Matrix Metalloproteinase Inhibition Attenuates Aortic Calcification

【摘要】  Objective- Arterial calcification has been associated with matrix metalloproteinase (MMP)-mediated elastin degradation. In this study, we investigated whether inhibiting MMP activity could reduce calcium accumulation in rodent models of aortic calcification.

Methods and Results- Aortic calcification was first induced in male Sprague-Dawley rats by administration of vitamin D 3. Treatment with doxycycline decreased aortic calcium and phosphorus accumulation, and it reduced aortic gelatinase levels; however, it also prevented the bone resorption associated with high doses of vitamin D 3. Using an in vivo model of localized aortic calcification, systemic doxycycline treatment reduced aortic calcium accumulation without affecting serum calcium levels, suggesting a more specific effect of doxycycline in the arterial wall. In organ culture, doxycycline limited aortic calcification caused by exposure to alkaline phosphatase and inorganic phosphate. When GM6001, a synthetic and specific inhibitor of MMPs, was used instead of doxycycline, it had a similar effect. In vivo, periadventitial delivery of GM6001 to calcifying arteries significantly reduced calcification compared with controls.

Conclusions- These results suggest that MMPs are involved in aortic calcification, and inhibiting MMP activity could reduce calcium accumulation in the arterial wall.

Arterial calcification is associated with MMP-mediated matrix degradation. Inhibiting MMP activity with doxycycline or the synthetic MMP-inhibitor GM6001 prevents calcium accumulation in the arterial wall.

【关键词】  artery calcification MMPs doxycycline GM rat

Introduction

Vascular calcification occurs pathologically in diabetes and chronic renal disease as well as during the normal aging process. 1-3 When calcification is identified in the coronary arteries, it is associated with an increased risk of cardiac events. 4,5 In patients with renal failure, it is associated with significantly shorter survival, 6 and in patients with diabetes, it occurs at an accelerated rate and is a strong predictor of morbidity and mortality. 7 Although vascular calcification is associated with atherosclerosis, recent data suggest that it is independently regulated, and calcifying vessels share features with osteogenesis. 2,8 Consistent with this hypothesis is that diseased human arteries have been shown to express proteins usually seen in bone, including members of the bone morphogenetic protein family, noncollagenous bone matrix proteins, matrix Gla protein, decorin, the osteoblastic regulator osteoprotegrin, and the matrix metalloproteinases (MMPs). 9-15

MMPs are involved in multiple processes in the vascular wall. 16 They are upregulated in human atherosclerotic and restenotic lesions, and they are involved in aneurysm formation. 17-21 Inhibiting MMP activity prevents injury-induced arterial remodeling in rodent models. 22,23 Recent studies demonstrated a correlation between MMP-mediated elastin degradation and aortic calcification. Inhibiting elastin degradation with aluminum ions prevented calcification in the aortas of rats after calcium chloride-mediated injury; and mice deficient in MMP-2 and MMP-9 did not develop calcification in a similar model. 24 Additionally, Lee et al demonstrated increased MMP activity associated with increased levels of soluble elastin peptides in a rat subdermal model of elastin calcification. 25 In the present study, we sought to determine whether inhibiting MMP activity could prevent calcium accumulation in experimental models of aortic calcification. We demonstrate that doxycycline, an antibiotic that inhibits MMP activity and reduces MMP levels, and GM6001, a synthetic selective MMP inhibitor, can inhibit arterial calcification both in organ culture and in vivo.

Materials and Methods

A detailed Methods section is available in the online supplement, available at http://atvb.ahajournals.org.

Experimental Procedures

Arterial calcification was induced in male Sprague-Dawley rats by subcutaneous injection of 7.5 mg/mL vitamin D 3 for 3 consecutive days. Rats received daily doxycycline by subcutaneous injection beginning at 2 days before vitamin D 3 injections and ending at the time of harvest. In another group of rats, aortic calcification was induced by periadventitial application of 0.15 mol/L CaCl 2 as described recently. 24

Calcium, Phosphorus, Bone Densitometry Measurements

Detailed methods for calcium, phosphorus, and bone densitometry measurements are available in the online supplement.

