Destabilizing Role of Cathepsin S in Murine Atherosclerotic Plaques
Objective— Lysosomal proteinases have been implicated in a number of pathologies associated with extracellular matrix breakdown. Therefore, we investigated the possibility that the lysosomal proteinase cathepsin S may be involved in atherosclerotic plaque destabilization.
Methods and Results— Atherosclerotic plaques in the brachiocephalic arteries of fat-fed apolipoprotein E/cathepsin S double knockout mice had 73% fewer acute plaque ruptures (P=0.026) and were 46% smaller (P=0.025) than those in age-, strain-, and sex-matched apolipoprotein E single knockout controls. When the incidence of acute plaque rupture was normalized for plaque size, the reduction in the double knockouts was 72% (P=0.039). The number of buried fibrous layers, indicative of an unstable plaque phenotype, was reduced by 67% in the double knockouts (P=0.008). The cysteine proteinase inhibitor, egg white cystatin, was biotinylated and used as an active-site-directed probe for cathepsins. Biotinylated cystatin selectively detected cathepsin S in extracts of human carotid atherosclerotic plaque. Active cathepsin S was detectable in extracts of human atherosclerotic plaque but not in nondiseased carotid arteries. Active cathepsins were especially prominent in macrophages in the shoulder regions of plaques, areas considered to be vulnerable to rupture. Cathepsin S protein colocalized with regions of elastin degradation in human coronary plaques.
Conclusion— These data provide direct evidence that an endogenous proteinase, cathepsin S, plays an important role in atherosclerotic plaque destabilization and rupture.
The highly thrombogenic gruel in the core of an atherosclerotic plaque is luminally covered by a fibrous cap, weakening of which leads to plaque rupture and thrombus formation. Macrophages and T cells accumulate at sites of plaque disruption.1 Degradation of collagen by macrophage-derived matrix metalloproteinases has been reported,2 but less is known of other classes of proteinase, which may be released by activated macrophages.
The lysosomal cathepsins have been implicated in the development and progression of atherosclerosis. Increased levels of cathepsins F, K, and S are present in atherosclerotic lesions,3,4 whereas levels of the major extracellular inhibitor of cysteine proteinases, cystatin C, are decreased.5 In humans, an association has been shown between a genetically determined decrease in cystatin C levels and the severity of coronary artery disease.6
Cathepsin S/low-density lipoprotein (LDL) receptor double knockout mice have impaired atherogenesis when compared with LDL receptor single knockout controls.7 Grading of atherosclerosis in the aortic arch showed that there was a delay in lesion progression in the double knockouts. For example, 12 weeks of feeding atherogenic diet to the double knockouts resulted in an average lesion severity similar to that seen in single knockout controls after just 8 weeks of feeding, and 26 weeks of atherogenic diet feeding in the double knockouts produced lesions similar to those observed after 12 weeks in the single knockouts. This suggests that cathepsin S is involved in the rate of progression of lesions.
There is also evidence supporting a role for cathepsins in plaque destabilization. Human monocyte-derived macrophages secrete the proactive and active forms of cathepsins B, L, and S in vitro.8 Cathepsins S and K have similar elastinolytic potency to neutrophil elastase at pH 7.5,9 and cathepsin K can cleave the triple helix of collagens I and III, so cathepsins can attack the structural proteins important for preservation of the integrity of the fibrous cap. Cathepsins have been detected extracellularly in plaques,10 further implicating them in fibrous cap degradation and plaque destabilization.
The development of an animal model of spontaneous plaque destabilization,11,12 and in particular the development of a model in which spontaneous plaque destabilization can be monitored in the short term,13 has made it possible to conduct intervention studies in which plaque stability is measured directly. In this model, plaques develop in the brachiocephalic artery of fat-fed apolipoprotein E (apoE) knockout mice. We tested the hypothesis that cathepsin S is involved in plaque destabilization by cross-breeding these mice with cathepsin S knockouts.14
For materials, please see the online supplement, available at http://atvb.ahajournals.org.
Homozygous C57BL/6–129 apoE knockout mice and homozygous cathepsin S knockout mice on a pure DBA strain background14 were used. The initial pairings generated apoE+/− cathepsin S+/− double heterozygotes on a mixed C57BL/6–129–DBA strain background. After interbreeding of the double heterozygotes, male offspring were selected, and polymerase chain reaction analysis of tail-tip DNA was used to identify apoE−/− cathepsin S−/− double knockouts and their apoE−/− cathepsin S+/+ single knockout controls, which were thus age and sex matched and on the C57BL/6–129–DBA mixed strain background. This approach resulted in random distribution of strain-related allelic variation between groups. The housing and care of the animals and all the procedures used in these studies were performed in accordance with the regulations of the University of Bristol and the United Kingdom home office.
