Published as doi: 10.1096/fj.06-7924com.
(The FASEB Journal. 2007;21:3029-3041.)
© 2007 FASEB
Cathepsin cysteine proteases in cardiovascular disease
Suzanne P. M. Lutgens,
Kitty B. J. M. Cleutjens,
Mat J. A. P. Daemen and
Sylvia Heeneman1
Department of Pathology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University, Maastricht, The Netherlands
1Correspondence: Department of Pathology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands. E-mail: s.heeneman{at}path.unimaas.nl
 |
ABSTRACT
|
|---|
Extracellular matrix (ECM) remodeling is one of the underlying mechanisms in cardiovascular diseases. Cathepsin cysteine proteases have a central role in ECM remodeling and have been implicated in the development and progression of cardiovascular diseases. Cathepsins also show differential expression in various stages of atherosclerosis, and in vivo knockout studies revealed that deficiency of cathepsin K or S reduces atherosclerosis. Furthermore, cathepsins are involved in lipid metabolism. Cathepsins have the capability to degrade low-density lipoprotein and reduce cholesterol efflux from macrophages, aggravating foam cell formation. Although expression studies also demonstrated differential expression of cathepsins in cardiovascular diseases like aneurysm formation, neointima formation, and neovascularization, in vivo studies to define the exact role of cathepsins in these processes are lacking. Evaluation of the feasibility of cathepsins as a diagnostic tool revealed that serum levels of cathepsins L and S seem to be promising as biomarkers in the diagnosis of atherosclerosis, whereas cathepsin B shows potential as an imaging tool. Furthermore, cathepsin K and S inhibitors showed effectiveness in (pre) clinical evaluation for the treatment of osteoporosis and osteoarthritis, suggesting that cathepsin inhibitors may also have therapeutic effects for the treatment of atherosclerosis.—Lutgens, S. P. M., Cleutjens;, K. B. J. M., Daemen, M. J. A. P., Heeneman, S. Cathepsin cysteine proteases in cardiovascular disease.
Key Words: atherosclerosis • extracellular matrix • lipid metabolism • diagnostic tool • therapeutic potential
|
INTRODUCTION
|
|---|
REMODELING OF THE extracellular matrix (ECM) is an important feature of many physiological and pathological processes. The ECM consists of elastins, collagens, and proteoglycans and is largely synthesized by smooth muscle cells (SMCs). It provides anchorage, support, and structure to tissue. Proteolytic enzymes, such as matrix metalloproteinases (MMPs) and cathepsin cysteine proteases, can degrade the ECM and therefore contribute to (patho) physiological processes, like atherosclerosis, aneurysm formation, neointima formation, and neovascularization.
In vitro and in vivo studies have been performed to establish the contribution of proteolytic enzymes to the etiology of cardiovascular disease. The contribution of several MMPs to ECM degradation in cardiovascular diseases has been highlighted in several excellent reviews (1
2
3)
. MMPs are not the only proteins required for ECM degradation; other proteases like cathepsin cysteine proteases are needed as well. Cathepsins of the cysteine protease family are localized in lysosomes and endosomes, and degrade intracellular or endocytosed proteins. These cathepsins have been proven to play an important role in cardiovascular diseases (4
, 5)
. In this review we will focus on the role of the cysteine protease family of the cathepsins in cardiovascular disease.
|
Cathepsin cysteine proteases in a nutshell
|
|---|
Proteases are enzymes that catalyze the irreversible hydrolysis of amide bonds (6)
. Different proteases catalyze this reaction with different strategies. These different properties are used to distinguish four major groups of proteases: cysteine, serine, aspartate, and metallo-proteases (7)
. The cysteine proteases can be subgrouped into several families including the family of enzymes related to interleukin-1ß-converting enzyme (ICE), the calpain family, and the papain family (7
, 8)
. The cysteine proteinases of the papain family are the most abundant of all cysteine proteases. The family consists of papain, related plant proteinases, and lysosomal cathepsins B, C, F, H, K, L, O, S, V, W, and X (8
, 9)
. Cathepsin N and T have also been described, but a detailed characterization is still lacking (8)
. Most cathepsins are endopeptidases (8
, 10)
, although cathepsins B and H may also function as a dipeptidyl carboxypeptidase (11)
and as an aminopeptidase (12)
, respectively. Cathepsin C is an aminodipeptidase (8
, 10)
and cathepsin X is a carboxy-mono or -dipeptidase (13)
. The activation process of cathepsins has been thoroughly reviewed by Turk et al. (9)
. In short, cathepsins are synthesized as preproenzymes. Procathepsin is formed after removal of the prepeptide during the passage to the endoplasmic reticulum. Subsequently, the active cathepsin can be produced after proteolytic removal of the propeptide in the acidic environments of late endosomes or lysosomes. This last process is accompanied by the action of several proteases, such as pepsin, neutrophil elastase, and various cysteine proteases. The propeptide is described to serve several functions such as stability, proper folding, targeting of cathepsins, and prevention of inappropriate protease activity (9)
. Most cathepsins like cathepsins B, F, H, K, L, and V are optimally active in acidic environments and are only weakly active at neutral pH (9)
. In contrast, cathepsin S activity is optimal at neutral pH (9)
.