Organ Culture

Segments of rat aorta measuring 1 cm underwent calcification as described previously. 26 Doxycycline (100 µg/mL) was pre-equilibrated in medium and then added to the culture medium. In a separate series of experiments, the synthetic MMP inhibitor GM6001 (12.5 or 25 µmol/L) or GM6001-negative control was used instead of doxycycline.

In Vivo Delivery of GM6001

For in vivo evaluation of GM6001, rats underwent periadventitial application of CaCl 2 solution or normal saline. The synthetic MMP inhibitor GM6001 or GM6001-negative control was then administered via the catheter on a daily basis until the time of aortic harvest.

Results

Doxycycline Inhibits Vitamin D 3 -Induced Aortic Calcification In Vivo

Male Sprague-Dawley rats (n=44) were divided into 6 groups that received 3 injections of 7.5 mg/kg vitamin D 3 or vehicle and either 0, 30, 60, or 120 mg/kg doxycycline by daily subcutaneous injection. Histological evaluation of treated and control specimens by von Kossa staining demonstrated extensive medial destruction and calcium staining in vitamin D 3 -injected specimens, but this was decreased with increasing dosages of doxycycline. Additionally, at the highest dose of doxycycline used, there was preservation of the elastin structure and medial vascular morphology (supplemental Figure I, available online at http://atvb.ahajournals.org). After 12 days, aortas from animals injected with vitamin D 3 had increased aortic calcification compared with control animals (0.29±0.13 µmol/L/mg calcium in control versus 9.54±1.15 µmol/L/mg in vitamin D 3 -injected rats; P <0.05). In vitamin D 3 -injected animals that received increasing amounts of doxycycline, there was a dose-dependent decrease in aortic calcium content. Aortic calcium content was 9.54±1.15 µmol/L/mg for 0 mg/kg doxycycline, 5.12±0.28 for 30 mg/kg doxycycline, 3.11±1.12 for 60 mg/kg doxycycline, and 1.49±0.63 for 120 mg/kg doxycycline (ANOVA, P <0.05; vitamin D 3 alone compared with vitamin D 3 plus doxycycline groups, P <0.05; Figure 1 ). This was mirrored by a dose-dependent decrease in aortic phosphorus content from 0.10±0.01 µmol/L/mg in rats receiving vitamin D 3 to 0.039±0.001 for 30 mg/kg doxycycline, 0.022±0.003 for 60 mg/kg doxycycline, and 0.021±0.004 for 120 mg/kg doxycycline. The average weight change in vitamin D 3 -injected animals was different from rats that did not receive vitamin D 3 (-16±9.3 g for vitamin D 3 -injected versus 120±13.6 for control animals; P <0.05), and this weight change was not affected by doxycycline treatment (-22±5.0 for vitamin D 3 plus doxycycline; P =NS).

Figure 1. Doxycycline (doxy) decreases aortic calcium and phosphorus accumulation and preserves arterial morphology in vitamin D 3 (Vit D3)-injected rats. Aortic calcium and phosphorous were measured in 6 groups of rats that received vehicle or vitamin D 3 and 0 (n=9), 30 (n=5), 60 (n=6), or 120 (n=8) mg/kg doxycycline by subcutaneous injection. ANOVA for all groups P <0.05; * P <0.05 compared with negative controls (first group); ** P <0.05 compared with vitamin D 3 -treated rats (third group).

Effect of Doxycycline on Vitamin D 3 -Induced Aortic MMP Expression, Serum Calcium, and Bone Mineral Density

In rats given vitamin D 3, we observed a decrease in aortic gelatinase levels from rats treated with 60 mg/kg doxycycline compared with those treated with vehicle ( Figure 2 A). The band corresponded to the band known to be MMP-9 found in conditioned medium from the human HT 1080 cell line (supplemental Figure II). No consistent changes were observed in bands located where MMP-2 is usually seen.

Figure 2. Effects of doxycycline (doxy) on aortic MMP-9 activity, serum calcium, and bone density in vitamin D 3 (Vit-D3)-injected rats. A, aortic MMP-9 levels in vitamin D 3 -injected rats (n=7), controls (n=6), and vitamin D 3 -injected rats that were treated with doxycycline (n=7). B, Serum calcium concentration in vitamin D 3 -injected rats and rats treated with doxycycline. C, Mineral bone density of midshaft femur in rats at 6 days after vitamin D 3 injection. * P <0.05 compared with control; ** P <0.05 compared with vitamin D 3 -injected rats.