At 6 to 8 weeks of age, male apoE−/− cathepsin S−/− double knockouts and apoE−/− cathepsin S+/+ controls were fed high-fat rodent diet containing 21% pork lard and supplemented with 0.15% cholesterol (Special Diets Services) for 12 weeks. At termination, animals were exsanguinated and pressure perfusion fixed as described previously.12
For termination, please see the online supplement.
For histology, please see the online supplement.
Sections stained for elastin were inspected for disruption of the cap of the plaque. Acute plaque disruption was accepted when a visible defect in the cap was accompanied by intrusion of erythrocytes into the region below it. In many animals, ≥1 smooth muscle cell– and elastin-rich layers, usually overlain with foam cells, were seen within the body of the plaque and were also counted.
For morphometric analysis, please see the online supplement.
Lipid, Elastin, Collagen, and Macrophage Content
For lipid, elastin, collagen, and macrophage content, please see the online supplement.
Fibrous Cap Thickness
For fibrous cap thickness, please see the online supplement.
Human carotid endarterectomy specimens (n=20) were obtained within 1 hour of carotid endarterectomy surgery. Frozen sections of the specimens were examined qualitatively for the presence and extent of calcification, the size of the lipid core, and the thickness of the fibrous cap. Macroscopically normal (n=5) and diseased human coronary arteries (n=5) were obtained postmortem. The studies were approved by the institutional ethical review board, and subjects or their next-of-kin gave informed consent.
Preparation of Biotinylated Cystatin
For preparation of biotinylated cystatin (bCystatin), please see the online supplement. Egg white cystatin was biotinylated with d-biotinyl-ε-aminocaproic acid N-hydroxysuccinimide ester. bCystatin was purified by column fractionation with monitoring of cathepsin S inhibitory activity using Z-Val-Val-Arg-7-amino-4-methylcoumarin (AMC) substrate.15
Some sections were developed by incubation with a goat polyclonal antibody to human cathepsin S or an isotype-matched nonimmune IgG control (1 μg/mL) overnight at 4°C. The secondary antibody was biotinylated rabbit anti-goat IgG (1:400; 30 minutes), developed using the Vectastain ABC kit and DAB. The specificity of the anti–cathepsin S antibody was assessed by Western blotting using extracts of the J774 mouse macrophage cell line showing that only cathepsin S was recognized, and that there was no reaction with other proteins including cathepsins B, H, and L that are known to be expressed by J774 cells.16 For full details, please see the online supplement.
For human immunohistochemistry, please see the online supplement. Cryosections of carotid endarterectomy plaque were incubated with bCystatin (60 μg/mL) for 60 minutes followed by a 30-minute incubation with Extravidin–horseradish peroxidase (HRP). Peroxidase activity was visualized using DAB. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide.
Sections developed with an antibody to human CD68 were blocked with goat serum and incubated with mouse anti-human CD68 (4 μg/mL) for 60 minutes. The secondary antibody was biotinylated goat anti-mouse IgG (1:1000), developed using Extravidin–HRP.
Cryosections of coronary plaque were blocked with rabbit serum in PBS containing pontamine sky blue 6BX (0.5%) for 30 minutes and incubated with goat anti-human cathepsin S (1 μg/mL). The secondary antibody was biotinylated rabbit anti-goat IgG (1:400), developed using Extravidin–fluorescein isothiocyanate. Sections were counterstained with propidium iodide.
Values are expressed as mean±SEM. Data were analyzed using the programs InStat and Prism (GraphPad Software). A check was first made for similar variances and normality of distribution. If these were both passed, then an unpaired 2-sample 2-tailed Student t test was used. If the distributions were not normal, then a nonparametric Mann–Whitney test was used. If the normality test was passed but the variances differed significantly, then Welch corrected t test was used. Contingency data (presence or absence of acute plaque rupture) were analyzed by Fisher exact test. Discontinuous data (number of buried fibrous layers) were analyzed using the Mann–Whitney test. Significance was concluded when P<0.05.
Acute Plaque Rupture Is Inhibited in ApoE/Cathepsin S Double Knockout Mice
The proportion of mice with acute plaque ruptures was decreased by 73% in apoE/cathepsin S double knockouts (P=0.026; Table 1).