Cysteine protease inhibitors can be subgrouped in several superfamilies, of which the cystatins are the most important (14)
. The cystatin superfamily can be divided into three groups. Type 1 cystatins are stefin A and B, also called cystatin A and B, which are localized mostly intracellular. Type II cystatins, among which is cystatin C, act extracellularly. Type III cystatins comprise the kininogens, the circulating proteins (14
, 15)
. The function of these cystatins is to protect against lysosomal proteins, such as cathepsins, which have accidentally escaped from lysosomes or which are occasionally released during, for example, apoptosis or phagocyte degranulation. Cystatin C shows the highest inhibiting properties to cathepsins L and S, followed by cathepsins B and H (16)
.
In humans, several cathepsin deficiencies have been described. A loss of function mutation in the cathepsin C gene leads to Papillon-Lefèvre syndrome, an autosomal recessive disorder characterized by palmoplantar hyperkeratosis and severe early onset periodontitis (17)
. Deficiency of cathepsin K leads to pycnodysostosis, an autosomal recessive osteochondrodysplasia characterized by osteosclerosis and short stature (18)
. Cathepsin deficiencies have also been generated in mice. Although a role for cathepsin B was suggested in antigen presentation, the phenotype of cathepsin B-deficient mice appeared normal compared with wild-type littermates and antigen presentation was not affected (19)
. Deficiency of cathepsin F causes lysosomal storage defects (neuronal lipofuscinosis) and late-onset neurological disease in mice (20)
. An osteopetrotic phenotype with excessive trabeculation of the bone marrow space has been described for mice deficient in cathepsin K (21)
. Mice lacking cathepsin L develop periodic hair loss, with alteration of hair follicle morphogenesis and cycling as well as hyperplasia and hyperkeratosis of the epidermis, which is attributed to hyperproliferation of hair follicle epithelial cells and basal keratinocytes. Although cathepsin S-deficient mice have normal fertility and show no gross abnormalities or major developmental abnormalities in their lymphocytes, they demonstrate inhibition of invariant chain (Ii) degradation, resulting in slow peptide loading by MHC class II/Ii p10 (22
, 23)
. Ondr et al. reported no abnormalities for cathepsin W-deficient mice (24)
.
|
Cathepsin cysteine proteases: mechanisms of action on a cell biological level
|
|---|
In this section we highlight some of the mechanisms by which cathepsins influence ECM turnover, inflammation, and apoptosis on a cell biological level. By influencing these processes cathepsin cysteine proteases may contribute to cardiovascular diseases including atherosclerosis, neointima formation, and aneurysm formation.
ECM degradation by cathepsin cysteine proteases contributes to a variety of physiological and pathological conditions, including cancer, bone remodeling, and cardiovascular disease (4
, 5
, 7)
. Cathepsin cysteine proteases are folded into two relatively large globular domains surrounding a cleft that contains the active site residues. Cleavage is achieved by substrate entry into the cleft (9)
. Individual cathepsins have specific substrate preferences. Cathepsin B and L have been described to degrade rat collagen type II, IX, and XI at acidic pH values (25)
. Cathepsin L was somewhat more efficient than cathepsin B. Cathepsin K exerted strong elastolytic activity and was once suggested to be the most potent elastase at neutral pH (26)
and collagenase at acidic pH (26
, 27)
. However, more recent findings indicate cathepsin V as a more potent elastase than cathepsin K (28)
. Cathepsin S was also shown to be a strong elastase (29)
.
Besides its role in ECM degradation, cathepsins are involved in major histocompatibility complex (MHC) class II antigen presentation. MHC class II molecules are expressed on the surface of antigen-presenting cells (APCs); they bind exogenous proteins and present them to CD4+ T cells. MHC class II 
ß heterodimers assemble in the endoplasmic reticulum (ER) with the assistance of the invariant chain (Ii). The Ii cytoplasmic tail targets the MHC class II-Ii complex to the endosomal pathway and prevents early loading of antigenic proteins on MHC class II with class II-associated invariant chain peptide (CLIP). Maturation of the early endosome leads to activation of lysosomal enzymes, including cathepsins, which degrade Ii. The invariant chain Ii is sequentially degraded in IiP22, IiP10, and CLIP. After Ii degradation, the MHC class II peptide binding groove remains occupied by CLIP, preventing premature peptide loading. Removal of CLIP and loading of peptides is mediated by MHC-like molecule HLA-DM in humans and H-2M in mice (reviewed in refs. 30
, 31
). In professional APCs, cathepsin F (macrophages), L (cortical thymic epithelial cells), and S (B cells, dendritic cells and macrophages) have been described to degrade MHC class II-associated Ii (23
, 32
33
34
35)
, while cathepsin B seems not to play a pivotal role in antigen presentation (19)
. Cathepsin S, but not L, also efficiently degrades MHC class II-associated Ii in nonprofessional APCs, such as (intestinal) epithelial cells (36
, 37)
.
