Author + information
- Received April 10, 2007
- Revision received August 23, 2007
- Accepted August 24, 2007
- Published online December 11, 2007.
- Masayoshi Sarai, MD⁎,1,
- Dagmar Hartung, MD⁎,†,1,
- Artiom Petrov, PhD⁎,⁎ (, )
- Jun Zhou, MD⁎,
- Navneet Narula, MD⁎,
- Leo Hofstra, MD, PhD‡,
- Frank Kolodgie, PhD§,
- Satoshi Isobe, MD⁎,
- Shinichiro Fujimoto, MD⁎,
- Jean-Luc Vanderheyden, PhD∥,
- Renu Virmani, MD, FACC§,
- Chris Reutelingsperger, PhD‡,
- Nathan D. Wong, PhD⁎,
- Sudhir Gupta, MD, PhD⁎ and
- Jagat Narula, MD, PhD, FACC⁎
- ↵⁎Reprint requests and correspondence:
Dr. Artiom Petrov, University of California, Irvine, School of Medicine, C116 Medical Science I, Irvine, California 92697.
Objectives The purpose of this study was to evaluate the role of caspase inhibitors on acute resolution of apoptosis in atherosclerotic lesions as evaluated by imaging with annexin A5.
Background Extensive apoptosis of macrophages has been reported at the site of plaque rupture in patients dying of acute coronary syndrome.
Methods Of 31 New Zealand White atherosclerotic rabbits, 6 received broad caspase, 3 received caspase-1, 3 received caspase-3, 3 received caspase-8, and 4 received caspase-9 inhibitors; 12 animals did not receive any caspase inhibitors (treatment control group). Six unmanipulated rabbits were used for comparison (disease control group). Technetium-99m–labeled annexin A5 was used for imaging atherosclerotic lesions; 6 of the 12 uninhibited atherosclerotic rabbits received 99mTc-labeled mutant annexin A5 (radiotracer control group). Gamma images were obtained, and quantitative radiotracer uptake was compared with pathologic findings.
Results Atherosclerotic lesions were best visible in untreated atherosclerotic rabbits. Quantitative annexin uptake, defined as the percent of injected dose per g of abdominal aorta tissue, was significantly higher in untreated atherosclerotic animals (mean ± SD = 0.0515 ± 0.0099) compared with the normal rabbits (0.0065 ± 0.0008; p < 0.0001) or atherosclerotic rabbits receiving mutant annexin (0.014 ± 0.0024; p < 0.0001). Among all caspase inhibitor-treated rabbits, uptake was 39% lower (0.0314 ± 0.0151) than in untreated atherosclerotic animals (p < 0.01). Uptake was also significantly lower in rabbits receiving broad caspase (0.0206 ± 0.0058; p < 0.0001) or caspase-1, -3, or -9 (0.0272 ± 0.0088, p < 0.01; 0.0286 ± 0.0095, p < 0.01; 0.0300 ± 0.0021, p < 0.01, respectively) inhibitors. Caspase-8 inhibitor did not affect apoptosis (0.0618 ± 0.0047; p = NS). Upon histologic characterization, a substantial decrease in macrophage apoptosis was observed in caspase-inhibited animals.
Conclusions Molecular imaging, using radiolabeled annexin A5, allows the detection of acute resolution of apoptosis as a result of caspase inhibition in experimental atherosclerosis. If proven clinically, this may allow development of novel intervention strategies in acute vascular events.
Autopsy studies have revealed a high prevalence of apoptosis in vulnerable and ruptured atherosclerotic plaques, which is associated with activation of caspase-1 and -3 (1,2). Apoptosis of smooth muscle cells leads to attenuation of the fibrous cap (3,4), and apoptosis of foam cells results in enlargement of the necrotic core (5); both thin caps and large cores are important determinants of plaque vulnerability. In addition, extensive apoptosis of foam cells is seen at the site of fibrous cap rupture (1,2). Therapeutic interventions known to substantially reduce acute coronary events, such as dietary modification and statin treatment, reduce the prevalence of apoptosis in atherosclerotic lesions (6,7). However, a recent clinical trial of high-dose statin treatment, immediately following acute coronary syndrome, demonstrated that the benefit of statins may be observed only after a lapse of 4 to 5 weeks, a period when recurrent events are most frequent (8). Therefore, it is logical to assume that if apoptosis contributes to plaque rupture, acute repression of apoptosis in acute coronary events should help prevent early recurrent coronary events. It should be possible to inhibit apoptosis acutely by pharmacologic interventions targeted at the signaling cascade of apoptosis.
