Cathepsin B

Protein Function

Cathepsin B is a member of the papain-like family of cysteine proteases and is synthesized as a preproenzyme of 339 amino acids with a calculated molecular weight of approximately 38 kDa [reviewed in (Kirschke et al. 1995; Mort and Buttle 1997)]. Cathepsin B is a bilobal protein with its catalytic site located at the interface between the two lobes (Musil et al. 1991). The amino acids cysteine, histidine and aspartic acid comprise the catalytic triad of the enzyme, with cysteine on the left lobe interacting with histidine on the right lobe to catalyze peptide bond cleavage. Cathepsin B, in contrast to other cysteine cathepsins of the papain-like family, exhibits both endopeptidase and exopeptidase activity. As an endopeptidase, cathepsin B favors amino acids with a large hydrophobic side chain in the P2 site of the protein/peptide substrate (i.e., two residues N terminus of the scissile bond), although it will also accept arginine at this site (Hasnain et al. 1993; Gosalia et al. 2005; Choe et al. 2006). As an exopeptidase, cathepsin B can remove two amino acids (dipeptide) from the C terminus of a polypeptide substrate, thus classifying the enzyme as a peptidyldipeptidase (Cezari et al. 2002; Cotrin et al. 2004). Unlike other cysteine cathepsins, cathepsin B also contains an occluding loop, which consists of 18 amino acids between Pro107 and Asp124. Two salt bridges between the occluding loop and the mature enzyme (at positions His110–Asp22 and Arg116–Asp224) partially block the substrate-binding cleft, thereby promoting its peptidyldipeptidase activity (Musil et al. 1991; Illy et al. 1997). Displacement of the occluding loop is pH dependent and results in the modulation of cathepsin B activity, with exopeptidase and endopeptidase activities being favored at acidic and neutral/alkaline pH values, respectively. Endopeptidase activity at neutral/alkaline pH values is due to disruption of the salt bridges between the occluding loop and the mature enzyme and unfolding of the enzyme (Nägler et al. 1997). In procathepsin B, the occluding loop cannot interact with the active site cleft of the enzyme due to the propeptide that passes directly across the active site cleft (Turk et al. 1996).

In vivo studies also indicate that cathepsin B functions as both an endopeptidase and an exopeptidase. Genetic ablation of cathepsin B is not lethal, nor is genetic ablation of the cysteine protease cathepsin L, which is solely an endopeptidase; however, genetic ablation of both cathepsins B and L is lethal (Sevenich et al. 2006). On the other hand, in a transgenic mouse model for mammary cancer, genetic ablation of cathepsin B and therefore its absence from the mammary cancer cell membrane is compensated for by a redistribution of the cysteine protease cathepsin X, which is solely a carboxypeptidase, to the membrane of these cells (Vasiljeva et al. 2006).

Under normal physiological conditions, active cathepsin B is localized to the endosomal/lysosomal compartment and is primarily involved in routine turnover of both intracellular and extracellular proteins, thus maintaining homeostatic metabolic activity within cells. Other cellular functions for cathepsin B include the regulation of pro-hormone and pro-enzyme activation, antigen processing, inflammatory responses against antigens, tissue remodeling and apoptosis [reviewed in (Mort and Buttle 1997; Reiser et al. 2010)]. For example, in the thyroid, proteolytic processing of the prohormone thyroglobulin (Tg) by secreted cathepsin B (i.e., nonlysosomal) generates the thyroid hormones thyroxine (T4) and triiodothyronine (T3) (Brix et al. 1996; Linke et al. 2002).