Furthermore, cathepsin cysteine proteases are involved in apoptosis. Under physiological circumstances, cathepsins are located in the lysosomes. However, lysosomal permeabilization due to exogenous oxidants (reactive oxygen species) may induce lysosomal leakage leading to release of cathepsins into the cytoplasm (38)
. In vitro studies using cultured macrophages showed that both 7ß-hydroxycholesterol and oxidized LDL (oxLDL) induced lysosomal destabilization, leading to leakage of cathepsins B and L to the cytoplasm, activation of caspases, and subsequent apoptosis. This suggests that cathepsins B and L may act as cleaving enzymes during the apoptotic process (39
, 40)
. Furthermore, suppression of cathepsin B by serine protease inhibitor 2A (Spi2A, with cross-reactivity to cathepsin B) inhibits the lysosomal apoptosis pathway and protects cells from TNF- induced apoptosis (41)
. NF-
B can protect cells from death after TNF receptor stimulation by reducing cathepsin B activity in the cytosol, mediated by the up-regulation of Spi2A. Thus, NF-B protects cells from apoptosis by inhibition of cathepsin B activity.
|
Cathepsins in atherosclerosis
|
|---|
More than a decade ago it was shown that human macrophages secreted active cathepsins B, L, and S and showed elastolytic activity (42)
. More recently, both mRNA and protein levels of cathepsin cysteine proteases B, L, and S were found to be increased in murine atherosclerotic lesions. Cathepsin protein expression was localized in macrophages and/or in lipid-rich areas (43)
. Cathepsin K and S were the first cathepsins that were found to be expressed in human atherosclerotic lesions (44)
.
Below we will extensively describe the role of cathepsins in atherosclerosis and therefore we have subdivided the paragraphs into segments describing the expression, knockout models and ECM remodeling, shear stress, and lipid metabolism and inflammation.
|
Cathepsins in atherosclerosis: expression
|
|---|
Expression patterns of cathepsins in atherosclerosis are summarized in Table 1
. Cathepsin B mRNA and protein levels were found to be increased in atherosclerotic lesions of apoE-deficient mice, and cathepsin B immunoreactivity was highest in areas next to the lumen and in macrophages (45)
. Cathepsin F was only weakly expressed in normal human arteries, but in human atherosclerotic lesions it was localized in macrophages devoid of intracellular lipid and in SMCs and ECs but not in T-lymphocytes (46)
. Protein levels of cathepsin L were also increased in human atheroma. Cathepsin L was localized mainly in SMCs, ECs, and macrophages in advanced atherosclerotic lesions (47)
. Cathepsin K expression in normal arteries is low. Early human atherosclerotic lesions showed cathepsin K expression in the intima and medial SMCs. In advanced atherosclerotic plaques, cathepsin K was localized mainly in macrophages and SMCs of the fibrous cap. Cathepsin K protein levels were increased in atherosclerotic lesions when compared with normal arteries, whereas cathepsin K mRNA levels were similar in both atherosclerotic and normal tissues (44)
. We recently described that cathepsin K mRNA and protein levels were highest in advanced but stable human atherosclerotic plaques compared with early atherosclerotic lesions and lesions containing a thrombus. In these lesions, cathepsin K was localized in SMCs and macrophages, and also in ECs (48)
(Fig. 1
). The most extensively described cysteine protease in atherosclerosis is cathepsin S. Protein expression studies showed that normal human arteries sparsely expressed cathepsin S. Early human atherosclerotic lesions, or fatty streaks, showed cathepsin S expression in the intima and in medial SMCs. In advanced human atherosclerotic plaques, cathepsin S was localized in macrophages and SMCs of the fibrous cap (44)
. ECs lining the lumen of the vessel itself and the plaque microvessels also expressed cathepsin S (4)
. Furthermore, cathepsin S mRNA and protein levels were increased in human atheroma compared with normal arteries (44)
. Recently, cathepsin V was also found to be expressed in human atheroma (28)
.
View larger version (52K):
[in this window]
[in a new window]
|
Figure 1. Overview of cathepsin expression and activity in atherosclerotic plaque. Cathepsins are expressed in endothelial cells (EC), smooth muscle cells (SMC), and macrophages (M ). The ECM, containing elastin and collagen, is degraded by cathepsins L, K, S, and V. Cathepsin S may even induce plaque rupture. Cathepsins F and S contribute to macrophage foam cell formation by reducing cholesterol efflux, which is counteracted by cathepsin K as a result of increasing lipid uptake.