There are 2 major signaling pathways of apoptosis (Fig. 1):the death receptor pathway (9–11) and the mitochondrial pathway (12,13). In the death receptor pathway, the signal is provided by the interaction between the ligand and death receptor, recruitment of adapter proteins, and activation of proximal (caspase-8) and executioner caspases. In the mitochondrial signaling pathway, a number of molecules (including cytochrome c) are released from the mitochondrial intermembrane space into the cytoplasm. There they interact with adapter proteins and activate a distinct initiator caspase (caspase-9) that then activates common executioner caspases, resulting in apoptosis. Caspase-3 is a prototype of the executioner caspases, and it cleaves a number of cytoplasmic and nuclear substrates, which results in the morphologic and biochemical characteristics of apoptosis (11). Activation of proximal caspases distinguishes between death receptor (caspase-8) and mitochondrial (caspase-9) pathways of apoptosis.
Caspase-3 activation is closely associated with phosphatidylserine (PS) expression on the apoptotic cell surface (14). Annexin A5, a naturally occurring protein, binds to PS with high affinity and has been used as an early marker of apoptosis. Radiolabeled annexin A5 has been successfully employed for the noninvasive detection of macrophage apoptosis in experimental atherosclerotic lesions in rabbit (15) and mouse aorta (16), porcine coronary arteries (17), and carotid arteries in patients who had suffered transient ischemic episodes (18).
The present study was performed to evaluate if acute repression of apoptosis was possible by the using caspase inhibitors, as determined by noninvasive imaging with technetium-99m (99mTc)–labeled annexin A5. For inhibition of apoptosis, we employed intravenously infused broad caspase inhibitors, executioner caspase-3–specific inhibitor, and inhibitors of proximal caspases-8 and -9 to distinguish between death receptor and mitochondrial pathways of apoptosis. Inhibitor of caspase-1 was used because activated caspase-1 has been observed in vulnerable plaques and ruptured lesions in patients dying of acute coronary events (2).
The experimental protocol was approved by the Institutional Animal Research and Care Committee at the University of California-Irvine. Thirty-seven male New Zealand White rabbits were obtained from Western Oregon Farm (Philomath, Oregon). After a 1-week quarantine, 6 unmanipulated animals were set aside as disease controls, and they received normal rabbit chow for 16 weeks (Fig. 2).These animals were imaged with 99mTc-annexin A5 16 weeks later.
Atherosclerotic lesions were induced in the remaining 31 rabbits. These animals were fed a 0.5% cholesterol diet custom-mixed in 6% peanut oil. One week after the start of the high-cholesterol diet, they underwent balloon de-endothelialization of their infradiaphragmatic aortas and continued on a high-cholesterol, high-fat diet for 15 additional weeks (6). Of these, 25 atherosclerotic animals were imaged with 99mTc-annexin A5 at 16 weeks and classified as follows. Six of these 25 atherosclerotic animals were imaged in the absence of any caspase inhibitor (treatment control), and 19 animals received caspase inhibitors. The caspase inhibitors were administered by intravenous push for 6 h (3 mg) and 1 h (4 mg) before radiolabeled annexin A5 injection: broad (nonselective) caspase inhibitor (n = 6), caspase-1 inhibitor (n = 3), caspase-3 inhibitor (n = 3), caspase-8 inhibitor (n = 3), or caspase-9 inhibitor (n = 4). The remaining group of 6 atherosclerotic rabbits was imaged with mutant annexin A5 at week 16 (radiotracer control); mutant annexin A5 has ineffective PS-binding sites (6).