Expression and activity of cathepsin B have been correlated with a number of pathologies, including cancer, arthritis, pancreatitis, cardiovascular disease and Alzheimer’s disease. In cancer, overexpression of cathepsin B has been reported in breast, prostate, brain, esophageal, gastric, lung, ovarian, thyroid, skin and colon cancers [reviewed in (Roshy et al. 2003; Jedeszko and Sloane 2004; Mohamed and Sloane 2006)]. Upregulation of cathepsin B expression is predictive of poor prognosis in several tumors, e.g., colon carcinoma (Campo et al. 1994) and ovarian carcinoma (Scorilas et al. 2002). Moreover, a positive correlation between cathepsin B expression and metastasis of carcinoma cells to lymph nodes has been shown in breast (Foekens et al. 1998), prostate (Sinha et al. 2002) and gastric (Czyzewska et al. 2008) cancers. Overexpression of cathepsin B is often accompanied by an altered trafficking of the enzyme to the plasma membrane and secretion into the extracellular milieu [reviewed in (Cavallo-Medved and Sloane 2003; Roshy et al. 2003; Mohamed and Sloane 2006)]. Cathepsin B retains activity at the neutral pH found in the extracellular milieu, particularly in the presence of large extracellular matrix (ECM) proteins (Buck et al. 1992). Cathepsin B directly degrades and remodels the ECM, thereby promoting migration and invasion of tumor cells. In addition, cathepsin B participates in proteolytic networks as an activator of downstream serine proteases and metalloproteases that also contribute to tumorigenesis and invasion. Possible mechanisms for the overexpression of cathepsin B in cancer cells are discussed below in the section ‘Regulation of Concentration’.

Upregulation of cathepsin B has been shown in the synovium of rheumatoid arthritis patients, and in particular in synovial cells at the site of joint destruction (Hansen et al. 2000). In the synovial fluid, cathepsin B participates in collagen degradation, which is inhibited by CA074, a highly selective inhibitor of cathepsin B, consistent with a role for cathepsin B in the progression of the disease (Hashimoto et al. 2001). In osteoarthritis, cathepsin B and its RNA variant CB(−2) (i.e., a cathepsin variant lacking exon 2) are overexpressed in osteoarthritic cartilage and osteophytes as compared with normal cartilage, an effect correlated with an increase in secretion of procathepsin B and its proteolytic activation (Berardi et al. 2001).

Cathepsin B has a role in acute pancreatitis. In this disease, cathepsin B prematurely activates latent precursor digestive enzymes such as trypsinogen within zymogen granules of pancreatic acinar cells, thus leading to destruction of the gland [reviewed in (Lerch and Halangk 2006; van Acker et al. 2006)]. In vitro assays have demonstrated that cathepsin B can activate one of the resident zymogens, trypsinogen, into its active form, trypsin. Moreover, in vivo studies using cathepsin B-deficient mice revealed that 90% of the intrapancreatic trypsinogen activation during pancreatitis is dependent upon cathepsin B activity (Halangk et al. 2000; Van Acker et al. 2002; van Acker et al. 2006).

In atherosclerosis, cathepsin B has roles in lipid metabolism [reviewed in (Lutgens et al. 2007)]. Uptake of modified (i.e., oxidized) low-density lipoprotein (LDL) particles results in localization of cathepsin B within the cytoplasm, most likely via the disruption of lysosomal membranes, and results in the activation of caspases and subsequent apoptosis (Li et al. 2001; Li and Yuan 2004). In atherosclerotic lesions of apolipoprotein E (apoE)-deficient mice, cathepsin B mRNA and protein levels are increased in areas adjacent to the lumen and within macrophages (Chen et al. 2002). In vivo imaging of cathepsin B localized the enzyme in atherosclerotic plaques (Chen et al. 2002), thus suggesting that the enzyme might be a useful diagnostic tool for atherosclerosis. Indeed, anti-atherosclerotic therapies using atorvastatin and glucosamine reduce the cathepsin B-related imaging signal in apoE-deficient mice (Kim et al. 2009). In contrast, a protective role for cathepsin B in atherosclerosis has also been suggested. In smooth muscle cells, inhibition of cathepsin B activity reduces lysosomal degradation of modified LDL and this accumulation of LDL induces foam cell formation (Tertov and Orekhov 1997).

Cathepsin B has also been localized to secretory vesicles of neuronal cells, and its activity within these vesicles produces β-amyloid (Aβ) peptides by cleavage of amyloid precursor protein (APP) at the wild-type (wt) β-secretase site (Hook et al. 2005; Hook et al. 2008). In transgenic mice expressing human APP containing the wt β-secretase site, which is present in most Alzheimer’s disease patients, inhibition of cathepsin B activity results in a reduction of Aβ peptide accumulation, amyloid plaque load, and an improvement in memory deficit (Hook et al. 2008). In cathepsin B knockout mice, there is a decrease in Aβ production and C-terminal β-secretase fragmentation, supporting cathepsin B as a potential therapeutic target for inhibitors to reduce brain Aβ generated from wt APP associated with Alzheimer’s disease (Hook et al. 2009).