|
|
|
Cathepsins in atherosclerosis: knockout models and ECM remodeling
|
|---|
The effects of cathepsin deficiency on ECM remodeling in in vivo mouse models are summarized in Table 1
. Deficiency of cathepsin S in LDL receptor-deficient mice resulted in a reduction in atherosclerotic plaque area (60% reduction after 12 wk atherogenic diet), a reduction in plaque stage development, and a reduction in the number of elastin breaks and elastase activity (49)
. Furthermore, cathepsin S deficiency led to a reduction in SMC and collagen content and fibrous cap thickness. Cathepsin S-deficient macrophage showed reduced transmigration through an EC monolayer and collagen type I and IV in vitro (49)
. In another study using cathepsin S-deficient apoE–/– mice, no differences in the number of elastin breaks were found, although plaque elastin content was reduced by 49% (50)
. Cathepsin S deficiency reduced atherosclerotic plaque size by 46% (after 12 wk atherogenic diet) and reduced the number of plaque ruptures (defined as visible defects in the cap, accompanied by intrusion of erythrocytes into the region below it) by 73%. In concordance with previous findings (44)
, active cathepsin S was found in extracts of human atherosclerotic lesions, but not in normal arteries, and cathepsin S expression in macrophages colocalized with areas of elastin fragmentation (50)
. These data all point to a protective role of cathepsin S inhibition in atherogenesis by decreasing degradation of ECM components.
Cathepsin K deficiency in apoE–/– mice resulted in a 42% reduction in atherosclerotic plaque area; although the total number of plaques remained unchanged, there was a relative increase in early lesions and a relative decrease of the number of advanced lesions when cathepsin K was absent. Furthermore, cathepsin K deficiency led to an increase in collagen content and macrophage size, a decrease in elastin breaks and macrophage content, but unchanged T-lymphocyte content and lipid core area. The increased macrophage size resulted from an increase in lipid uptake in macrophages, which was stored in lysosomes that had increased in size (48)
. Subsequent microarray and pathway analysis suggested that cathepsin K deficiency altered plaque phenotype not only by decreasing proteolytic activity, but also by stimulating TGF-ß signaling, and suggested a role for caveolin-1 and CD36 in the lipogenic phenotype of cathepsin K-deficient atherosclerotic lesions (51)
. These data suggest that cathepsin K deficiency may have a protective role in atherosclerosis by increasing fibrosis; however, cathepsin K deficiency also aggravates foam cell formation, at least in mice, which may affect plaque stability.
Table 1
summarizes the effects of cathepsins on ECM remodeling activities in vitro. It was shown that cathepsin V shows the highest elastolytic activity, followed by cathepsins K, S, F, L, and B. However, in cultured macrophages cathepsins K, S, and V contribute equally (20%) to elastolytic activity, of which 20% takes place intracellularly and 40% extracellularly (28)
. Extracts from human atheroma showed a twofold increase in elastolytic activity compared with normal arteries, which could be inhibited up to 40% by a cysteine protease inhibitor (E64). Cytokine-stimulated SMCs also showed elastolytic activity (in vitro), which could be inhibited for >80% by a selective cathepsin S inhibitor (LHVS) or a cysteine protease inhibitor (E64) (44)
. Other studies demonstrated that cathepsin S inhibition in (unstimulated) SMCs by either LHVS, or the cysteine protease inhibitor E64 cystatin C reduced invasion through an elastin gel by 
90%, while invasion through a collagen type I gel was reduced by only 30%. No effect on adhesion and migration was found, indicating that cathepsin S could exert its effect in atherosclerotic lesions via elastolysis (52)
. Cathepsin S also exerted collagenolytic activities although less potent than elastolytic activities. The elastolytic and collagenolytic activity (type I) of cathepsin L in SMCs and ECs has already been mentioned to increase after bFGF stimulation (47)
. Cathepsin K not only contributes to elastinolysis, although cathepsin L and S may take over when cathepsin K is absent (53)
, but also plays an important role in the degradation of collagen type I, and therefore its collagenolytic activity is unique among proteases (27)
.
|
Cathepsins: shear stress sensitive proteases
|
|---|
During the initiation phase of atherosclerosis, hemodynamic forces at branch points act on the endothelial layer (shear stress) leading to increased permeability of the endothelium to lipoproteins, expression of endothelial adhesion molecules, and induction of leukocyte migration into the arterial wall. Oscillatory shear stress corresponds to atheroprone areas, while laminar shear stress corresponds to atheroprotected areas (54)
. It was recently suggested that secreted cathepsin L is a shear-dependent matrix protease (55)
. Inhibition of cathepsin L by a cysteine protease inhibitor or cathepsin L silencing RNA (siRNA) treatment inhibited oscillatory shear stress-induced gelatinase and elastase activity by mouse aortic ECs, but not laminar shear stress-induced activity (55)
. This suggests that cathepsin L is a shear-sensitive protease with potential importance in vascular remodeling and atherosclerosis. The same researchers also showed that cathepsin K is shear stress regulated, since cathepsin K siRNA reduced oscillatory shear stress-induced gelatinase and elastase activity by mouse aortic ECs (56)
. Thus, it can be suggested that cathepsin K and L activity is increased by oscillatory shear stress, which leads to increased ECM remodeling and atherosclerosis.
|
Cathepsins in atherosclerosis: lipid metabolism and inflammation
|
|---|
The role of cathepsins in lipid metabolism is summarized in Table 1
. If lipids are oxidized, the resulting oxysterols, hydroxyperoxides, and their toxic carbonylic fragments may affect lysosomal enzymes and membranes. Uptake of modified (oxidized) LDL damages the lysosomal membrane and, as a result, the acidic components will flow into the cytosol (57)
. Modified LDL relocates cathepsins B and L from the lysosome into the cytosol, probably as a result of disruption of the lysosomal membrane. It also reduces overall cathepsin L activity, but induces the relative cytoplasmic activity. Lysosomal cathepsins that are translocated to the cytoplasm may act as cleavage enzymes during apoptosis (39)
. These harmful effects of modified LDL on lysosomal membrane damage and cathepsin L relocation and inactivation can be partly prevented by high density lipoprotein (HDL) and/or vitamin E (57)
.