We used a cell-permeable nonselective caspase inhibitor (ZVAD-fmk; BIOMOL International, LP, Plymouth Meeting, Pennsylvania), caspase-1 inhibitor (YVAD-cho; BIOMOL International, LP), caspase-3 inhibitor (DEVD-cho; BIOMOL International, LP), caspase-8 inhibitor (ZIETD-fmk; Calbiochem, San Diego, California), and caspase-9 inhibitor (ZLEHD-fmk; BIOMOL International, LP) for acute suppression of apoptosis. The inhibitors were dissolved in dimethylsulphoxide and diluted with saline for intravenous administration. For radionuclide imaging, human recombinant annexin A5 was obtained from Theseus Imaging Corporation (Boston, Massachusetts), and mutant annexin A5 was from Mosa Medics (Maastricht, the Netherlands).
Noninvasive imaging of atherosclerotic lesions
Radiolabeling of Annexin A5 With 99mTc
Human recombinant annexin A5 was produced by expression in Escherichia coliand was labeled with 99mTc, as described (19). Recombinant annexin A5 has been shown to retain PS-binding activity equivalent to that of native annexin A5. Before radiolabeling, annexin A5 was derivatized with the nicotinic acid analog hydrazinonicotinamide (HYNIC) (AnorMED, Incorporated, Langley, Canada) at a 0.9 mol/mol ratio by gentle mixing. Hydrazinonicotinamide is a bifunctional molecule with an affinity for lysine residues of proteins on 1 moiety and for the conjugates of 99mTc on the other; the stable complex formed by this molecule does not affect protein bioreactivity. For binding 99mTc to the HYNIC-annexin A5 conjugate, a reduced tin (stannous ion) and tricine solution was added to 99mTc pertechnetate with an aliquot of HYNIC-annexin A5 under anoxic conditions. The final specific radioactivity was 10 to 200 μCi/μg protein (1 μCi = 37 kBq). Thin-layer chromatography using the solvent NaCl showed a radiopurity between 92% and 97%.
For molecular imaging, each preparation of annexin A5 (0.1 mg) radiolabeled with 5.5 ± 1.3 mCi of 99mTc was injected into the left marginal ear vein of rabbits. Three hrs later animals were anesthetized with ketamine and xylazine (0.1 mg/ml each, 10:1 vol/vol). Radionuclide imaging was performed using a Vertex PLUS gamma camera (ADAC Laboratories, Milpitas, California) for planar imaging (n = 30) or a dual-head gamma camera (n = 7) with micro-computed tomography (CT) X-SPECT (Gamma Medica, Inc., Northridge, California) for SPECT images. Planar images of the abdomen were obtained for 30 min in a 128 × 28 matrix using a 1-mm pinhole collimator. Single photon emission computed tomography (SPECT) images of the abdomen were acquired in a 64 × 64 scaffold, 32 steps at 120 s/step on a 140-keV photopeak of 99mTc with a 15% window. After SPECT acquisition, a micro-CT image of the rabbit abdomen was acquired without moving the animal position. The micro-CT used an X-ray tube operating at 50 kVp and 0.8 mA, and images were captured for 0.5 s/view for 256 views in 360° rotation. The micro-SPECT images were converted to a 256 × 256 scaffold, and micro-CT studies were fused, allowing the achievement of simultaneous molecular and anatomic information in all tomographic scans in 3 different spatial axes. After in vivo imaging, animals were sacrificed with an overdose of sodium pentobarbital (120 mg/kg). Aortas were carefully dissected, and planar images of ex vivo aortas were obtained for 15 min in a 128 × 128 matrix using a 1-mm pinhole collimator.
Quantitative radiotracer uptake in tissue specimens
After ex vivo imaging, each aorta was segmented at 1-cm intervals. These segments were weighed and counted in an automatic well-type gamma counter (1480 Wizard 3″, Wallac, Turku, Finland) for determination of the percent injected annexin dose per gram (%ID/g) of tissue. Biodistribution in the blood, lung, liver, spleen, kidney, and skeletal muscles was also evaluated. Aortas were preserved for histologic and immunohistochemical investigations, as reported earlier (15).