We then evaluated the changes in serum calcium concentration caused by vitamin D 3 injection and 60 mg/kg doxycycline treatment. Serum calcium was significantly increased in vitamin D 3 -injected rats compared with controls, and this was not affected by treatment with doxycycline. The calcium concentration was 10.58±0.54 mg/dL in control rats versus 15.23±0.44 in vitamin D 3 -injected rats ( P <0.05 versus controls) and 15.95±0.84 in vitamin D 3 -injected plus doxycycline ( P =NS compared with vitamin D 3 alone; Figure 2 B).

Bone density was significantly lower in vitamin D 3 -injected rats than in controls, but doxycycline treatment partially reversed this decrease. Bone density was 0.120±0.002 g/cm 2 in controls compared with 0.087±0.002 in vitamin D 3 -injected rats ( P <0.05 versus control) and 0.101±0.005 in vitamin D 3 -injected plus doxycycline rats ( P <0.05 versus controls; P <0.05 versus vitamin D 3 -injected; Figure 2 C). This suggests that some effects of doxycycline on aortic calcification could occur through altering the increased bone resorption caused by high doses of vitamin D 3.

Doxycycline Inhibits Aortic Calcification Caused by Periadventitial CaCl 2 Administration

We next evaluated whether the effects of doxycycline could be reproduced in a model of arterial calcification that is not characterized by alterations in bone resorption. We studied the effects of doxycycline treatment in rats that underwent periadvential administration of 0.15 mol/L CaCl 2. Animals not treated with doxycycline demonstrated medial and adventitial accumulation of calcium by von Kossa staining. However, animals treated with doxycycline had nearly complete inhibition of calcification as assessed by the von Kossa stain and by measurement of aortic calcium in harvested specimens. ( Figure 3A and 3 B). The aortic calcium content was 0.06±0.015 µmol/L/mg for NaCl control rats (n=3), 0.33±0.06 for CaCl 2 rats not given doxycycline (n=4), and 0.13±0.11 µmoles/mg for rats treated with 60 mg/kg doxycycline (n=5; P <0.05; Figure 3 C). Additionally, immunohistochemical staining of these specimens using the macrophage-specific antibody ED-1 demonstrated macrophage infiltration in the adventitia of untreated rats, but this was not see in doxycycline-treated rats (supplemental Figure III). Serum calcium levels in doxycycline-treated rats were not different from control rats (12.0±0.13 for controls versus 11.27±0.35 for doxycycline-treated rats; P =NS).

Figure 3. Doxycycline reduces aortic calcium accumulation and adventitial macrophage infiltration in rats undergoing periadventitial administration of CaCl 2. A and B, von Kossa-stained aortic sections from rat that underwent periadventitial application of calcium chloride and either no doxycycline (-doxy; A) or doxycycline (+doxy; B). Arrows indicate medial calcium; m, medium; a, adventitia. C, Calcium in aortas from rats that underwent application of NaCl (n=3), CaCl 2 (n=4), and CaCl 2 rats treated with 60 mg/kg doxycycline (n=5). * P <0.05 compared with NaCl control rats; ** P <0.05 compared with CaCl 2 rats not given doxycycline.

In Vitro Effect of Doxycycline and GM6001 on Arterial Calcification

We next investigated whether the inhibitory effects of doxycycline on in vivo calcification could be reproduced in an organ culture system. This has the benefit of excluding the effects of doxycycline on systemic, circulating factors. Aortas were calcified in an organ culture system by addition of calf intestinal alkaline phosphatase and sodium phosphate to DMEM without serum (calcification medium). 26 von Kossa-stained sections demonstrated calcium accumulation in the adventitia and outer medial layers of vessels cultured in calcification medium but little calcification in those also treated with doxycycline (supplemental Figure IV). The calcium content of vessels cultured in DMEM alone was 0.284±0.090 µmol/L, and there was a 3-fold increase in calcium content to 0.953±0.280 µmol/L for those cultured in calcification medium. Arteries cultured in calcification medium with 100 µg/mL doxycycline demonstrated a significant decrease in calcium accumulation (0.365±0.028; P <0.05 compared with calcification medium alone; Figure 4 A). Gelatinase levels in the medium of doxycycline-treated samples were also reduced ( Figure 4 B).