Regulation of Activity

Cathepsin B is an important participant within cellular proteolytic networks as determined mainly from in vitro data. One of the activating proteases for cathepsin B is cathepsin D, an aspartic cathepsin (van der Stappen et al. 1996). Cathepsin B is also activated by a number of other proteases, including the serine proteases cathepsin G, urokinase-type plasminogen activator (uPA), tissue-type plasminogen activator (tPA) and elastase (Dalet-Fumeron et al. 1993; Dalet-Fumeron et al. 1996). Additionally, the interaction between cathepsin B and uPA is reciprocal, as cathepsin B can also activate pro-uPA (Kobayashi et al. 1991). Crystal structure data has contributed significantly to understanding the mechanism of procathepsin B activation (Turk et al. 1996; Podobnik et al. 1997; Pungercar et al. 2009). Cathepsin B can undergo autocatalytic activation within the acidic environment of the late endosome and lysosomes. This is a bimolecular process that results in proteolytic removal of the propeptide with subsequent activation of the enzyme (Podobnik et al. 1997; Rozman et al. 1999) being triggered by proenzyme activity (Pungercar et al. 2009). Autoactivation of cathepsin B can be accelerated by binding of the enzyme to polyanionic polysaccharides such as dextran sulfate and naturally occurring glycosaminoglycans, including heparin, heparin sulfate and chondroitin sulfate (Caglic et al. 2007).

Activity of cathepsin B is also modulated by the position of its occluding loop. The occluding loop has 2 histidine residues that at acidic pH interact with the substrate binding cleft and exhibit carboxypeptidase activity (i.e., exopeptidase) of the enzyme. At acidic pH, cathepsin B has been shown to degrade types II, IX and XI collagen (Maciewicz et al. 1990). Displacement of the occluding loop away from the active site cleft, as observed under neutral/alkaline conditions, allows protein substrates access to the active site and the enzyme to function as an endopeptidase (Musil et al. 1991; Illy et al. 1997). Interaction of the occluding loop of cathepsin B with heparin sulfate at the cell surface structurally stabilizes the enzyme under alkaline conditions and potentiates endopeptidase activity (Almeida et al. 2001). This interaction is significant in possibly associating cathepsin B to the plasma membrane and regulating its enzymatic activity.

The final level of regulation of cathepsin B activity is by endogenous inhibitors such as cystatins (Barrett et al. 1986; Turk and Bode 1991; Turk et al. 1997). Cystatins are inhibitors of papain-like cysteine proteases and are subdivided into three families (i.e., cystatin I, II and III) based on their distinct structural details, distribution in the body and physiological roles [reviewed in (Abrahamson et al. 2003)]. The actions of these inhibitors require that the occluding loop of cathepsin B be displaced, allowing the inhibitor to bind to the active site cleft. Binding of cystatin C, an extracellular inhibitor belonging to the cystatin II family, to cathepsin B occurs in two steps: 1) an initial weak binding of the inhibitor to the enzyme, followed by 2) a conformational change in the enzyme that is required to displace the occluding loop and allow for a tighter inhibitor-enzyme bond (Musil et al. 1991; Nycander et al. 1998). Inhibition of cathepsin B by cystatin A/stefin A, a cytosolic cystatin I family inhibitor, has also been reported to occur via a similar two-step mechanism (Pavlova et al. 2000). Due to the occluding loop, inhibition of cathepsin B by protein inhibitors is several orders of magnitude weaker than their inhibition of other endopeptidases such as cathepsins L, V and S. A structural analysis of cystatin A/stefin A binding to cathepsin B suggests that the relocation of the occluding loop is an energy-efficient event (Renko et al. 2010). Other inhibitors of cathepsin B activity include cystatin B/stefin B, also a member of the cystatin I family (Turk et al. 1997); thyropins including equistatin from sea anemone (Lenarcic et al. 1997), saxiphilin from bullfrog (Lenarcic et al. 2000) and chum salmon egg cysteine proteinase inhibitor (Yamashita and Konagaya 1996); and clitocypin from fungi (Brzin et al. 2000). Spi2A, a member of the anti-chymotrypsin family of serine protease inhibitors (serpins), has also been shown to inhibit cathepsin B activity and protect cells from caspase-dependent apoptosis (Liu et al. 2003). Unlike other inhibitors of the anti-chymotrypsin family, Spi2A is not secreted and resides in the cytoplasm.

