa pro-form of CTSB in the conditioned cell culture media of PymT+/0;Tg(CTSB)+/0 tumor cells as the major bands (Figure 1d). The cathepsin B activity of PymT+/0;Tg(CTSB)+/0 tumor cells compared with PymT+/0;wt was increased six- and threefold in lysates and conditioned media, respectively (Figure 1e). This indicates that a small fraction of the CTSB released from the tumor cells is indeed proteolytically active.

Figure 1
figure1

Expression and proteolytic activity of CTSB. (a) Detection of CTSB in tumor sections by immunohistochemistry. Incubation with IgG served as an isotype control. The scale bar represents 50 μm. (b) Detection of CTSB expression by western blotting in primary PymT tumor homogenates of PymT+/0;wt [1] and PymT+/0;Tg(CTSB)+/0 [2]. α-Tubulin served as loading control. (c) Measurement of hydrolysis of z-Phe-Arg-AMC in the presence of the cathepsin B specific inhibitor CA074 in PNS of primary PymT tumor homogenate of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 from 10- (n=4–5 per group) and 14-week-old mice (n=7–9 per group). (d) Detection of CTSB in PNS and conditioned cell media (CCM) of primary PymT tumor cells from PymT+/0;wt [1], PymT+/0;Tg(CTSB)+/0 [2]. α-Tubulin and colloidal Coomassie-stained gel served as loading controls. (e) Measurement of cathepsin B proteolytic activity by hydrolysis of the fluorogenic peptide z-Phe-Arg-AMC in PNS of primary PymT tumor cell lysates (n=6) and CCM (n=3) of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 cells. Cathepsin B activity was measured as the hydrolysis rate of z-Phe-Arg-AMC that is sensitive to the selective cathepsin B inhibitor CA074. Hence, activity was determined and calculated as the difference of AMC release in aliquots with and without CA074. Data are presented as mean and standard error; Student's t-test was used for statistical analyses.

Progression and histopathology of PymT-induced mammary carcinomas

To analyze the effects of CTSB expression on the development of PymT-induced carcinomas, all 10 mammary glands per mouse were palpated for the occurrence of tumors every second day from day 28 to day 44 after birth (Figure 2a). However, there was no CTSB-dependent difference in the course of early tumor development: PymT+/0;Tg(CTSB)+/0 and PymT+/0;wt mice exhibited palpable tumors in 5 of the 10 mammary glands by day 41 and day 42 after birth, respectively. To analyze the progression of established tumors, the diameter of the individual tumors was continuously determined from week 8 to week 14. In contrast to early tumor development, tumor progression of established tumors was accelerated by CTSB expression, with PymT+/0;Tg(CTSB)+/0 mice developing bigger tumors significantly earlier, as indicated for 10 and 14 weeks of age (Figure 2b). These findings were confirmed by measurement of the cumulative tumor weight at these time points. We already observed a tendency towards a greater weight for PymT+/0;Tg(CTSB)+/0 tumors by 10 weeks of age. At 14 weeks of age the cumulative tumor weight of PymT+/0;Tg(CTSB)+/0 mice was 45% higher compared with PymT+/0;wt animals (P<0.001) (Figure 2c). The development of PymT-induced carcinomas is marked by the stepwise development of distinct histopathological stages (Lin et al., 2003). In order to analyze the impact of CTSB expression on tumor grade, we classified pre-malignant and malignant lesions into defined histopathological stages: atypical ductal hyperplasia, ductal carcinoma in situ and invasive ductal carcinoma, which was further subdivided into grade GI–GIII with regard to mitotic activity and cellular differentiation (Elston and Ellis, 1991). Already by 10 weeks of age each of the analyzed tumors contained invasive parts, which were graded as invasive ductal carcinoma GII (moderately differentiated) or GIII (poorly differentiated). Nevertheless, at 10 weeks of age 50% of PymT+/0;Tg(CTSB)+/0 tumors were poorly differentiated compared with 20% in the PymT+/0;wt group (Figure 2d). At 14 weeks 90% of the PymT+/0;Tg(CTSB)+/0 carcinomas were poorly differentiated as compared with 53% in the PymT+/0;wt group (P<0.01) (Figure 2d).

Figure 2
figure2

Progression, tumor burden and histopathology of mammary carcinomas (a) Detection of first palpable tumors of PymT+/0;wt (n=7) and PymT+/0;Tg(CTSB)+/0 (n=9) by palpation of all ten mammary glands from day 28 to 44. The time point at which each genotype developed five tumors is indicated; statistical analysis by Student's t-test. (b) Estimation of the tumor size by palpation. Distribution of tumor size (0, <0.5, 0.5–1, >1 cm) per genotype is shown for the time points 10 and 14 weeks with PymT+/0;wt (n=23) and PymT+/0;Tg(CTSB)+/0 (n=21); The chi square test was used for statistical analyses. (c) The tumor weight of all ten tumors per mouse was measured for PymT+/0;wt (n=7) and PymT+/0;Tg(CTSB)+/0 (n=9) at 10 and 14 weeks of age (n=16 per group). Data are presented as statistical box plots; boxes include data between the 25th and 75th percentiles; statistics by Mann–Whitney U-test. (d) Histopathological grading of HE stained tumor sections of PymT+/0;wt (n=10 and 15) and PymT+/0;Tg(CTSB)+/0 (n=9 and 22) at 10 and 14 weeks of age, respectively. The chi-square test was used for statistical analyses.