Figure 4. Addition of doxycycline (doxy) to calcifying aortas in organ culture inhibits calcification and reduces MMP levels. A, Aortic calcium content measured in HCl extracts from 1-cm aortic specimens taken after 9 days in organ culture with the specified treatments. B, MMP levels in conditioned medium determined by gelatin zymography from cultured aortas. D0 indicates DMEM without serum; CM, calcification medium. n=5 for each group; * P <0.05 compared with control; ** P <0.05 compared with CM groups. This experiment was performed twice with similar results.

Although doxycycline has been useful for evaluating a drug with MMP-inhibiting properties, its experimental use is limited because it has several biological effects that are not related to MMPs. To determine whether the effects of doxycycline could be reproduced using a compound with more specific anti-MMP activity, we used the synthetic MMP inhibitor GM6001. Histological evaluation of these segments by von Kossa?s stain failed to demonstrate calcium accumulation in GM6001-treated arteries ( Figure 5A and 5 B). Aortas exposed to calcification medium showed an increase in calcium content compared with control vessels (0.211±0.016 µmol/L for control compared with 2.525±0.670 for calcification medium). When GM6001 was added to the calcification medium, there was a dose-dependent reduction in aortic calcium accumulation. Aortic calcium was 2.331±0.456 µmol/L for 12.5 µmol/L ( P =NS) and 0.434±0.088 for 25 µmol/L GM6001 ( P <0.05 versus no GM6001; Figure 5 C). Viability of cultured segments at 7 days was not affected by calcification medium or GM6001, as evidenced by measurement of lactic dehydrogenase secretion into the medium, which was not different from control segments (data not shown).

Figure 5. The MMP inhibitor GM6001 inhibits in vitro aortic calcification. A and B, von Kossa-stained section of artery incubated in calcification medium (A) or in an artery incubated in calcification medium with 25 µmol/L GM6001 (B). C, Graph of calcium accumulation in artery wall of vessels cultured in calcification medium (CM) with 12.5 or 25 µmol/L GM6001. m indicates medium; arrows, calcium; a, adventitia (n=6 for each group; * P <0.05 compared with control; ** P <0.05 compared with CM-treated groups). This experiment was performed 3 times with similar results.

GM6001 Inhibits Aortic Calcification In Vivo

Based on our findings in organ culture, we next evaluated whether GM6001 could be used to inhibit aortic calcification in vivo. Daily injections of GM6001 or its negative control (250 µg per day) were administered by catheter into the periaortic space after periadventitial application of calcium chloride. Aortas from control rats that underwent painting with NaCl had minimal calcification after 7 days. However, CaCl 2 -painted arteries 10-fold increase in calcium accumulation, and this was significantly decreased by local delivery of GM6001 (0.166±0.026 µmol/L for normal saline controls; 1.983±0.477 for CaCl 2 with GM6001-negative control; and 0.656±0.331 for CaCl 2 painted vessels treated with GM6001; P <0.05). Aortic histology from rats that received the negative control compound demonstrated extensive calcification in the adventitia and in the outer medial layers, whereas there was only occasional calcium staining in aortas from rats treated with GM6001 ( Figure 6 ).

Figure 6. The MMP inhibitor GM6001 inhibits aortic calcification in vivo. Rats were subjected to periadventitial application of CaCl 2 or normal saline, and then they were treated with either GM6001 or its negative control compound (neg-cont). A and B, von Kossa staining shows calcium in aortas from rats that did not receive GM6001 (-GM6001; A) and in aortas from rats that received local treatment with GM6001 (+GM6001; B). arrows indicate calcium; m, medium; a, adventitia. C, Aortic calcium concentration in saline controls (n=6), CaCl 2 -painted rats that received GM6001-negative control (n=11), and CaCl 2 -painted vessels that received GM6001 (n=9); ANOVA P <0.05; * P <0.05 compared with saline control; ** P <0.05 compared with rats receiving negative-control compound.