Interactions with Ligands and Other Proteins

As a constitutively expressed lysosomal protease involved in normal protein turnover, cathepsin B interacts with and hydrolyzes a number of protein substrates. Redistribution of cathepsin B to the cell surface and extracellular milieu allows the enzyme access to ECM protein substrates such as laminin, fibronectin and collagen (Maciewicz et al. 1990; Buck et al. 1992). Although in vitro data has revealed many potential substrates for cathepsin B, with the exception of E-cadherin, little is known about in vivo substrates for this enzyme (Gocheva et al. 2006).

Using a yeast-two hybrid system, several proteins including S100A10, i.e., the light chain of the annexin II heterotetramer (AIIt) (Mai et al. 2000), DP1 and MAGE-3 (Mai and Sloane, unpublished data) have been identified as cathepsin B binding partners. Further in vitro and in vivo analyses (i.e., glutathione S-transferase (GST)-pull down and co-immunoprecipitation assays) revealed that S100A10 binds procathepsin B, but not the mature form of the enzyme (Mai et al. 2000). More specifically, these proteins were shown to co-localize on the surface of both tumor cells (Mai et al. 2000) and endothelial cells (Cavallo-Medved et al. 2009) and within their caveolin-enriched membrane microdomains (Cavallo-Medved et al. 2003; Cavallo-Medved et al. 2005; Cavallo-Medved et al. 2009). AIIt at the cell surface has also been shown to bind heparin (Kassam et al. 1997), which along with heparin sulfate binds cathepsin B at the cell surface (Almeida et al. 2001). The binding of cathepsin B to either heparin or heparin sulfate occurs at the His111 residue of the occluding loop and results in the stabilization of the enzyme at alkaline pH (Almeida et al. 2001). AIIt also binds the serine proteases plasminogen (Kassam et al. 1998) and tPA (Hajjar et al. 1994), which is an activator of cathepsin B. As previously noted, cathepsin B activates pro-uPA (Kobayashi et al. 1991), which in turn is hypothesized to initiate proteolytic events involving the activation of plasminogen, tPA and matrix metalloproteinases (MMPs) at the cell surface [reviewed in (Cavallo-Medved and Sloane 2003). Other binding partners for cathepsin B at the cell surface are α2-macroglobulin and its receptor, which also binds both pro-PA and its inhibitor PAI-1 (Arkona and Wiederanders 1996).

Cathepsin B has been reported to play an important role in apoptosis and has been shown to cleave the proapoptotic Bid and degrade a number of anti-apoptotic proteins: Bcl-2, Bcl-xL, Mcl-1, Bak and BimEL (Cirman et al. 2004; Droga-Mazovec et al. 2008). Using a yeast-two hybrid method, the human homologue of SETA binding protein 1 (hSB1) (Liu et al. 2006b), bikunin and TSRC1 (Liu et al. 2006a) have been identified as potential cathepsin B binding proteins. hSB1 binds the multifunctional adaptor protein SETA; bikunin is a member of the Kunitz-type protease inhibitor family and inhibits trypsin, plasmin and leukocyte elastase, whereas the function of TSRC1 remains unknown. Interactions between these proteins and cathepsin B were confirmed using in vitro GST pull-down assays and in vivo co-immunoprecipitation experiments (Liu et al. 2006a; Liu et al. 2006b). Bikunin, hSB1 and TSRC1 also co-localize with cathepsin B in cellular lysosomes; however, their roles in regulating cathepsin B activity remain unknown. Overexpression of hSB1 and bikunin has been shown to suppress tumor necrosis factor (TNF)-triggered apoptosis in OV-90 ovarian cancer cells, although only bikunin expression reduced cathepsin B activity in these cells (Liu et al. 2006a; Liu et al. 2006b).