Formation of lung metastases

PymT-induced mammary carcinomas are prone to form lung metastases. We quantified the percentage of mice in each group that bore lung metastases at 10 weeks of age as the time point at which the first micro-metastases were microscopically detectable. Notably, we observed micro-metastases in only 12.5% of PymT+/0;wt mice compared with 67% of PymT+/0;Tg(CTSB)+/0 mice (P<0.05) (Figure 3a). At 14 weeks of age 100% of PymT-expressing female mice had developed lung metastases. We estimated the metastatic burden by histomorphometry with respect to the number and average size of the metastases. PymT+/0;Tg(CTSB)+/0 mice developed 55% more lung metastases compared with PymT+/0;wt mice (P<0.05) (Figure 3b). Remarkably, the most pronounced effect of CTSB expression was observed on the average size of the metastases, with a 2.5-fold increase in PymT+/0;Tg(CTSB)+/0 mice compared with PymT+/0;wt mice (P<0.001) (Figure 3b).

Figure 3
figure3

Formation of lung metastases (a) Incidence of micro-metastases in lungs was analyzed at 10 weeks of age for PymT+/0;wt (n=8) and PymT+/0;Tg(CTSB)+/0 (n=6); statistical analysis by chi-square test. (b) Representative images of lung sections and quantification of metastatic burden in lungs of 14-week-old mice. Metastatic burden in lungs was assessed as number and average size of metastases for PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 (n=15 per genotype); Statistics by Mann–Whitney U-test. Analyses of metastases were performed on three independent HE-stained sectional planes per lung.

Tumor cell autonomous characteristics

Alterations in the balance of proliferation and cell death of cancer cells, the potential for tumor cell colony formation, as well as cell migration and invasion each contribute to tumor progression and metastasis (Hanahan and Weinberg, 2000). To examine the impact of tumor-cell-derived CTSB on these processes, we isolated primary PymT tumor cells from multiple 14-week-old PymT+/0;Tg(CTSB)+/0 and PymT+/0;wt donor mice and analyzed the cell cultures during their third passage. CTSB expression had no effect on proliferation as measured by BrdU incorporation, or on TNF-α-induced apoptosis analyzed by annexin V/propidium iodide flow cytometry (Supplementary Figure 1a and b). To assess the effects of CTSB on the potential to form colonies, we embedded single tumor cells in soft agar. After 28 days, we quantified the number of colonies that exceeded 50 μm in diameter. However, quantification of the number of colonies revealed no CTSB-dependent difference in colony formation (Supplementary Figure 1c). Cell migration was estimated by measuring the distance of scratch closure per hour within the first 15 h. However, cell migration was unaffected by CTSB expression (Supplementary Figure 1d). Assuming that higher CTSB levels promote tumor cell invasion, we embedded spheroids composed of primary PymT+/0;Tg(CTSB)+/0 or PymT+/0;wt cells into a collagen I matrix. The number and length of the multicellular strands growing into the collagen matrix within 24 h represent a measure of the invasive potential of the tumor cells in the spheroid. We detected a tendency towards more and longer strands in PymT+/0;Tg(CTSB)+/0 tumor cell spheroids (Supplementary Figure 1e). However, due to large variations no statistically significant difference (P=0.17) could be detected. In summary, tumor cell autonomous characteristics show no or only marginal genotype-dependent differences, suggesting a major impact for CTSB derived from cells of the tumor microenvironment on tumor promotion and metastasis.

Cellular composition of the tumor and metastatic microenvironment

Tumors develop in a complex microenvironment that functionally modulates tumor growth. Infiltrating immune cells are an important component of the tumor microenvironment. During the premalignant transition from atypical ductal hyperplasia to ductal carcinoma, increasing numbers of inflammatory cells, particularly macrophages, are recruited to the tumor stroma (Supplementary Figure 2). In invasive cancers very few inflammatory cells are present within the cancer cell mass, although the close vicinity of inflammatory and tumor cells can be observed at the border zone between the cancer cell mass and the stroma (Supplementary Figure 2).