Discussion

In this study, we found that inhibiting MMP activity is associated with decreased calcium accumulation in the arterial wall. We began our studies by using doxycycline, an antibiotic with known MMP-inhibiting properties, to demonstrate an association between decreased aortic gelatinase levels and reduced aortic calcification. To explore the importance of MMPs in arterial calcification more directly, we used the synthetic MMP-specific inhibitor GM6001 and were able to demonstrate that it decreased aortic calcification both in vitro and in vivo. These findings suggest that MMPs play an important role in arterial calcification, and further, that inhibiting MMP activity may prevent arterial calcification in the clinical setting.

The changes in arterial histology associated with hypervitaminosis D have been well characterized. 27-29 Rats exposed to sublethal doses of vitamin D initially develop calcium phosphate crystals along the elastic laminae followed by spreading of calcification into the surrounding medial smooth muscle cells. 27 The model is characterized by increased bone resorption and increased intestinal uptake of ingested calcium, leading to elevated serum calcium levels. 28 Although the mechanisms related to arterial calcification in this model have not been fully elucidated, it is thought to occur through an interaction between degraded medial elastin and circulating factors from resorbed bone. Price et al demonstrated recently that a fetuin-matrix Gla protein-mineral complex is increased in the circulation of rats exposed to toxic doses of vitamin D, whereas serum levels of fetuin, a proposed inhibitor of calcification, are decreased. 30 They propose that high levels of fetuin in complex may prevent it from inhibiting the growth of mineral components in the elastic layer of the vessel wall. Alternatively, the fetuin-mineral complex might serve as a marker for a presently unknown causative factor.

In our experiments, administration of vitamin D 3 increased MMP-9 levels in the aortic wall and increased serum calcium levels. Doxycycline treatment reduced aortic MMP-9 levels back to normal, but it had no effect on serum calcium levels. This suggests that doxycycline may act in a manner that is similar to ibandronate or osteoprotegrin, which inhibit bone resorption but do not affect serum calcium levels in vitamin D 3 -treated rats. 31-33 However, we have not excluded the possibility that the effects of doxycycline are related to other mechanisms such as increases in inhibitory factors, decreases in positive regulatory factors, direct binding of doxycycline to hydroxyapatite, antibiotic properties, or other yet unknown factors. Additionally, there are known interactions between doxycycline and calcium, and these may have occurred in our model despite the lack of effect on serum calcium levels. Also notable is the significant weight loss seen in vitamin D 3 -treated rats. This weight loss was not reversed by treatment with doxycycline, which suggests that vitamin D 3 administration might have indirect consequences. For example, leptin levels may have been affected in this model, and this may have resulted in indirect effects on vascular calcification. 34,35

The importance of MMPs in arterial calcification has been suggested previously by Basalyga et al, who demonstrated that mice deficient in MMP-2 or MMP-9 failed to develop aortic calcification after perivascular application of CaCl 2. 24 Additionally, Lee et al recently demonstrated by gelatin zymography and RT-PCR that MMP-2 and MMP-9 levels are increased in subdermally placed elastin that is undergoing active calcification. 25 In the present series of experiments, we demonstrated that MMP-9 levels are increased in calcifying aortas, and that decreased gelatinase levels are associated with a reduced amount of aortic calcium accumulation. However, other MMPs are also capable of degrading medial elastin, and it is possible that MMPs other than the gelatinases could be responsible for the elastin disruption seen in this model.

A mechanism for elastin calcification that involves a positive feedback loop between degraded elastin and MMP gene expression has been proposed recently. 25 According to this model, physical or biochemical injury may disrupt the glycoproteins normally surrounding elastin. This may serve to expose elastin to cells that, in turn, produce MMPs or other serine proteases that can degrade elastin. MMPs can then further degrade both the protective glycoproteins in the arterial wall and the exposed elastin. Elastin degradation may also release matrix-bound cytokines that can recruit inflammatory cells and cause smooth muscle cells to undergo a phenotypic modulation into more osteoblastic-type cells. Our data on aortic gelatinase levels during calcification are consistent with this model and furthermore suggest that the enzymes required for elastin degradation can be generated within the arterial wall.