Regulation of Concentration

Cathepsin B expression can be regulated at multiple levels. The cathepsin B gene locus is on chromosome 8 (8p22) (Fong et al. 1992), the site of an amplicon that contributes to the malignant progression of esophageal carcinoma. Overexpression of cathepsin B associated with this amplicon is hypothesized to contribute to tumor progression (Hughes et al. 1998). Amplification of the cathepsin B gene has also been found in transformed rat ovarian cells (Abdollahi et al. 1999).

The cathepsin B gene consists of 12 exons (Gong et al. 1993). In human gastric adenocarcinoma cells, there are an additional two small exons, 2a and 2b (Berquin et al. 1995). There have been at least three promoter regions identified as sites for the initiation of transcription [reviewed in (Yan and Sloane 2003)]. The major promoter is located just upstream of exon 1 with other promoter regions upstream of exons 3 and 4. This promoter is TATA-less and GC-rich with a single transcription start site, initially classifying cathepsin B as a housekeeping gene (Berquin et al. 1995). This promoter also contains six Sp1, four Ets, and one USF (E-box) binding sites all located 200 bp upstream of the transcription start site (Yan et al. 2000). Transcriptional activation of the cathepsin B promoter by several transcription factors such as USF1, USF2, Sp1, Sp3, and Ets1 has been shown to have an important role in the regulation of cathepsin B expression (Yan et al. 2000; Yan and Sloane 2003). Sp1 binding to GC-rich regions of the cathepsin B promoter upregulates cathepsin B expression (Yan et al. 2000). In murine melanoma cells, this increase in cathepsin B expression results in increased cell invasion (Szpaderska et al. 2004). Binding of USF1 and USF2 to the E-box is required for cathepsin B promoter activity in both normal and tumor cells (Jane et al. 2002a; Jane et al. 2002b; Yan et al. 2003). On the other hand, a splice variant of USF2, USF2c, binds to the E-box element within the cathepsin B promoter and represses cathepsin B expression (Yan and Sloane 2004). Regulation of cathepsin B expression by Est1 is intriguing as this transcription factor is also upregulated in tumors and is associated with inflammatory and endothelial cells (Yan et al. 2000). Furthermore, Est1 induces expression of other proteases linked to malignant progression (Wasylyk and Wasylyk 1992)

Posttranscriptional regulation of cathepsin B gene expression has also been shown. For example, treatment of cells with phorbol ester increases cathepsin B expression by increasing the stability of the mRNA transcript (Berquin et al. 1999). Alternative splicing of the pre-mRNA of cathepsin B is another regulatory mechanism that mediates expression of the enzyme (Gong et al. 1993; Tam et al. 1994; Berquin et al. 1995). Indeed, biosynthesis of cathepsin B from mRNA transcripts missing exons 2 and 3 is more efficient in comparison to its full length counterpart (Zwicky et al. 2003). Two major mRNA species produced from the cathepsin B gene are 2.3 kb and 4.0 kb transcripts, the latter with an extended 3′ untranslated region (UTR). Alternative splicing in the 5′-UTR results in six splice variants and splicing in the 5′-translated regions results in 2 cathepsin B transcripts, one lacking exon 2 [CB(−2)] and another lacking both exons 2 and 3 [CB(−2,3)] (Gong et al. 1993; Tam et al. 1994). Since the start codon is located in exon 3, the translated product of CB(−2,3) lacks both the signal peptide and the propeptide and therefore is not transported to intracellular vesicular compartments (Gong et al. 1993). Instead the variant remains cytosolic (Mehtani et al. 1998). Preferential expression of the CB(−2,3) splice variant has also been observed in osteoarthritic cartilage from joints with active disease (Berardi et al. 2001; Baici et al. 2006). Although catalytically inactive (Müntener et al. 2005), the truncated forms of cathepsin B have also been shown to be targeted to mitochondria where they induce cell death and to the cytoplasm where they induce nuclear fragmentation in human chondrocytes and HeLa cells (Müntener et al. 2003; Müntener et al. 2004). Functions of these cathepsin B splice variants are discussed below in the section ‘Splice Variants’.

Regulation of cathepsin B at the posttranslational level has also been reported. Transfection of colon epithelial cells with the K-ras4Bval12 oncogene increases cathepsin B expression levels without altering mRNA levels (Yan et al. 1997). Similar results are observed in breast epithelial cells transfected with the c-Ha-ras oncogene (Rozhin et al. 1994). Other posttranslational mechanisms for regulation of cathepsin B expression include altered trafficking of cathepsin B to non-lysosomal compartments as discussed below in the section ‘Subcellular Localization’.