To further characterize the cellular composition of the primary tumor microenvironment, we quantified several tumor-associated immune cell types in breast carcinomas from 14-week-old mice by flow cytometry, including T and B cells, macrophages, and neutrophil granulocytes. Mast cells were identified by chloroacetate esterase staining (Leder stain; Figure 4d). The relative numbers of CD4+ and CD8+ T cells, macrophages and neutrophils were similar in PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 tumors (Table 1). Interestingly, the numbers of B cells and mast cells were significantly increased in tumors from PymT+/0;Tg(CTSB)+/0 mice, with 3 times more B cells (P<0.05) and 1.7 times more mast cells (P<0.01) compared with PymT+/0;wt mice (Table 1). Histological detection of macrophages, B cells and mast cells in both experimental groups confirmed the higher frequency of B cells and mast cells in CTSB-overexpressing tumors, as well as a similar localization in the tumor stroma as was observed for macrophages (Figures 4a, b and d).

Figure 4
figure4

Localization of inflammatory cells (a, b) Representative images of immunofluorescence staining of the macrophage-specific marker CD301 (red staining) and the B cell specific marker CD19 (red staining) and nuclear stain by Hoechst (blue staining) in mammary carcinomas of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 from 14-week-old mice. (c) Detection of immunoglobulin deposits by direct immunofluorescence labeling with rabbit anti-mouse IgG/IgM (green staining) and detection of FcγRII/III by staining with rat anti-mouse CD32/16 (red staining). (d) Histochemical staining of mast cells in tumor sections from PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 from 14-week-old mice and quantification of degranulated mast cells. Data are presented as mean and standard error; Student's t-test was used for statistical analyses.

Table 1 Tumor-associated immune cells

To analyze whether the immune cell infiltration that was observed in the primary tumors could also be found in the lungs of tumor-bearing mice, we quantified immune cell numbers in the metastases-bearing lungs of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 mice at 14 weeks of age. However, we observed no significant CTSB-related changes in the frequency of leukocytes, B cells and macrophages, whereas mast cells were not detectable in the lungs (Supplementary Figure 3).

Although B cells are present at low frequency in the tumors, immunoglobulins (Igs) are present throughout the stroma, indicating that tumor-associated B cells are active. Furthermore, FCγ-receptor (FCγR)-expressing cells, that is, macrophages, mast cells and neutrophils, are recruited to sites of Ig deposition and activated by the FCγR (Andreu et al., 2010). The Ig deposits in the cancer stroma of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 mice colocalize with the presence of the FCγR (Figure 4c), suggesting an activation of FCγR-expressing immune cells. Murine mast cells are known to express FCγRII and FCγRIII. Among these, FCγRIII has been shown to induce mast cell degranulation (Katz and Lobell, 1995; Takai, 2005) in addition to the FcɛR-dependent mechanisms. In agreement with this argument, the number of degranulated mast cells was found to be doubled in PymT+/0;Tg(CTSB)+/0 tumors compared with PymT+/0;wt (P<0.05; Figure 4d). Hence, human CTSB expression results in increased numbers and activity of mast cells, an immune cell type known to promote cancer and angiogenesis (Coussens et al., 1999).

Angiogenesis and cytokine production in CTSB-expressing tumors

Formation of a tumor vasculature by angiogenesis is essential for tumors to grow beyond 1–2 mm3 in size (Folkman, 2002). To analyze the effects of CTSB overexpression on angiogenesis we detected endothelial cells by immunofluorescence staining of the marker CD31, and quantified the number of CD31+ cells using the automated microscope-based imaging platform Scan^R (Olympus, Hamburg, Germany). At 14 weeks of age tumors from PymT+/0;Tg(CTSB)+/0 mice showed 45% more CD31+ cells compared with PymT+/0;wt (P<0.05; Figures 5a and b). Angiogenesis is induced by various mediators, particularly by vascular endothelial growth factor (VEGF) (Nagy et al., 2007). We therefore measured the amount of VEGF in phosphate-buffered saline-extracts of tumors and in the serum of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 mice by enzyme-linked immunosorbent assay. Notably, VEGF levels were elevated about 80% in the tumor homogenates and 75% in the serum (P<0.01) of PymT+/0;Tg(CTSB)+/0 mice (Figure 5c). To examine whether elevated VEGF levels were due to increased VEGF expression, we quantified VEGF mRNA levels in primary tumors by quantitative RT–PCR. VEGF mRNA levels were not different in the PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 tumors (Figure 5d), suggesting that elevated VEGF levels in CTSB-overexpressing mice are due to enhanced VEGF processing or mobilization.