A recently described model of aortic calcification involves the periadventitial application of CaCl 2. 24 Because calcification in this model is related to local factors, it is not associated with altered circulating levels of calcium, phosphate, or other products of bone resorption. Using this model, we were also able to demonstrate a significant reduction in aortic calcium accumulation by systemic treatment with doxycycline. The inhibitory effects of doxycycline on aortic calcification in this model argue against a mechanism related specifically to circulating mineral complexes; however, there are ongoing low levels of bone resorption that may be inhibited by doxycycline, and even these minimal changes may be sufficient to affect arterial calcification. Additionally, we noted strong ED-1-specific staining in the adventitia of calcifying aortas, but this was not seen in doxycycline-treated rats. Although the role of macrophages in this model has not been defined, it is possible that they are another source of matrix-degrading enzymes necessary for the calcification to proceed. The potential role of inflammatory cells and their secreted mediators on arterial calcification has been reviewed recently, 36 and further studies in this regard are needed.

Our experiments in organ culture allowed us to focus on the direct effects of MMP inhibition on the aortic wall. In this system, calcification is thought to occur because of inhibition of pyrophosphate production and increased calcium x phosphorus product in the medium. 26 Doxycycline reduced aortic calcium levels, and this effect was associated with decreased MMP levels in the culture medium. This likely occurred through the previously described inhibitory effects of doxycycline on MMP synthesis. 37 However, the use of doxycycline in mechanistic studies is limited because of its diverse actions, many of which are not related to altering MMP levels. For this reason, we extended our experiments to include the synthetic compound GM6001 that specifically inhibits the MMPs. We found that addition of GM6001 to the calcification medium resulted in a dose-dependent decrease in arterial calcium accumulation. On histology, GM6001-treated vessels had less calcium deposition in the adventitia and outer medial layers. GM6001 is a potent, hydroxamic acid-based MMP inhibitor with activity against several MMPs, including collagenase, the gelatinases, and stromelysin. Both the gelatinases and stromelysin have elastolytic activity, and a possible role of MMPs in medial calcification is in initiating the process of elastin degradation that may serve to provide a nidus for hydroxyapatite crystals to develop. These in vitro results suggested that synthetic MMP inhibitors might prevent arterial calcification in vivo.

Our final series of experiments involved the use of GM6001 in vivo. For these studies, GM6001 was delivered locally, near the aorta at a concentration that was similar to that used in organ culture, and we found that daily administration reduced aortic calcium content compared with negative control treated vessels. Although the source of MMP activity in this model is presently unknown, the potential sources include medial smooth muscle cells, adventitial fibroblasts, or infiltrating inflammatory cells, all of which would have been affected by local delivery of the inhibitor. GM6001 is a synthetic inhibitor with activity against several MMPs, and thus, the identity of MMPs involved in arterial calcification remains to be determined. Based on the current model, it is likely to be an MMP with elastolytic properties that can be produced either within the aortic wall or by cells that can reside in the adventitia. One possibility is that GM6001 has direct effects on the CaCl 2 solution used to inhibit calcification. However, the accumulation of aortic calcium we observed with the negative control compound argues against this point. Another limitation of this technique is that compound delivery is limited to 5 to 7 days because of the formation of a fibrous capsule around the catheter. However, the fact that catheter delivery was successful suggests that aortic calcification occurs early in this model and that it can be inhibited by drug administration during the first several days. Alternatively, sufficient compound may have diffused through the capsule to have an effect at later time points. Efforts to understand the temporal aspects of this model and the relative contributions of MMPs during the various phases of arterial calcification are under way.

Together, our data suggest that MMPs are involved in arterial calcification and that inhibiting MMPs may be a clinically useful method for reducing calcification in the vessel wall. Arterial calcification is a complex phenomenon that is likely regulated at several levels, and further efforts to understand the role of MMPs in this pathologic process are needed.

Acknowledgments

We thank Dr Richard Hoover for his support and assistance in producing this manuscript.

Sources of Funding

This research was supported by mentored clinical scientist development award HL069926 from the National Heart, Lung, and Blood Institute (R.G.) and grants from the Lifeline Foundation and the William J. von Liebig Foundation.

Disclosures

None.

【参考文献】
  Burke SK. Arterial calcification in chronic kidney disease. Semin Nephrol. 2004; 24: 403-407.

Abedin M, Tintut Y, Demer LL. Vascular calcification: mechanisms and clinical ramifications. Arterioscler Thromb Vasc Biol. 2004; 24: 1161-1170.

Speer MY, Giachelli CM. Regulation of cardiovascular calcification. Cardiovasc Pathol. 2004; 13: 63-70.