Subcellular Localization

Co-translational translocation of cathepsin B into the rough endoplasmic reticulum (rER) results in the formation of a preproenzyme containing 339 amino acids with the subsequent removal of the signal peptide (17 amino acids). Procathepsin B (43/46 kDa) is N-glycosylated at 2 potential sites (one in the propeptide region and another in the heavy chain of the enzyme). Procathepsin B is shuttled to the Golgi apparatus where further modification of the oligosaccharides to unique high-mannose carbohydrates occurs (Takahashi et al. 1984). Phosphorylation of the mannose residues directs the enzyme to the endosomal/lysosomal compartment via the mannose-6-phosphate receptor pathway. In the endosomal compartment, the enzyme is dissociated from the receptor and the 62-amino acid propeptide is removed from the N terminus, along with 6 amino acids from the C terminus, generating the mature single chain form of cathepsin B (31 kDa). Further processing may occur in the lysosomes, where 2 amino acids between residues 47 and 50 are removed generating a mature double chain form consisting of a heavy chain (25/26 kDa) and a light chain (5 kDa) that are linked via a disulfide bridge [reviewed in (Mort and Buttle 1997)].

Redistribution of cathepsin B to the cell surface has been observed in both normal cells such as cytotoxic T lymphocytes (Balaji et al. 2002) and endothelial cells (Cavallo-Medved et al. 2009) as well as in a number of different types of cancer cells [reviewed in (Cavallo-Medved and Sloane 2003; Roshy et al. 2003)]. Cathepsin B has also been localized to focal adhesions in tumor cells (Rempel et al. 1994; Sameni et al. 1995) and podosomes in transformed fibroblasts (Tu et al. 2008). Moreover, the metastatic potential of three melanoma cell lines is positively correlated with an increase in cathepsin B activity in plasma membrane fractions (Sloane et al. 1986). Subcellular fractionation of metastatic murine melanoma cells revealed the presence of the 31 kDa single chain form of cathepsin B on the plasma membrane as three isozymes with pI values of 5.33, 5.2 and 5.1 (Sloane et al. 1986; Moin et al. 1998). Activity of cell surface cathepsin B has been demonstrated using both real-time enzymatic assays and live-cell proteolysis assays with confocal imaging (Linebaugh et al. 1999; Sameni et al. 2000). Trafficking of cathepsin B to the cell surface can be mediated by several factors including the expression of oncogenic Ha-ras (Sloane et al. 1994) and Ki-ras (Cavallo-Medved et al. 2003). Specific localization of cathepsin B to caveolae on the cell surface was identified in colorectal and breast carcinoma cells as well as endothelial cells (Cavallo-Medved et al. 2003; Cavallo-Medved et al. 2005; Cavallo-Medved et al. 2009). In these cells, cathepsin B is associated with AIIt within these membrane microdomains along with other proteases including uPA and MMP-2 (Cavallo-Medved et al. 2005; Cavallo-Medved et al. 2009). Moreover, a decrease in expression of caveolin-1, the main structural protein of caveolae, leads to a decrease in cathepsin B distribution to caveolae and a decrease in pericellular ECM degradation and invasion (Cavallo-Medved et al. 2005).

Cathepsin B secretion by normal and cancer cells has been shown to occur via both constitutive and inducible pathways (Kuliawat and Arvan 1994; Linebaugh et al. 1999). Secretion of cathepsin B has been suggested to be due to exocytosis of lysosomes, supported by the evidence that treatment of cells with pH 6.5 alters the distribution of vesicles containing cathepsin B to the cell surface resulting in increased secretion of the enzyme (Rozhin et al. 1994). Membrane vesicles shed from B16 melanoma cells, likely to be what are now known as exosomes, contain cathepsin B (Cavanaugh et al. 1983). Other factors including phorbol ester (Guo et al. 2002), 12-(S)-hydroxyeicosatetraenoic acid (12-S-HETE) (Honn et al. 1994) and interferon-γ (Lemaire et al. 1997) also increase cathepsin B secretion.