DiI-labeled modified (oxidized) LDL uptake studies showed an increased uptake of modified LDL in cathepsin K-deficient apoE–/– bone marrow (BM) -derived macrophages (48)
. Subsequent pathway analysis revealed that the increased lipid uptake is mediated by both CD36 and caveolins (51)
. In addition, cathepsin K-deficient apoE–/– BM-derived macrophages showed an increase in cholesterol ester storage compared with apoE–/– BM-derived macrophages, which was stored in large lysosomal compartments (48)
. These data indicate that deficiency of cathepsin K aggravates foam cell formation.
Cathepsins may also play a role in (modified) LDL degradation. Inhibition of cathepsin B inhibited naturally occurring modified LDL degradation in human aortic SMC lysates by 41% (pH 4.0 and pH 5.5). Decreased lysosomal degradation may lead to LDL accumulation in SMCs and subsequent foam cell formation (58)
. In vitro studies showed that recombinant cathepsin F extensively degraded apoB-100 (60% degradation at pH 6.0), while cathepsins K and S showed less extensive degradation (10% and 20% respectively at pH 6.0) (46)
. Increasing pH led to a reduction in the degrading capacity of cathepsins F and K but affects cathepsin S to a much lesser extent. Degradation (proteolytic modification) of apoB-100 by cathepsin F, but not cathepsins K and S, led to aggregation and fusion of LDL particles and increased the ability of LDL to bind proteoglycans, subsequently leading to the accumulation of extracellular lipid droplets. These data suggest that cathepsins B, F, K, and S contribute to extracellular lipid accumulation in the arterial wall, a key feature of atherosclerosis.
Literature provides ample evidence that cathepsins are also involved in cholesterol efflux. Lindstedt et al. showed that cathepsins F and S reduced the ability of cholesterol efflux from macrophages by 50% in vitro due to proteolysis of preß-HDL (59)
. Furthermore, cathepsin S totally (100%) degraded lipid-free apolipoprotein A-1 (apoA-1) leading to complete loss of the ability of apoA-1 to stimulate cholesterol efflux. Cathepsins F and K also partially degraded apoA-1 leading to a reduction of cholesterol efflux of 30% and 15% respectively (59)
. These data suggest that reduction of cholesterol efflux by cathepsins F, K, and S mediated degradation of cholesterol acceptors may contribute to the preservation of foam cells in the atherosclerotic lesion. In a recent study from our laboratory, however, deficiency of cathepsin K in apoE–/– bone marrow-derived macrophages reduced both HDL and D37F (a specific apoA-1 mimetic peptide) mediated cholesterol efflux by 20% and 15%, respectively, compared with apoE–/– bone marrow-derived macrophages (S. Lutgens, unpublished observations). Although our results contradict the results of Lindstedt et al., these findings can be explained. Lindstedt et al. studied the extracellular capacity of cathepsin K to reduce cholesterol efflux by degradation of preß-HDL and apoA-1 (59)
. However, by using cathepsin K-deficient apoE–/– bone marrow derived macrophages, we studied the intracellular effect of cathepsin K deficiency on cholesterol efflux without prior degradation of the cholesterol acceptor by cathepsins. Thus, extracellular cathepsin K reduces cholesterol efflux by decreasing the amount of cholesterol acceptors, while intracellular deficiency of cathepsin K reduces the reverse cholesterol transporting capacity independent of cholesterol acceptor degradation.
The role of cathepsin cysteine proteases in lipid uptake, storage, and efflux has been partly elucidated. However, the most important question of whether the role of cathepsins in lipid metabolism is atherosclerosis-stimulating or -protective remains unanswered. Relocation of cathepsins B and L from the lysosome into the cytosol, where they may act as cleavage enzymes in apoptosis, may eventually contribute to the formation of the necrotic core and can be considered an atherosclerosis-stimulating role for cathepsins B and L. On the other hand, inhibition of cathepsin B reduced lysosomal degradation of modified LDL, thereby inducing foam cell formation, which can be regarded as an atherosclerosis-protective role for cathepsin B. A similar case can be made for the role of cathepsin K in cholesterol efflux. Lindstedt et al. suggested a role for cathepsin K in reducing cholesterol efflux (atherosclerosis-stimulating role for cathepsin K), while we suggested that cathepsin K deficiency reduced cholesterol efflux (an atherosclerosis protective role for cathepsin K).