ApoE/Cathepsin S Double Knockout Mice Have a More Stable Plaque Phenotype
The number of buried fibrous layers, a marker of an unstable plaque phenotype, was reduced by 67% from 1.33±0.29 buried layers/mouse in apoE single knockout mice to 0.44±0.10 buried layers/mouse in double knockouts (P=0.008; Table 1).
The double knockouts had fibrous caps that were 70% thicker: 7.2±0.8 μm, compared with 4.2±0.7 μm in single knockout controls (P=0.020; Table 2). The analysis was complicated by the fact that the plaques in many of the double knockouts had not developed sufficiently to form a fibrous cap. All 9 of the apoE single knockout controls but only 12 of the 27 double knockouts had discernible fibrous caps (P=0.005), lending weight to the suggestion that cathepsin S fosters plaque development as well as destabilization.
Atherosclerotic Plaque Size Is Decreased in ApoE/Cathepsin S Double Knockout Mice
Plaque area in apoE single knockouts was 101.5±16.6 ×103 μm2, but this was reduced by 46% to 55.1±9.9 ×103 μm2 in double knockouts (P=0.025; Table 2). There were no significant differences in media or lumen area. Representative plaques are shown in Figure 1A and 1B. In apoE single knockout mice, cathepsin S immunopositivity was seen in the lipid-rich regions of plaques (Figure 1E).
The Increase in Plaque Stability in ApoE/Cathepsin S Double Knockout Mice Is Independent of the Reduction in Plaque Size
When the number of acute plaque ruptures was normalized by dividing by the plaque area, there was still a significant reduction, from 9.4±4.0 mm2 in single knockouts to 2.7±1.4 mm2 in double knockouts (−72%, P=0.039; Table 1). These data suggest that cathepsin S is a mediator of plaque destabilization.
Plaque Composition Is Altered in ApoE/Cathepsin S Double Knockout Mice
ApoE/cathepsin S double knockouts had 49% less plaque elastin than single knockout controls (controls 10.5±1.1%; double knockouts 5.3±1.3%; P=0.005). There were no significant differences in medial elastin, plaque collagen, plaque lipid, or macrophage content. There was no difference in the number of breaks in the medial elastic laminae (for full details, see Table II, available online at http://atvb.ahajournals.org).
bCystatin Selectively Detects Cathepsin S in Tissue Extracts
For full details, see the online supplement. On Western blots using bCystatin for detection, 1 major band was present in all of the plaque extracts (n=15, 6 of which are shown in Figure 2). No corresponding band was present in nondiseased carotids (n=5, 2 of which are shown in Figure 2). Using a polyclonal antibody, it was demonstrated that the major band corresponded to cathepsin S (Figure 2).
Cysteine Proteinase Activity Is Present in Human Atherosclerotic Plaques
Significant increases in bCystatin binding (P=0.002; Figure 3A) and the rate of hydrolysis of a cathepsin S substrate (P=0.002; Figure 3B) were found in extracts of carotid plaque when compared with extracts of control arterial tissue. Although there was appreciable variability in cysteine proteinase activity between specimens, this did not appear to be correlated with the stage of development of the plaque. All of the endarterectomy plaques examined were similar in terms of calcification, lipid content, and fibrous cap thickness. None showed characteristics associated with plaque instability or rupture.
Cysteine Proteinase Activity Localizes to Macrophages in Human Atherosclerotic Plaques
Cryosections of human atherosclerotic plaque tissue (n=20) were examined for enzymically active cysteine proteinase molecules using the bCystatin probe. Isolated groups of cells were found to be strongly positive, mainly in the shoulder regions of the plaque and at the edges of the lipid core (Figure 4A and 4B). When serial sections were examined by immunohistochemistry, these cells were all shown to be lipid-rich macrophages (Figure 4C and 4D). Incubation of tissue sections with cysteine proteinase inhibitors before incubation with bCystatin completely abolished the reactivity to bCystatin. Some extracellular cathepsin S immunopositivity was also seen; in coronary arteries, this appeared to be closely focused at areas of elastin fragmentation (Figure 4E). Adjacent regions of the same section with normal elastin were negative for cathepsin S (Figure 4F).