Histologic assessment of experimental atherosclerosis
The aortic segments were fixed with HEPES-buffered formalin (4%) containing an additional 2 mmol/l Ca2+. Histologic and immunohistochemical characterization was performed in 10 atherosclerotic (4 untreated and 2 broad caspase, 1 caspase-1, 2 caspase-8, and 1 caspase-3 inhibited) animals and 2 normal controls. Each aorta was segmented at 1-cm intervals, and each 1-cm segment was subdivided into 3 equidistant sections and embedded on edge in paraffin. The tissue was then dehydrated in a graded series of ethanol. Serial 4-μm-thick sections were cut and mounted on charged slides (Superfrost; Fisher Scientific, Waltham, Massachusetts). Tissue sections were stained with hematoxylin-eosin and Movat pentachrome elastin stains. Atherosclerotic lesions were characterized using a classification scheme based on the recommendations of the American Heart Association (20).
For the immunohistochemical characterization of the cellular composition, primary antibodies directed against macrophages (using marker RAM-11, dilution 1:200 overnight incubation; Dako, Carpinteria, California) and smooth muscle cells (using a primary antibody against actin α and β isotopes, HHF-35, dilution 1:200 overnight incubation; Enzo Biochem, New York, New York) were employed as described previously (15). Primary antibodies were tracked with a biotinylated link anti-mouse antibody using a peroxidase-based labeled streptavidin biotin kit (Dako) and visualized by an 3-amino-9-ethylcarbazole substrate-chromogen system (Dako). The presence of apoptotic cells was evaluated by histologic detection of DNA fragmentation by terminal deoxyribonucleotide transferase TdT-mediated nick-end labeling (TUNEL) staining using an in situ apoptosis detection kit (TACS, Trevigen, Inc., Gaithersburg, Maryland) as described previously (15). A positive reaction was visualized with the chromogen substance diaminobenzidine tinted with CoCl2, which produces a black reaction product. The sections were counterstained with methylgreen (blue-green nuclei).
Based on our previous article (15), annexin uptake in untreated atherosclerotic animals was 0.054 ± 0.0095 %ID/g. With a sample of 6 untreated animals (treatment control group) in the present study, 3 animals per caspase inhibitor treatment group would provide 94% power (1-tailed alpha = 0.05) to detect a 40% reduction in annexin uptake (≤0.0324 %ID/g uptake). We felt this to be an important magnitude of reduction to demonstrate, based on our previous observation showing chronic statin treatment and dietary modification to reduce annexin update by this magnitude (6).
The gamma scintillation counts were calculated as %ID/g of tissue or blood. Results were expressed as mean ± standard deviation. The Student ttest was initially used to compare uptake between untreated atherosclerotic rabbits and all caspase inhibitor-treated groups combined. To determine the statistical significance of differences between individual caspase inhibitor and control groups, 1-way analysis of variance was performed followed by Scheffe’s post hoc test for multiple comparisons. An alpha level <0.05 after adjustment for multiple comparisons was considered statistically significant.
In vivo annexin imaging
Gamma imaging allowed noninvasive visualization of atherosclerotic lesions in the abdominal aorta; best visible were the untreated atherosclerotic plaques (Fig. 3).No annexin uptake was observed in normal (nonatherosclerotic, disease control) animals. Similarly, no tracer uptake was seen in atherosclerotic rabbits imaged with mutant annexin A5 (radiotracer control). Annexin uptake was significantly lower in the groups of animals treated with broad caspase inhibitor and caspase-3 inhibitor compared with the untreated atherosclerotic rabbits. Inhibition of caspase-1 and -9 was also associated with substantial reduction in annexin uptake. The uptake was not affected in the caspase-8 inhibitor-treated group. The ex vivo gamma images of the explanted aortic specimens verified the in vivo results.