Evidence for a role for cathepsins in the inflammation process in atherosclerosis is sparse. Sukhova et al. showed that cathepsin S deficiency in atherosclerotic apoE-deficient mice led to a reduction in macrophage and lipid content, in the number of T-cells, and in IFN-
content (49)
. We have already mentioned that cathepsins F, L, and S can degrade MHC class II-associated Ii. However, data defining the role of cathepsin cysteine proteases in antigen processing and presentation related to atherosclerosis are lacking. Besides a direct role for cathepsins in inflammation, they can also play an indirect role via reducing TGF-ß expression levels. Inhibition of TGF-ß in apoE–/– mice resulted in an inflammatory plaque phenotype with an increased inflammatory cell content (60)
. We recently showed that deficiency of cathepsin K induced an increased expression of genes involved in TGF-ß signaling in atherosclerotic lesions of apoE–/– mice, suggesting that cathepsin K deficiency may decrease inflammation by inducing TGF-ß activity (51)
.
|
Role of natural cathepsin inhibitors in atherosclerosis
|
|---|
The effects of cathepsins in atherosclerotic lesions are balanced by cystatin C. Cystatin C protein levels decreased in human atherosclerotic plaques (61)
. Elastinolytic activity of IFN- stimulated vascular SMCs was inhibited by cystatin C. TGF-ß1 stimulated SMC cystatin C secretion and subsequently blocked SMC elastase activity (61)
. Two groups studied the effects of cystatin C deficiency on atherosclerosis in apoE-deficient mice, but found contradictory effects (62
, 63)
. One group studied the effect of cystatin C deficiency in female apoE–/–-deficient mice fed a high-fat diet (21% cocoa fat, 0.15% cholesterol, and 0% sodium cholate) for 25 wk. They observed an increase in plaque size at the aortic root, whereas no effect on collagen content was observed. Lipid content tended to be larger and total macrophage content increased in the absence of cystatin C (62)
. In the aortic arches of cystatin C-deficient male apoE-deficient mice fed a "Western" type diet (20.1% saturated fat, 1.37% cholesterol, and 0% sodium cholate) for 12 wk, elastic lamina degradation, SMC, and collagen content increased, indicating that disruption of the elastic lamina may facilitate SMC migration. Lesion size did not differ between these cysC–/–/apoE–/– and apoE–/– mice. Levels of cathepsin B, L, and S in aortic extracts were increased in the absence of cystatin C. Cytokine or growth factor stimulated SMCs also showed increased production of these cathepsins when cystatin C was absent. Both aortic extracts and (stimulated) SMCs showed higher elastolytic activity in the absence of cystatin C. Lipid content was reduced, while macrophage content and T cell content were unchanged (63)
. Although data from both studies are contradictory in some respects, it seems that deficiency of cystatin C facilitates the atherosclerotic process either by increasing plaque size or by increasing elastolytic activity, and thus ECM degradation.
|
Cathepsins in restenosis and neointima formation
|
|---|
As mentioned, cathepsins contribute to ECM degradation, suggesting a possible role for cathepsins in restenosis and neointima formation. After percutaneous coronary intervention, luminal size may decrease as a result of constrictive remodeling. Placement of a stent may induce neointima formation and contribute to restenosis (64)
. In a balloon injury model of restenosis in hypercholesterolemic rabbits, cathepsin S mRNA and protein expression was increased. Protein expression was increased in SMCs and macrophages. Cystatin C mRNA and protein expression were only minimally up-regulated (65)
. A carotid balloon injury model of restenosis in rats on a normal diet also showed increased mRNA and protein levels of cathepsin S, but also cathepsin K, whereas cystatin C mRNA and protein were not increased (66)
. These studies imply that there is a relatively larger increase in cathepsin activity than cystatin activity. In vitro studies demonstrated that tissue extracts from whole-mount balloon injured carotid arteries showed an increase in elastolytic and a minor increase in collagenolytic activity (66)
. Furthermore, it was demonstrated that cathepsin S degrades laminin, fibronectin, and collagen type I and that SMC migration through an in vitro basement membrane matrix could be inhibited by a selective cathepsin S inhibitor (65)
.
As summarized in Table 2
, the increased expression level of cathepsins K and S during neointima formation and the increased ECM degrading potential of cathepsin S suggest that these proteases are involved in neointima formation. Degradation of the ECM may facilitate the migration and invasion of SMCs and macrophages, thereby contributing to the arterial remodeling as observed in neointima formation and restenosis. However, until now in vivo intervention studies using inhibitors or genetically modified mice to define the exact role of these proteases in neointima formation have been lacking, nor had expression patterns of cathepsin cysteine proteases been studied in human neointima formation.