Cathepsin B has also been localized in the cytoplasm where it acts as a proapoptotic mediator (Guicciardi et al. 2000). The presence of cathepsin B in the cytoplasm is a transient event as a result of damaged lysosomes. Cathepsin B plays a role in TRAIL-mediated apoptosis (Guicciardi et al. 2007) as well as apoptosis mediated by lysosomotropic reagents (Cirman et al. 2004; Droga-Mazovec et al. 2008). In addition, in hepatocytes exposed to TNF, caspase-8 activation is associated with the release of cathepsin B from acidic vesicles into the cytoplasm, which results in the subsequent release of cytochrome c from mitochondria and activation of caspases 9 and 3 (Guicciardi et al. 2000). Moreover, hepatocytes isolated from cathepsin B knockout mice are resistant to TNF-induced apoptosis (Guicciardi et al. 2000). Further studies also identified cathepsin B as a major contributor to an apoptotic cascade upstream of mitochondria in TNF-mediated hepatocyte apoptosis and the progression of the TNF-induced liver damage (Guicciardi et al. 2001). In contrast, others have found that ablation of cathepsin B in primary tumors of MMTV-PymT (mouse mammary tumor virus- polyoma virus middle T oncogene) mice has no effect on TNF-mediated apoptosis (Vasiljeva et al. 2008). In another study, genetic ablation of cathepsin B appeared to increase tumor-associated apoptosis, thus indicating that cathepsin B is not always a pro-apoptotic mediator (Gocheva et al. 2006; Bojic et al. 2007).

Nuclear distribution of cathepsin B has been observed for an artificially truncated form of the enzyme (Bestvater et al. 2005). Localization of truncated cathepsin B to the nucleus is mediated by a signal within the heavy chain domain of the enzyme and results in cell death. More recently, immunofluorescence and biochemical data also demonstrated nuclear localization of a proteolytically active variant smaller than procathepsin B in thyroid carcinoma cells (Tedelind et al. 2010).

Major Sites of Expression

Analysis of the promoter region of cathepsin B indicates that the enzyme is encoded by a housekeeping gene and thus is constitutively expressed in all tissues. Nonetheless, upregulation of cathepsin B mRNA and protein has been observed in cancer cells including prostate, colon, breast, esophageal, gastric, lung, ovarian and thyroid cancer and gliomas and melanomas [reviewed in (Cavallo-Medved and Sloane 2003; Roshy et al. 2003; Jedeszko and Sloane 2004)]. In particular, cathepsin B has been shown to localize to the leading invasive edges of tumors. Cathepsin B staining revealed that the enzyme is distributed to the basal pole of cancer cells. In colon cancer, this redistribution is observed in late adenomas (Yan et al. 1997), a stage subsequent to the activation of Ki-ras. Expression of cathepsin B in stromal cells (i.e., fibroblasts, endothelial cells, macrophages) within the tumor microenvironment has also been reported. In colon cancer, cathepsin B overexpression is observed in both inflammatory macrophages and tumor-associated fibroblasts (Campo et al. 1994). Macrophages associated with breast and prostate tumors also express high levels of cathepsin B (Castiglioni et al. 1994; Fernández et al. 2001). Tumors from transgenic mice expressing human cathepsin B that were crossed with transgenic polyoma virus middle T oncogene breast cancer mice [mouse mammary tumor virus-PymT (MMTV-PymT)] showed increased numbers of tumor-associated B cells, mast cells and CD31+ endothelial cells, which correlate with higher levels of vascular endothelial growth factor (VEGF) in the tumor and serum (Vasiljeva et al. 2006; Sevenich et al. 2011). These data suggest that cathepsin B facilitates increased immune cell infiltration and tumor angiogenesis.