Control apoE single knockout mice experienced an average of 0.56 acute plaque ruptures in the proximal part of the brachiocephalic artery (Table 1). This was reduced by 73% in animals that were additionally cathepsin S null (P=0.026), supporting the idea that this cysteine proteinase plays a role in atherosclerotic plaque destabilization. There was also a reduction in plaque size (−46%; P=0.025). In a previous study,7 cathepsin S/LDL receptor double knockout mice also had smaller plaques, but plaque stability was not directly examined. The current study is therefore the first to show directly that cathepsin S is involved in atherosclerotic plaque destabilization and rupture.
Is the reduction in the number of plaque ruptures just a reflection of impaired plaque progression? Because the plaques in the apoE/cathepsin S double knockouts were significantly smaller, they may simply have failed to develop to the stage at which instability could manifest itself. To address this, we normalized for plaque size by dividing the number of acute plaque ruptures by the plaque area. The normalized incidence of acute rupture was reduced by 71% in double knockouts (P=0.039), showing that the reduction in plaque ruptures reflects a real change in plaque stability.
Plaques in the double knockout mice contained significantly less elastin, which is surprising given that cathepsin S is elastinolytic. However, it is consistent with a recent report that apoE knockout mice transplanted with cystatin C null bone marrow show an increase in plaque elastin.17 It is possible that cathepsin S is involved in the normal turnover of elastin during atherogenesis, and that intimal elastin deposition is compromised in its absence. In contrast, normal elastogenesis in the media was unaffected, and there was no change in the number of breaks in the medial elastic laminae. This is consistent with the lack of cathepsin S immunopositivity in the media (Figure 1E).
In contrast to the report by Sukhova et al7 of a reduction in aortic arch plaque macrophage content in LDL receptor/cathepsin S double knockout mice, we saw no similar change in the brachiocephalic arteries of apoE/cathepsin S double knockouts. The macrophage content was reduced from &30% to &22% in the study of Sukhova et al7 but was only 19% in our single knockout control brachiocephalic arteries at a similar time point. At present, we are unable to explain the difference in macrophage accumulation between the 2 sites or why this should be cathepsin S sensitive in one but not the other.
The use of biotinylated inhibitors for the detection of active proteinases has so far been restricted to nitrocellulose membrane-bound proteinases after their separation by SDS/PAGE or isoelectric focusing.18,19 In the present study, methods have been developed for detecting active cysteine proteinases in cryosections of human tissue. This approach should be adaptable to proteinases of other classes provided that suitable inhibitors are available.
Because of similarities in their apparent molecular weights, it is difficult to separate cathepsins by 1D electrophoresis. Inactivation of other cathepsins by heating before electrophoresis allowed the selective detection of cathepsin S by bCystatin. If heat treatment was omitted, 3 bands were detected, the additional bands corresponding to cathepsins L and B (data not shown).
The lysosomal cysteine proteinases are optimally active at acidic pH. With the exception of cathepsin S, they are irreversibly inactivated after prolonged exposure to neutral pH or heating at neutral pH.20 The pH stability of cathepsin S has been used to increase the specificity of activity assays using peptide substrates. We took advantage of the pH stability of cathepsin S to allow its selective detection on affinity blots using bCystatin and developed a bCystatin binding assay to measure cathepsin S activity in tissue extracts. This correlated well with activity as assessed using a peptide substrate (see the online supplement). Increased cathepsin S activity was demonstrated in extracts of carotid plaques with both of these assays (Figure 3).
Because analysis of tissue extracts can only indicate gross changes in proteinase and proteinase inhibitor levels, we further used the active site-directed bCystatin probe to localize regions of active proteinase activity in tissue cryosections. Positive staining was located mainly between the fibrous cap and the lipid core and in the shoulder region of the plaque, areas vulnerable to rupture (Figure 4A and 4B). The positive cells were macrophage-derived foam cells (Figure 4D and 4E), supporting a link between the accumulation of modified lipids and proteins by macrophages and increased cathepsin activity. No positive staining with bCystatin was found in the sections from undiseased control carotid arteries. These data suggest that there is an increase in cysteine proteinase activity in human advanced carotid plaques, localized to specific regions rich in macrophage-derived foam cells. Cathepsin S protein also localized to these areas (Figure 1E). It has been reported previously that aortic sinus plaques in fat-fed apoE knockouts are primarily immunopositive for cathepsin S in the smooth muscle cell-rich fibrous cap and media.21 However, Sukhova et al found that in human plaques, cathepsin S protein was present in both foam cells (presumed to be macrophage derived) and smooth muscle cells, and that in more advanced plaques, cathepsin S immunopositivity in smooth muscle cells decreased.3 We reported previously that aortic sinus lesions in fat-fed apoE knockout mice do not rupture,13 so the lack of cathepsin S in plaque smooth muscle cells from our apoE knockout mice may reflect their more advanced phenotype.