Quantitative annexin A5 uptake in treated and untreated lesions
The quantitative annexin uptake in the atherosclerotic specimens, calculated as %ID/g of aortic tissues, should represent the extent of apoptotic activity in the lesions. Maximum annexin uptake was observed in the untreated atherosclerotic lesions (0.0515 ± 0.0099%), which was 10-fold higher than that in the aortic specimens obtained from disease control rabbits (Fig. 4).Among all caspase inhibitor-treated animals (combining broad and caspase-1, -3, -8, and -9 groups; n = 19), there was an overall 39% lower annexin uptake (0.0314 ± 0.0151%; p < 0.01 compared with untreated atherosclerotic rabbits). Among individual caspase inhibitor groups, the annexin uptake was significantly reduced in broad caspase (0.0206 ± 0.0058%; p < 0.0001) and caspase-3 inhibitor (0.0286 ± 0.0095%; p = 0.0034) groups (compared with untreated atherosclerotic lesions). Annexin uptake in the broad caspase inhibitor group was not significantly different than the annexin uptake in normal aortic specimens (p = NS), suggesting almost complete (and acute) abrogation of apoptosis in atherosclerotic lesions. The uptake also was reduced markedly in caspase-9 (0.03 ± 0.0021%; p = 0.0024) and -1 inhibitor (0.0272 ± 0.0088%; p = 0.0016) groups. Annexin uptake remained unchanged for the caspase-8 inhibitor group (0.0618 ± 0.0047%; p = NS compared with the untreated group and p < 0.0001 compared with the disease control group). Together, these data suggest involvement of the mitochondrial pathway of apoptosis in atherosclerosis. Finally, there was no uptake of mutant annexin in atherosclerotic lesions, and quantitative uptake was as low as annexin uptake in normal rabbit aortas (0.014 ± 0.0024%; p = NS). The biodistribution of radiotracer in nontarget organs showed maximal radiation burden in the renal cortex (6.2 ± 1.1 %ID/g), spleen (0.37 ± 0.14%), and liver (0.15 ± 0.05%).
Histology of atherosclerotic lesions
Our previous study had shown that specific annexin uptake preferentially occurred in the advanced American Heart Association type IV atherosclerotic plaque; the uptake was traced to the apoptotic foam cells by dual immunohistochemical staining techniques (15). Similarly, in the present study, the lesions were predominantly foam cell rich with minimal smooth muscle cell content and showed extensive apoptosis (Fig. 4). The animals treated with the broad caspase inhibitor also revealed foam cell-rich lesions histopathologically similar to those in untreated rabbits; however, almost no TUNEL-positive apoptotic nuclei were observed (Fig. 4). The lipid cores with predominant cholesterol clefts and abundant macrophages were visible in the lesions and remained unaffected after administration of caspase inhibitors. A similar phenomenon was observed in the animals treated with selective caspase-1 and -3 inhibitors. However, apoptosis was unaffected in the selective caspase-8 inhibitor group and was found to be similar to untreated animals (Fig. 5).
The present study demonstrated the feasibility of acute resolution of apoptosis in atherosclerotic plaques with broad and selective caspase inhibitors. This study also demonstrated the feasibility of employing radioactive annexin A5 for evaluation of apoptosis resolution by noninvasive imaging. Because apoptosis of macrophages contributes significantly to plaque rupture, it is logical to presume that acute repression of apoptosis may be of therapeutic benefit in acute coronary events.
The data support efficacy of broad caspase and caspase-9 and -3 inhibitors for acute repression of apoptosis in atherosclerosis. Because caspase-8 inhibition had no significant effect on apoptosis and caspase-9 inhibition significantly blocked it, the apoptosis of foam cells is likely to involve the mitochondrial pathway and not the death receptor pathway. In an earlier in vitro study, caspases-3 and -9 played a critical role in the apoptosis of cultured vascular smooth muscle cells in atherosclerotic lesions, and the inhibitors of both caspase-9 and -3 markedly reduced apoptosis (21).