Figure 5
figure5

Mean vessel density and VEGF levels (a) Representative images of immunofluorescence staining of the endothelial cell-specific marker CD31 (red staining) and nuclear stain by Hoechst (blue staining) in mammary carcinomas of PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 from 14-week-old mice. (b) Quantification of CD31+ cells as percentage of total cells calculated from 300 high power fields (HPF) on three independent sectional planes per tumor (n=8 per genotype). Data acquisition and analyses were performed using the automated microscope-based imaging platform Scan^R (Olympus, Hamburg, Germany). (c) Quantification of VEGF protein levels by ELISA in PBS-extracts from primary tumors (n=8 per group) and serum from PymT+/0;wt (n=5) and PymT+/0;Tg(CTSB)+/0 mice (n=6) at 14 weeks of age. (d) Quantification of VEGF mRNA levels by quantitative RT–PCR in mammary carcinomas of PymT+/0;wt (n=11) and PymT+/0;Tg(CTSB)+/0 (n=12). Data are presented as means and standard errors; statistical analysis by Student's t-test. Scale bar represents 50 μm.

Cathepsins have also been implicated in the production of other cytokines, for example, IL-1β, and tumor-associated immune cells produce a plethora of cyto- and chemokines (Joyce and Pollard, 2009). However, screening a panel of 12 inflammatory cytokines in phosphate-buffered saline extracts of primary breast cancers by Multianalyte enzyme-linked immunosorbent assay revealed similar cytokine profiles between cancer tissues from PymT+/0;wt and PymT+/0;Tg(CTSB)+/0 mice (Supplementary Figure 4a). Furthermore, the quantification of IL-1β by a second enzyme-linked immunosorbent assay also showed no significant differences in the IL-1β levels (Supplementary Figure 4b), indicating that IL-1β in PymT-induced breast cancers is not correlated to CTSB expression. Hence, the increased VEGF levels in tumors and in the serum of PymT+/0;Tg(CTSB)+/0 may point to a specific role of human CTSB in the release or mobilization of this angiogenic factor.

Discussion

Clinical studies have revealed that overexpression of CTSB correlates with a poor prognosis for cancer patients. We induced the expression of high levels of human CTSB in MMTV-PymT mice to mimic the clinical situation in a mouse breast cancer model in order to analyze the functional consequences of CTSB overexpression. In this study we demonstrate that CTSB expression promotes tumor development and metastasis and is associated with enhanced immune cell infiltration, angiogenesis and VEGF levels.

A growing number of studies demonstrate a positive correlation between inflammation and tumor promotion, thereby emphasizing the clinical relevance of the tumor microenvironment (Pollard, 2004; Mantovani et al., 2008) in addition to the intrinsic properties of the tumor cells themselves (Hanahan and Weinberg, 2000). CTSB has been implicated in tumor progression by its ability to degrade the ECM proteins laminin, fibronectin and collagen (Buck et al., 1992). Accordingly, the inhibition of murine or human cathepsin B by inhibitors or neutralizing antibodies, its knockdown by antisense RNA or genetic ablation, all result in reduced invasion (Premzl et al., 2003; Gondi et al., 2004; Tummalapalli et al., 2007; Sevenich et al., 2010), whereas CTSB overexpression in cell culture has been shown to facilitate invasion (Coulibaly et al., 1999). Strikingly, even if we observed enhanced tumor growth and metastasis in PymT+/0;Tg(CTSB)+/0 mice, CTSB overexpression in PymT tumor cells had either no or only marginal influence on tumor-cell autonomous characteristics such as proliferation and cell death, the potential for colony formation, migration or invasive strand formation. These data indicate a major role for host-derived CTSB on tumor progression. Recently, we demonstrated that the absence of host-derived cathepsin B in mice significantly reduces the formation of lung metastases, and that the cathepsin B expression is induced in macrophages upon their recruitment to metastatic foci in experimentally induced lung colonization (Vasiljeva et al., 2006). Consistent with these findings, histological analyses revealed particularly high expression of CTSB in cells of the tumor stroma in PymT+/0;Tg(CTSB)+/0 mice. The characterization of the cellular composition of tumor-associated immune cells revealed significantly increased numbers of B cells and mast cells in late-stage PymT-induced mammary carcinomas from CTSB-expressing mice. B cells are known to contribute to tumor promotion by establishing an angiogenic and pro-tumoral microenvironment through the secretion of Igs and cytokines to attract innate immune cells to the tumor site (DeNardo and Coussens, 2007). It was shown that genetic elimination of B and T cells limits the formation of hyperplasias during K14-HPV16-induced skin carcinogenesis because of impaired recruitment of innate immune cells, and that adoptive transfer of B cells or serum restores malignant progression (de Visser et al., 2005). Furthermore, the stromal Ig deposits are able to activate the effector functions of myeloid cells through their FCγR (Andreu et al., 2010). In the present study we demonstrate a higher number of tumor-associated B cells in the stroma of PymT cancers, which express high levels of CTSB. We also detected Ig deposits in the stroma of these cancers. The quantification of the stromal Igs was not possible in our setting; however, we were able to demonstrate co-localization of Igs with FCγR, which are known to activate various types of myeloid cells including mast cells (Katz and Lobell, 1995; Takai, 2005). Interestingly, we could detect an increased number of mast cells that also showed enhanced effector function (by counting degranulated mast cells) in the tumors expressing high levels of CTSB. Mast cells are known to promote angiogenesis and tumorigenesis in mouse models of cancer (Coussens et al., 1999). Furthermore, an infiltration by mast cells has been found in a variety of human cancers and has been associated with enhanced cancer growth and invasion (Ribatti et al., 2001). Therefore, we speculate that human CTSB in PymT-induced breast cancers might somehow interfere with the interaction of humoral immunity, that is, B cells and innate immune cells, that is, mast cells.