|
Cathepsins in aneurysm formation
|
|---|
By their contribution to ECM remodeling and inflammation, cathepsins may play an important role in aneurysm formation. Aortic abdominal aneurysm (AAA) formation is characterized by ECM degradation and chronic inflammation of the aortic wall, accounting for high morbidity and mortality rates (67
, 68)
. Risk factors include male sex, history of atherosclerotic vascular disease, lower HDL values, and higher LDL levels (69)
. Compared with normal arteries, human aortic aneurysms show increased protein levels of cathepsins K and S, whereas cystatin C protein levels were decreased (61)
(see Table 3
). These data imply that the increase in cathepsin activity exceeds the increase in cystatin activity. Cathepsins B, C, and L also showed increased activity in the aneurysm wall and thrombus of human aortic aneurysms when compared with normal arteries (70
71
72)
. Other in vitro studies showed that unstimulated vascular SMCs do not express cathepsin S and show hardly any elastolytic activity, whereas SMCs stimulated with IFN- secrete active cathepsin S and show elastolytic activity. This elastolytic activity could be inhibited by cystatin C (44)
. TGF-ß1, a known inducer of protease inhibitors, was demonstrated to stimulate SMC cystatin C secretion and subsequently block SMC elastase activity in vitro (61)
. Protein levels of cathepsin L were also increased in human AAA and atheroma (47)
. In vitro stimulation with the proinflammatory cytokines interleukin 1ß, interferon-, and tumor necrosis factor (IL-1ß, IFN-, TNF-) and the growth factors basic fibroblast growth factor and vascular endothelial growth factor (bFGF, VEGF) increased cathepsin L mRNA and protein expression in SMCs (induced by bFGF, IFN-, TNF-), endothelial cells (ECs) (similar induction as SMCs), and macrophages (induced by IFN-). Stimulation of both SMCs and ECs with bFGF increased their elastolytic and collagenolytic (type I) activity. The same was true for macrophages stimulated with IFN- (47)
. Furthermore, Sukhova et al. examined the effect of cystatin C deficiency in an atherosclerotic apoE-deficient mouse model on aneurysm formation (63)
. They found an increase in both aortic circumference and length when cystatin C was absent.
The increased expression patterns of several cathepsins, including cathepsins K, L, and S, suggest involvement of these proteins in aneurysm formation. Cathepsin activity might be enhanced by inflammation, since cathepsin L and S protease activity increases after stimulation with inflammatory cytokines. Until now, in vivo intervention studies using inhibitors or genetically modified mice to define the role of cysteine protease cathepsins in aneurysm formation have been lacking.
|
Cathepsins in neovascularization
|
|---|
Neovascularization is defined as the formation of new blood vessels, mediated by progenitor and/or ECs. Subsequently, tube formation will take place, eventually resulting in a stabilized new blood vessel. Cathepsin B and S have been described to contribute to neovascularization by stimulating the formation of capillary-like tubular structures (73)
or stimulating angiogenic islet formation and cell proliferation (74)
. The effect of these proteases on vessel formation has been studied extensively in tumors, but not in cardiovascular diseases.
ECs stimulated with bFGF developed capillary-like tubule structures, while cystatin C and a selective cathepsin S inhibitor (LHVS) reduced this tube formation by 80% and 50%, respectively (75)
. Although cathepsin S deficiency did not affect proliferation or adhesion of mouse ECs, cathepsin S-deficient ECs showed reduced elastolytic and type IV collagenolytic activity and a reduced capacity to invade matrigel or collagen type I gel membranes. Wounded skin shows increased expression of cathepsin S in macrophages, T-lymphocytes, and microvascular ECs. At the site of wound healing, cathepsin S-deficient mice showed an 80% reduction in microvessels despite normal levels of the angiogenic factors bFGF and VEGF (75)
. These data suggest that cathepsin S plays an important role in ECM degradation, thereby facilitating microvessel formation in physiological neovascularization (wound healing).
Attraction of endothelial progenitor cells (EPCs) to the ischemic tissue is also thought to play a role in neovascularization (76
, 77)
. Cathepsins H, L, and O were recently found to be highly expressed in EPCs compared with mature ECs (78)
. Cathepsin L activity was higher in EPCs than in mature ECs (human umbilical venous endothelial cells or HUVECs). Gelatin and collagen (type IV) degradation activity assays in EPC extracts and culture supernatants suggested that cathepsin L was required for ECM degradation, and thus for the invasive capacity of EPCs in vitro. Mice receiving EPCs pretreated with a cathepsin L inhibitor showed impaired recovery after limb ischemia, decreased capillary density, and decreased incorporation of EPCs into vascular structures. The same was evident when mice received cathepsin L-deficient bone marrow-derived cells, indicating that cathepsin L plays a crucial role in integrating EPCs into ischemic tissue. Irradiation of mice resulted in the abolishment of cathepsin L activity, indicating that cathepsin L activity was mediated by irradiation-sensitive cells that probably originated from the bone marrow. Finally, infusion of cathepsin L-transfected mature ECs (HUVECs) increased the recovery of limb perfusion after ischemia (78)
. In addition, cathepsin G, although not a cysteine protease, may play a role in neovascularization. EPCs may secrete cathepsin G to eventually promote early CXCR2-dependent EPC arrest on denuded SMCs or adherent platelets (79)
.