We observed that the caspase-1 inhibitor also blocked apoptosis to a level similar to the broad caspase and specific caspase-3 inhibitors, suggesting that caspase-1 may also be involved in apoptosis in atherosclerotic lesions (Fig. 1). A previous autopsy report also demonstrated the role of caspase-1 activation in acute coronary syndrome (2). Activation of caspase-1 in apoptosis in atherosclerosis is intriguing (22,23). The pro-domain of caspase-1 enhances caspase-8 activation during CD95 (fas)-mediated apoptosis (24), the specific caspase-1 inhibitor blocks CD95-mediated apoptosis in rat acinar cells (25), and caspase-1–deficient mice are resistant to CD95-mediated apoptosis (26). These studies suggested that caspase-1 may be involved in the death receptor or extrinsic pathway of apoptosis. However, we observed no effect of the caspase-8 inhibitor on apoptosis in atherosclerotic lesions, excluding an involvement of the extrinsic pathway. Caspase-1 involvement in the mitochondrial pathway of apoptosis has recently been demonstrated in traumatic brain injuries in children. Increased levels of cytochrome c, caspase-3, and caspase-1 have been observed in the cerebrospinal fluid, and caspase-3 and -1 activation have been observed in brain tissue (27). It has been proposed that reactive oxygen species generation is downstream of caspase-1 (28,29). Because reactive oxygen species generation mediates apoptosis via cytochrome c release and activation of caspase-9 and -3 (30), it is likely that caspase-1 is upstream of caspase-9 and -3. Our data may indirectly support these observations because the caspase-1 inhibitor blocks apoptosis in atherosclerotic lesion to a similar extent as the inhibitor of caspase-9. Interestingly, activation of caspase-11 has been demonstrated during ischemia-induced apoptosis in the brain (31), which was associated with subsequent activation of caspase-1 and -3. It is unclear if there is activation of caspase-11 in experimentally induced atherosclerotic lesions.
Acute resolution of apoptosis may have translational clinical implications. Because extensive apoptosis has been demonstrated in acute coronary events, acute repression of apoptosis may offer novel avenues for intervention. Such a proposal may be of added value because resolution of apoptosis has also been shown to be beneficial in limiting myocardial injury in experimental myocardial infarction.
The present study indicated that broad and specific caspase inhibitor therapy can acutely reduce the incidence of apoptosis in experimental atherosclerotic lesions. A mitochondrial pathway appears to be predominantly involved in induction of apoptosis, and a novel role of caspase-1 in apoptosis in atherosclerotic lesion is suggested. If verified in clinical trials, repression of apoptosis by specific caspase inhibitors may add to intervention strategies in acute coronary syndromes. Our findings also support the value of annexin A5 imaging as a noninvasive tool to monitor pharmacologic effects of therapies aimed to reduce apoptosis.
↵1 Drs. Sarai and Hartung contributed equally to this study.
This study was supported by the National Institutes of Health, grant #RO1 HL68657 (to Dr. Narula).
Dan Berman, MD, served as Guest Editor for this article.
- Abbreviations and Acronyms
- computed tomography
- percent injected dose per gram
- single photon emission computed tomography
- terminal deoxyribonucleotide transferase-mediated nick-end labeling
- Received April 10, 2007.
- Revision received August 23, 2007.
- Accepted August 24, 2007.
- American College of Cardiology Foundation
- Geng Y.J.,
- Libby P.
- Hartung D.,
- Sarai M.,
- Petrov A.,
- et al.
- Fuster V.,
- Gotto A.M.,
- Libby P.,
- Loscalzo J.,
- McGill H.C.
- Hengartner M.O.
- Ashkanazi A.,
- Dixit V.M.
- Gupta S.,
- Reutelingsperger C.,
- Narula J.
- Kietselaer B.L.,
- Reutelingsperger C.,
- Boersma H.H.,
- et al.
- Kolodgie F.D.,
- Petrov A.,
- Virmani R.,
- et al.
- Isobe S.,
- Tsimikas S.,
- Zhou J.,
- et al.
- Johnson L.L.,
- Schofield L.,
- Donahay T.,
- Narula N.,
- Narula J.
- Blankenberg F.G.,
- Katsikis P.D.,
- Tait J.F.,
- et al.
- Stary H.C.,
- Chandler A.B.,
- Glagov S.,
- et al.
- Chen J.,
- Mehta J.L.,
- Haider N.,
- Zhang X.,
- Narula J.,
- Li D.
- Syed F.M.,
- Hahn H.S.,
- Odley A.,
- et al.
- Rowe S.J.,
- Allen L.,
- Ridger V.C.,
- Hellewell P.G.,
- Whyte M.K.
- Tatsuta T.,
- Shiraishi A.,
- Mountz J.D.
- Kuida K.,
- Lippke J.A.,
- Ku G.,
- et al.
- Satchell M.A.,
- Lai Y.,
- Kochanek P.M.,
- et al.
- Schulz J.B.,
- Weller M.,
- Klockgether T.