To better define the role of human CTSB in tumor-associated immunity, we screened a panel of 12 cytokines, including IL-1β. However, the expression of human CTSB in PymT cancers did not alter the levels of these 12 cytokines in the tumors. It is noteworthy that cathepsin B has been implicated in IL-1β processing during inflammasome activation under distinct pathological conditions by mechanisms that require lysosomal damage to release cathepsin B into the cytosol to cleave IL-1β (Halle et al., 2008; Duewell et al., 2010). Our results show that the IL-1β release observed in tumors is unaffected by higher cathepsin B levels, indicating that cathepsin B has no critical impact on IL-1β processing in PymT-induced carcinogenesis. In the light of these unaffected cytokine levels it is important to note that only the concentrations of the VEGF were higher in the phosphate-buffered saline extracts of primary breast tumors from PymT+/0;Tg(CTSB)+/0 mice. This finding is associated with significantly increased numbers of endothelial cells in the human CTSB-expressing cancers, indicating that CTSB promotes tumor angiogenesis.

Angiogenic factors, most notably VEGF, are highly expressed in developing tumors. However, their bioavailability is limited as these factors are sequestered by ECM molecules (Hawinkels et al., 2008). Matrix-remodeling proteases, including cathepsins and MMPs, regulate the release of these factors and thereby activate angiogenesis (Green and Lund, 2005; Hawinkels et al., 2008). Genetic ablation of mouse cathepsin B significantly reduced tumor-associated vasculature during pancreatic islet carcinogenesis (Gocheva et al., 2006). In addition, elevated cathepsin B activities were measured during the formation of the first new vessels in neoplastic tissue (Chang et al., 2009). In light of the increased numbers of degranulated mast cells in human CTSB-expressing PymT-induced cancers, it is of note that the best-known role for mast cells in tumor progression is their ability to induce angiogenesis by providing angiogenic factors, such as VEGF, as well as proteases, including MMPs, cathepsins, mast cell chymase and tryptase (Crivellato et al., 2008). As we observed no alterations in VEGF mRNA levels, our data suggest that CTSB causes either a higher VEGF release or mobilization by proteolytic matrix remodeling, rather than transcriptional regulation of VEGF expression.

Taken together, our results suggest that elevated CTSB activity, either directly or by activating other matrix remodeling proteases, leads to increased secretion or mobilization of VEGF, resulting in enhanced induction of tumor angiogenesis. Increased accumulation of tumor-associated B cells and mast cells might support the establishment of a pro-tumor and angiogenic microenvironment. In consequence, angiogenesis-dependent processes like tumor growth and metastasis are amplified, resulting in an increased tumor and metastatic burden.

In summary, the present data on human CTSB gain-of-function by transgenic overexpression provide direct proof of pro-tumor effects of elevated CTSB levels in an in vivo model of stepwise developing and metastasizing breast cancer. This corroborates a number of previous loss-of-function studies in several cancer mouse models (Gocheva et al., 2006; Vasiljeva et al., 2006), and indicates that human and mouse cathepsin B exert similar tumor-promoting functions. The combined evidence from transgenic and knockout mice suggests CTSB as a prime drug target for cancer therapy among the members of the cysteine cathepsin protease family, and establishes human CTSB-expressing mice as an attractive tool for in vivo trials of cathepsin B inhibitors.

Materials and methods

Additional methods are presented in the online supplement

Mouse model for breast cancer. FVB/N mice harboring the genomic human cathepsin B construct (Tg(CTSB)+/0) (Houseweart et al., 2003) were crossed with the transgenic mouse strain FVB/N-TgN(MMTV-PyVT)634-Mul/J (abbreviated as MMTV-PymT), which develops invasive and metastatic mammary carcinomas (Guy et al., 1992). After the intercross mice were grouped according to their cathepsin B genotypes into mice expressing endogenous cathepsin B levels (PymT+/0;wt) and mice expressing mouse and human cathepsin B (PymT+/0;Tg(CTSB)+/0). The mouse husbandry and the experiments in this study were performed in accordance with the German law for Animal Protection (Tierschutzgesetz) as published on 25 May 1998.