Although the role of neovascularization in atherosclerosis is not completely unraveled, neovascularization is particularly prominent in complicated lesions. Furthermore, neovascularization has recently been identified as a marker of plaque vulnerability (77
, 80
81
82)
. Together, these data indicate that cathepsins are involved in neovascularization, but data defining the role of cathepsin cysteine proteases in neovascularization related to atherosclerosis are lacking.
|
Cathepsins in relation to MMPs
|
|---|
In literature there is little evidence about the differential role of MMPs and cathepsins in cardiovascular disease. Reddy et al. showed that human monocyte-derived macrophages synthesize both elastinolytic MMPs (MMP-7 and -9) and cathepsins B, L, and S. However, only the cathepsins were detected extracellularly as processed, active enzymes. Inhibition of cathepsins L and S, but not MMPs, almost completely blocked macrophage-mediated elastinolytic activity, indicating that cathepsins are the most potent elastases secreted by human macrophages (42)
. The relative contribution of MMPs and cathepsins in bone remodeling is studied more extensively. Cathepsin K-deficient mice showed an osteopetrotic phenotype with excessive trabeculation of the bone marrow space (21)
, indicating that cathepsin K plays a major role in osteoclastic bone resorption. However, cathepsin K-deficient mice continued to grow, suggesting that other proteases, such as MMPs, compensated for the loss of cathepsin K activity (83)
. It was suggested that osteoclasts lower the pH in the resorption area, after which the bone matrix was first digested by cysteine proteases. Digestion by MMPs followed at higher pH levels (84)
. Site specificity existed in relation to digestion of the bone matrix by osteoclasts, since osteoclastic resorption of calvarial bone (intramembranous bone) was dependent on both cysteine proteases and MMP activity, whereas long bone (endochondral bone) resorption was only dependent on cysteine protease activity (85)
. Others suggested that resorption of scapular bone (intramembranous bone) was more dependent on MMPs than cysteine proteases whereas resorption of long bone (endochondral bone) was more dependent on cysteine proteases than MMPs (86)
. Recently, it was shown that calvarial osteoclasts use other cysteine proteases in addition to cathepsin K and that long bone osteoclasts use MMPs in the absence of cathepsin K (87)
. Deficiency of other cathepsins, including cathepsins B and L, do not show osteoclast-related effects in bones. In the absence of cathepsin L activity, osteoclasts do not use MMPs for the resorption of calvarial bone matrix, suggesting that cathepsin L plays a role in osteoclast-mediated bone matrix resorption by activating MMPs (87)
.
Both MMPs (MMP-1, -2, -3, -7, -9, -11, -12, and -13) and TIMPs (TIMP-1 and -2) have been extensively studied in atherosclerosis using overexpression and knockout studies (1
2
3)
. For example, Johnson et al. studied atherosclerotic plaque stability in apoE-deficient mice lacking MMP-3, -7, -9, or -12 and found that members of the MMP family had differential effects on atherogenesis, some having a protective role (MMP-3 and -9) and others having a stimulating role (MMP-12), while deficiency of MMP-7 showed no apparent effect on atherosclerotic plaque stability (MMP-7) (88)
. The same group showed that adenoviral infection of TIMP-2, but not TIMP-1, resulted in the reduction of lesion area and macrophage content and in an increase in SMC and elastin content (89)
. For the cathepsin cysteine protease family, only cathepsins K and S knockout studies have been described. While recent studies in bone remodeling indicate that deficiency of cathepsin K is compensated for by MMP activity and that cathepsins and MMPs show site specificity, more in vivo studies are needed to compare the relative contribution and the differential role of MMPs and cathepsins in cardiovascular diseases.
|
Cathepsins and cystatins as a diagnostic tool in vascular pathology
|
|---|
Several studies have evaluated the use of serum cathepsin levels as a diagnostic tool for AAAs. Using a DNA expression array, cathepsin H (30-fold up-regulated) was found to be up-regulated in human AAAs compared with normal aorta (90)
. Further evaluation of a role for cathepsin H in the etiology of AAA, using genetic linkage analysis, did not confirm a linkage for cathepsin H with AAA (91)
. A comparison of serum levels of patients with AAA vs. normal aorta revealed decreased levels of serum cystatin C (61)
, but no difference in serum cathepsins B and L levels was found (92)
. These data suggest that measurement of serum cystatin C levels may be helpful in the diagnosis of AAA, but this requires further exploration.
Recently, the feasibility of the use of cathepsins as a tool in the diagnostic imaging of atherosclerosis was explored. In vivo imaging of cathepsin B using a cathepsin B imaging probe showed colocalization with cathepsin B immunoreactivity in atherosclerotic plaques (45)
. Several other studies have suggested that cathepsins might be a useful diagnostic tool for atherosclerosis. Cathepsin L serum levels were found to be increased in patients with >10% stenosis in at least one of the coronary arteries when compared with patients without stenosis. This increase persisted after adjustment for the confounders sex, smoking, age, and glucose levels (47)
. Serum cathepsin S levels were also increased in patients with atherosclerotic stenosis in at least one of the coronary arteries compared with patients without stenosis (93)
TOP
ABSTRACT
INTRODUCTION