In summary, both plaque size and the incidence of acute plaque rupture in the brachiocephalic arteries of fat-fed apoE/cathepsin S double knockout mice are reduced significantly. The reduction in the incidence of acute plaque rupture was independent of the effect on plaque size. This is the first direct demonstration that cathepsin S is involved in plaque rupture. A specific active site-directed probe was used to show that human carotid artery plaques contain enzymically active cysteine proteinases that are clustered in the shoulder regions and around the lipid core, localizing to lipid-rich macrophages. These data support the idea that cathepsin S contributes to tissue destruction in the plaque and ultimately to atherosclerotic plaque rupture.
This work was supported by Pfizer (K.J.R.), Organon Laboratories Limited (D.J.W.), and the British Heart Foundation (A.L.M.). We gratefully acknowledge the assistance of the staff of the Vascular Studies Unit, Department of Surgery, Bristol Royal Infirmary, in obtaining carotid endarterectomy specimens.
- Received July 5, 2005.
- Accepted December 28, 2005.
Van der Wal AC, Becker AE, Van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994; 89: 36–44.
Oorni K, Sneck M, Bromme D, Pentikainen MO, Lindstedt KA, Mayranpaa M, Aitio H, Kovanen PT. Cysteine protease cathepsin F is expressed in human atherosclerotic lesions, is secreted by cultured macrophages, and modifies low density lipoprotein particles in vitro. J Biol Chem. 2004; 279: 34776–34784.
Eriksson P, Deguchi H, Samnegard A, Lundman P, Boquist S, Tornvall P, Ericsson CG, Bergstrand L, Hansson LO, Ye S, Hamsten A. Human evidence that the cystatin C gene is implicated in focal progression of coronary artery disease. Arterioscler Thromb Vasc Biol. 2004; 24: 551–557.
Reddy VY, Zhang Q-Y, Weiss SJ. Pericellular mobilization of the tissue-destructive cysteine proteinases, cathepsins B, L, and S, by human monocyte-derived macrophages. Proc Nat Acad Sci U S A. 1995; 92: 3849–3853.
Hakala JK, Oksjoki R, Laine P, Du H, Grabowski GA, Kovanen PT, Pentikainen MO. Lysosomal enzymes are released from cultured human macrophages, hydrolyze LDL in vitro, and are present extracellularly in human atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2003; 23: 1430–1436.
Williams H, Johnson JL, Carson KG, Jackson CL. Characteristics of intact and ruptured atherosclerotic plaques in brachiocephalic arteries of apolipoprotein E knockout mice. Arterioscler Thromb Vasc Biol. 2002; 22: 788–792.
Johnson JL, Carson KGS, Williams H, Karanam S, Newby AC, Angelini GD, George SJ, Jackson CL. Plaque rupture after short periods of fat-feeding in the apolipoprotein E knockout mouse: model characterization, and effects of pravastatin treatment. Circulation. 2005; 111: 1422–1430.
Nakagawa TY, Brissette WH, Lira PD, Griffiths RJ, Petrushova N, Stock J, McNeish JD, Eastman SE, Howard ED, Clarke SRM, Rosloniec EF, Elliott EA, Rudensky AY. Impaired invariant chain degradation and antigen presentation and diminished collagen-induced arthritis in cathepsin S null mice. Immunity. 1999; 10: 207–217.
Claus V, Jahraus A, Tjelle T, Berg T, Kirschke H, Faulstich H, Griffiths G. Lysosomal enzyme trafficking between phagosomes, endosomes, and lysosomes in J774 macrophages: enrichment of cathepsin H in early endosomes. J Biol Chem. 1998; 273: 9842–9851.
Rodgers KJ, Melrose J, Ghosh P. Biotin-labeled potato chymotrypsin inhibitor-1: a useful probe for the detection and quantitation of chymotrypsin-like serine proteinases on Western blots and its application in the detection of a serine proteinase synthesised by articular chondrocytes. Anal Biochem. 1995; 227: 129–134.
Maciewicz RA, Etherington DJ. A comparison of four cathepsins (B, L, N and S) with collagenolytic activity from rabbit spleen. Biochem J. 1988; 256: 433–440.