CTSB immunoblotting. Frozen tumor tissue was disrupted in lysis buffer (200 mM sodium acetate, 1 mM EDTA, 0.05% Brij, pH 5.5) using an Ultraturrax disperser (Ika, Staufen, Germany) followed by Dounce homogenization (Wheaton, Millville, USA). Primary PymT tumor cells were broken up in lysis buffer by Dounce homogenization. Postnuclear supernatants were produced by centrifugation at 1500 g for 10 min. Cell conditioned media was collected after 24-h cell culture in serum-free medium and concentrated by centricon tubes. Post nuclear supernatant and cell conditioned media (10 μg protein) were resolved by 12% SDS–PAGE and transferred onto polyvinylidene difluoride membranes. For immunodetection of CTSB, membranes were probed with goat anti-human CTSB (R&D Systems, Wiesbaden, Germany; 1:500 dilution) and secondary antibody anti-goat-POD (Sigma, Hamburg, Germany; 1:5000 dilution), followed by detection with enhanced chemiluminescence reagent (Thermo Scientific, Rockford, IL, USA) and visualization using the Lumi-Imager system and Lumi-Analyst software (Roche, Mannheim, Germany).

Detection of cathepsin B enzyme activity. Cathepsin B proteolytic activity was determined by hydrolysis of the fluorogenic substrate z-Phe-Arg-7-amino-4-methylcoumarin (z-Phe-Arg-AMC) (Bachem, Bubendorf, Switzerland; 25 μmol/l). The release of 7-amino-4-methylcoumarin was monitored by spectrofluorometry for 45 min at excitation and emission wavelengths of 360 and 460 nm, respectively. As z-Phe-Arg-AMC is also degraded by other proteases, that is, cathepsin L, measurements of each sample were performed in the presence and in the absence of the selective cathepsin B inhibitor CA074 (Bachem, Bubendorf; 150 nmol/l). Cathepsin B activity was defined as the CA074-sensitive z-Phe-Arg-AMC hydrolysis and was calculated as the difference of amino-4-methylcoumarin release in aliquots with and without CA074.

Detection of VEGF. Frozen tumor tissue was disrupted in phosphate-buffered saline using an Ultraturrax disperser (Ika) and centrifuged at 1500 g for 5 min at 4°C. Blood samples were allowed to clot for 2 h at room temperature and centrifuged for 20 min at 2000 g to obtain serum. Tumor extract (50 μg protein) and serum (1:5 diluted) were assayed to quantify VEGF protein levels using the Quantikine mouse VEGF immunoassay (R&D Systems) as described by the manufacturer. Optical density was measured at 450 nm with wavelength correction set to 540 nm on a spectrophotometer (Tecan Systems, San Jose, CA, USA).

Data presentation and statistical analyses. The quantitative data are presented as means with their respective standard error, or as statistical box plots. Multiple group comparisons were carried out by ANOVA. For post hoc range tests and pair-wise multiple comparisons the Student's t-test (two-sided) or Mann–Whitney U-test was used. Proportions were compared by the chi-square test. P0.05 was considered to be statistically significant.

References

  1. Allavena P, Sica A, Solinas G, Porta C, Mantovani A . (2008). The inflammatory micro-environment in tumor progression: the role of tumor-associated macrophages. Crit Rev Oncol Hematol 66: 1–9.

  2. Andreu P, Johansson M, Affara NI, Pucci F, Tan T, Junankar S et al. (2010). FcRgamma activation regulates inflammation-associated squamous carcinogenesis. Cancer Cell 17: 121–134.

  3. Barrett AJ . (1992). Cellular proteolysis. An overview. Ann NY Acad Sci 674: 1–15.

  4. Berdowska I . (2004). Cysteine proteases as disease markers. Clin Chim Acta 342: 41–69.

  5. Brix K, Dunkhorst A, Mayer K, Jordans S . (2008). Cysteine cathepsins: cellular roadmap to different functions. Biochimie 90: 194–207.

  6. Buck MR, Karustis DG, Day NA, Honn KV, Sloane BF . (1992). Degradation of extracellular-matrix proteins by human cathepsin B from normal and tumour tissues. Biochem J 282 (Part 1): 273–278.

  7. Caglic D, Kosec G, Bojic L, Reinheckel T, Turk V, Turk B . (2009). Murine and human cathepsin B exhibit similar properties: possible implications for drug discovery. Biol Chem 390: 175–179.

  8. Chang SH, Kanasaki K, Gocheva V, Blum G, Harper J, Moses MA et al. (2009). VEGF-A induces angiogenesis by perturbing the cathepsin-cysteine protease inhibitor balance in venules, causing basement membrane degradation and mother vessel formation. Cancer Res 69: 4537–4544.

  9. Coulibaly S, Schwihla H, Abrahamson M, Albini A, Cerni C, Clark JL et al. (1999). Modulation of invasive properties of murine squamous carcinoma cells by heterologous expression of cathepsin B and cystatin C. Int J Cancer 83: 526–531.

  10. Coussens LM, Raymond WW, Bergers G, Laig-Webster M, Behrendtsen O, Werb Z et al. (1999). Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev 13: 1382–1397.