Phenotypes

Cathepsin B knockout mice are both viable and fertile and do not demonstrate any phenotypes that distinguish them from their wild-type (wt) counterparts (Deussing et al. 1998). This study did show that cathepsin B is not a critical factor in major histocompatibility complex II-mediated antigen presentation, as had been proposed (Deussing et al. 1998). However, in cathepsin B knockout mice under challenging conditions, such as experimental pancreatitis, the premature and intracellular activation of trypsinogen that is followed by acinar cell necrosis is largely reduced in the absence of cathepsin B (Halangk et al. 2000). Deletion of the cathepsin B gene in mice reduces TNF-associated hepatocyte apoptosis, by inhibiting mitochondrial release of cytochrome c and caspase 9 and 3 activation and TNF- liver damage and animal mortality (Guicciardi et al. 2001). Excessive accumulation of saturated free fatty acids in liver cells directly induces mitochondrial dysfunction and oxidative stress. This stress was shown to be dependent upon lysosomal disruption and activation of cathepsin B in cathepsin B knockout mice (Li et al. 2008). In transgenic mice expressing human wt-APP, the knockout of cathepsin B significantly reduces the generation of brain Aβ and the APP-derived C-terminal β-secretase fragment, thus supporting the hypothesis that cathepsin B participates in Aβ production from APP containing the wt-β-secretase site (Hook et al. 2008). In a Rip-Tag-2 model, deletion of cathepsin B gene reduces the frequency of initial angiogenic switching in dysplastic progenitors, and impairs subsequent development of the tumor vasculature, leading to reduced tumor growth (Joyce et al. 2004). Moreover, null mutations in any one of the three cathepsins B, L, or S disrupts the progression of tumors to invasive carcinoma, indicating that each enzyme has an important, non-redundant role in the process of tumor invasion (Gocheva et al. 2006). On the contrary, mice deficient in both cathepsins B and L die shortly after birth, with neuronal loss and severe brain atrophy (Felbor et al. 2002), an effect not observed in mice deficient in only cathepsin B or cathepsin L. Clearly these data indicate redundancy between the two enzymes in the maintenance of the central nervous system.

In mammary tumors of transgenic mice expressing human cathepsin B that have been crossed with transgenic MMTV-PymT mice, there is a 20-fold increase in cathepsin B activity (Sevenich et al. 2011). Although cathepsin B overexpression does not affect tumor onset, there is accelerated tumor growth with an increase in tumor weight and a decline in their histopathological grades. In addition, tumors from cathepsin B-overexpressing mice exhibit increased numbers of tumor-associated B cells, mast cells and endothelial cells (Vasiljeva et al. 2006; Sevenich et al. 2011).

Splice Variants

Although the mechanisms for regulating cathepsin B mRNA splicing remain unclear, cathepsin B variants can be subdivided into two subpopulations that give rise to two distinct translation products (Gong et al. 1993; Berquin et al. 1995; Mehtani et al. 1998; Berardi et al. 2001). The first group of variants lacks exon 2 [CB(−2)] and was originally observed in malignant tumors overexpressing cathepsin B (Gong et al. 1993). Although CB(−2) encodes the same enzyme as the full-length CB transcript and its distribution is unaltered, its translation rate is doubled (Gong et al. 1993). This altered rate of translation and hence increased expression may augment pathologies including tumorigenesis (Gong et al. 1993) and osteoarthritis (Berardi et al. 2001). The second group lacks exons 2 and 3 [CB(−2,3)], thus translation starts at an initiation codon at position 53 within exon 4. This mRNA variant is translated into a naturally truncated form of cathepsin B (Δ51CB) that is not distributed to intracellular vesicles, i.e., endosomal/lysosomal compartments, and has no cathepsin B enzymatic activity (Müntener et al. 2004; Müntener et al. 2005). Δ51CB is more prominently expressed in tumors (Mehtani et al. 1998) and arthritic tissues/cells (Berardi et al. 2001). Δ51CB is also found to be associated with the nucleus, the cytoplasmic side of intracellular membranes (Mehtani et al. 1998) and mitochondria (Müntener et al. 2004). Furthermore, its overexpression stimulates nuclear fragmentation and cell death (Müntener et al. 2003; Bestvater et al. 2005).

Antibodies

Several monoclonal and polyclonal antibodies to both human and murine cathepsin B are available both commercially and from individual laboratories. Polyclonal antibodies are available that recognize all forms of the enzyme. Monoclonal antibodies have been developed that recognize specific epitopes in the mature enzyme, as well as others that recognize only epitopes within the propeptide. These antibodies detect cathepsin B in immunoblots and in both cultured cells and histological tissue sections. In addition, a monoclonal antibody 2A2 has been developed that regulates cathepsin B activity by inducing a change in conformation of the enzyme. This change results in stabilization of the exopeptidase conformation and neutralization of the endopeptidase activity (Mirković et al. 2009).