  11. Crivellato E, Nico B, Ribatti D . (2008). Mast cells and tumour angiogenesis: new insight from experimental carcinogenesis. Cancer Lett 269: 1–6.

  12. de Visser KE, Korets LV, Coussens LM . (2005). De novo carcinogenesis promoted by chronic inflammation is B lymphocyte dependent. Cancer Cell 7: 411–423.

  13. Del Rosso M, Fibbi G, Pucci M, D'Alessio S, Del Rosso A, Magnelli L et al. (2002). Multiple pathways of cell invasion are regulated by multiple families of serine proteases. Clin Exp Metastasis 19: 193–207.

  14. DeNardo DG, Coussens LM . (2007). Inflammation and breast cancer. Balancing immune response: crosstalk between adaptive and innate immune cells during breast cancer progression. Breast Cancer Res 9: 212.

  15. Dennemarker J, Lohmuller T, Mayerle J, Tacke M, Lerch MM, Coussens LM et al. (2010). Deficiency for the cysteine protease cathepsin L promotes tumor progression in mouse epidermis. Oncogene 29: 1611–1621.

  16. Du R, Lu KV, Petritsch C, Liu P, Ganss R, Passegue E et al. (2008). HIF1alpha induces the recruitment of bone marrow-derived vascular modulatory cells to regulate tumor angiogenesis and invasion. Cancer Cell 13: 206–220.

  17. Duewell P, Kono H, Rayner KJ, Sirois CM, Vladimer G, Bauernfeind FG et al. (2010). NLRP3 inflammasomes are required for atherogenesis and activated by cholesterol crystals. Nature 464: 1357–1361.

  18. Egeblad M, Werb Z . (2002). New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2: 161–174.

  19. Elston CW, Ellis IO . (1991). Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 19: 403–410.

  20. Folkman J . (2002). Role of angiogenesis in tumor growth and metastasis. Semin Oncol 29: 15–18.

  21. Gocheva V, Wang HW, Gadea BB, Shree T, Hunter KE, Garfall AL et al. (2010). IL-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. Genes Dev 24: 241–255.

  22. Gocheva V, Zeng W, Ke D, Klimstra D, Reinheckel T, Peters C et al. (2006). Distinct roles for cysteine cathepsin genes in multistage tumorigenesis. Genes Dev 20: 543–556.

  23. Gondi CS, Lakka SS, Yanamandra N, Olivero WC, Dinh DH, Gujrati M et al. (2004). Adenovirus-mediated expression of antisense urokinase plasminogen activator receptor and antisense cathepsin B inhibits tumor growth, invasion, and angiogenesis in gliomas. Cancer Res 64: 4069–4077.

  24. Green KA, Lund LR . (2005). ECM degrading proteases and tissue remodelling in the mammary gland. Bioessays 27: 894–903.

  25. Guy CT, Cardiff RD, Muller WJ . (1992). Induction of mammary tumors by expression of polyomavirus middle T oncogene: a transgenic mouse model for metastatic disease. Mol Cell Biol 12: 954–961.

  26. Halle A, Hornung V, Petzold GC, Stewart CR, Monks BG, Reinheckel T et al. (2008). The NALP3 inflammasome is involved in the innate immune response to amyloid-beta. Nat Immunol 9: 857–865.

  27. Hanahan D, Weinberg RA . (2000). The hallmarks of cancer. Cell 100: 57–70.

  28. Hawinkels LJ, Zuidwijk K, Verspaget HW, de Jonge-Muller ES, van Duijn W, Ferreira V et al. (2008). VEGF release by MMP-9 mediated heparan sulphate cleavage induces colorectal cancer angiogenesis. Eur J Cancer 44: 1904–1913.

  29. Houseweart MK, Pennacchio LA, Vilaythong A, Peters C, Noebels JL, Myers RM . (2003). Cathepsin B but not cathepsins L or S contributes to the pathogenesis of unverricht-lundborg progressive myoclonus epilepsy (EPM1). J Neurobiol 56: 315–327.

  30. Jedeszko C, Sloane BF . (2004). Cysteine cathepsins in human cancer. Biol Chem 385: 1017–1027.

  31. Joyce JA, Pollard JW . (2009). Microenvironmental regulation of metastasis. Nat Rev Cancer 9: 239–252.

  32. Katz HR, Lobell RB . (1995). Expression and function of Fc gamma R in mouse mast cells. Int Arch Allergy Immunol 107: 76–78.

  33. Lin EY, Jones JG, Li P, Zhu L, Whitney KD, Muller WJ et al. (2003). Progression to malignancy in the polyoma middle T oncoprotein mouse breast cancer model provides a reliable model for human diseases. Am J Pathol 163: 2113–2126.

  34. Mai J, Finley Jr RL, Waisman DM, Sloane BF . (2000). Human procathepsin B interacts with the annexin II tetramer on the surface of tumor cells. J Biol Chem 275: 12806–12812.

  35. Mantovani A, Allavena P, Sica A, Balkwill F . (2008). Cancer-related inflammation. Nature 454: 436–444.

  36. Mohamed MM, Sloane BF . (2006). Cysteine cathepsins: multifunctional enzymes in cancer. Nat Rev Cancer 6: 764–775.

  37. Nagy JA, Dvorak AM, Dvorak HF . (2007). VEGF-A and the induction of pathological angiogenesis. Annu Rev Pathol 2: 251–275.

  38. Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T, Naeem R et al. (2005). Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121: 335–348.

  39. Pollard JW . (2004). Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4: 71–78.

  40. Premzl A, Zavasnik-Bergant V, Turk V, Kos J . (2003). Intracellular and extracellular cathepsin B facilitate invasion of MCF-10A neoT cells through reconstituted extracellular matrix in vitro. Exp Cell Res 283: 206–214.

  41. Rawlings ND, Barrett AJ, Bateman A . (2010). MEROPS: the peptidase database. Nucleic Acids Res 38: D227–D233.

  42. Ribatti D, Vacca A, Nico B, Crivellato E, Roncali L, Dammacco F . (2001). The role of mast cells in tumour angiogenesis. Br J Haematol 115: 514–521.

  43. Rozhin J, Sameni M, Ziegler G, Sloane BF . (1994). Pericellular pH affects distribution and secretion of cathepsin B in malignant cells. Cancer Res 54: 6517–6525.

  44. Sevenich L, Schurigt U, Sachse K, Gajda M, Werner F, Muller S et al. (2010). Synergistic antitumor effects of combined cathepsin B and cathepsin Z deficiencies on breast cancer progression and metastasis in mice. Proc Natl Acad Sci USA 107: 2497–2502.

  45. Shin M, Kadowaki T, Iwata J, Kawakubo T, Yamaguchi N, Takii R et al. (2007). Association of cathepsin E with tumor growth arrest through angiogenesis inhibition and enhanced immune responses. Biol Chem 388: 1173–1181.

  46. Takai T . (2005). Fc receptors and their role in immune regulation and autoimmunity. J Clin Immunol 25: 1–18.

  47. Tummalapalli P, Spomar D, Gondi CS, Olivero WC, Gujrati M, Dinh DH et al. (2007). RNAi-mediated abrogation of cathepsin B and MMP-9 gene expression in a malignant meningioma cell line leads to decreased tumor growth, invasion and angiogenesis. Int J Oncol 31: 1039–1050.

  48. Turk V, Kos J, Turk B . (2004). Cysteine cathepsins (proteases)—on the main stage of cancer? Cancer Cell 5: 409–410.

  49. Van Damme J, Struyf S, Opdenakker G . (2004). Chemokine-protease interactions in cancer. Semin Cancer Biol 14: 201–208.

  50. Vasiljeva O, Korovin M, Gajda M, Brodoefel H, Bojic L, Kruger A et al. (2008). Reduced tumour cell proliferation and delayed development of high-grade mammary carcinomas in cathepsin B-deficient mice. Oncogene 27: 4191–4199.

  51. Vasiljeva O, Papazoglou A, Kruger A, Brodoefel H, Korovin M, Deussing J et al. (2006). Tumor cell-derived and macrophage-derived cathepsin B promotes progression and lung metastasis of mammary cancer. Cancer Res 66: 5242–5250.

  52. Vasiljeva O, Reinheckel T, Peters C, Turk D, Turk V, Turk B . (2007). Emerging roles of cysteine cathepsins in disease and their potential as drug targets. Curr Pharm Des 13: 387–403.

  53. Werb Z, Vu TH, Rinkenberger JL, Coussens LM . (1999). Matrix-degrading proteases and angiogenesis during development and tumor formation. Apmis 107: 11–18.

Download references

Acknowledgements

The Tg(CTSB)+/0 mice were kindly provided by Dr L Pennacchio, Genomics Division, Lawrence Berkeley Laboratory. We thank Anne Schwinde, Ulrike Reif, Lisa Christiansson and Anna Heisswolf for excellent technical assistance. This work was supported by European Commission FP7 Grant 201279 (Microenvimet), Deutsche Forschungsgemeinschaft SFB 850 Project B7, and by the Excellence Initiative of the German Federal and State Governments (EXC 294 and GSC-4, Spemann Graduate School).

Author information

Correspondence to T Reinheckel.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Sevenich, L., Werner, F., Gajda, M. et al. Transgenic expression of human cathepsin B promotes progression and metastasis of polyoma-middle-T-induced breast cancer in mice. Oncogene 30, 54–64 (2011). https://doi.org/10.1038/onc.2010.387

Download citation

Keywords

  • protease
  • tumor microenvironment
  • tumor-associated inflammation
  • lysosome

Further reading