Arad Y, Spadaro LA, Goodman K, Newstein D, Guerci AD. Prediction of coronary events with electron beam computed tomography. J Am Coll Cardiol. 2000; 36: 1253-1260.

Taylor AJ, Bindeman J, Feuerstein I, Cao F, Brazaitis M, O?Malley PG. Coronary calcium independently predicts incident premature coronary heart disease over measured cardiovascular risk factors: mean three-year outcomes in the Prospective Army Coronary Calcium (PACC) Project. J Am Coll Cardiol. 2005; 46: 807-814.

London GM, Guerin AP, Marchais SJ, Metivier F, Pannier B, Adda H. Arterial media calcification in end-stage renal disease: impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant. 2003; 18: 1731-1740.

Wolfe ML, Iqbal N, Gefter W, Mohler ER III, Rader DJ, Reilly MP. Coronary artery calcification at electron beam computed tomography is increased in asymptomatic type 2 diabetics independent of traditional risk factors. J Cardiovasc Risk. 2002; 9: 369-376.

Vattikuti R, Towler DA. Osteogenic regulation of vascular calcification: an early perspective. Am J Physiol Endocrinol Metab. 2004; 286: E686-E696.

Bostrom K, Watson KE, Horn S, Wortham C, Herman IM, Demer LL. Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest. 1993; 91: 1800-1809.

Shanahan CM, Cary NR, Metcalfe JC, Weissberg PL. High expression of genes for calcification-regulating proteins in human atherosclerotic plaques. J Clin Invest. 1994; 93: 2393-2402.

Bini A, Mann KG, Kudryk BJ, Schoen FJ. Noncollagenous bone matrix proteins, calcification, and thrombosis in carotid artery atherosclerosis. Arterioscler Thromb Vasc Biol. 1999; 19: 1852-1861.

Shanahan CM, Cary NR, Salisbury JR, Proudfoot D, Weissberg PL, Edmonds ME. Medial localization of mineralization-regulating proteins in association with Monckeberg?s sclerosis: evidence for smooth muscle cell-mediated vascular calcification. Circulation. 1999; 100: 2168-2176.

Dhore CR, Cleutjens JP, Lutgens E, Cleutjens KB, Geusens PP, Kitslaar PJ, Tordoir JH, Spronk HM, Vermeer C, Daemen MJ. Differential expression of bone matrix regulatory proteins in human atherosclerotic plaques. Arterioscler Thromb Vasc Biol. 2001; 21: 1998-2003.

Fischer JW, Steitz SA, Johnson PY, Burke A, Kolodgie F, Virmani R, Giachelli C, Wight TN. Decorin promotes aortic smooth muscle cell calcification and colocalizes to calcified regions in human atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2004; 24: 2391-2396.

Schoppet M, Al-Fakhri N, Franke FE, Katz N, Barth PJ, Maisch B, Preissner KT, Hofbauer LC. Localization of osteoprotegerin, tumor necrosis factor-related apoptosis-inducing ligand, and receptor activator of nuclear factor-kappaB ligand in Monckeberg?s sclerosis and atherosclerosis. J Clin Endocrinol Metab. 2004; 89: 4104-4112.

Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251-262.

Halpert I, Sires UI, Roby JD, Potter-Perigo S, Wight TN, Shapiro SD, Welgus HG, Wickline SA, Parks WC. Matrilysin is expressed by lipid-laden macrophages at sites of potential rupture in atherosclerotic lesions and localizes to areas of versican deposition, a proteoglycan substrate for the enzyme. Proc Natl Acad Sci U S A. 1996; 93: 9748-9753.

Nikkari ST, Geary RL, Hatsukami T, Ferguson M, Forough R, Alpers CE, Clowes AW. Expression of collagen, interstitial collagenase, and tissue inhibitor of metalloproteinases-1 in restenosis after carotid endarterectomy. Am J Pathol. 1996; 148: 777-783.

Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest. 1994; 94: 2493-2503.

Allaire E, Forough R, Clowes M, Starcher B, Clowes AW. Local overexpression of TIMP-1 prevents aortic aneurysm degeneration and rupture in a rat model. J Clin Invest. 1998; 102: 1413-1420.

Petrinec D, Liao S, Holmes DR, Reilly JM, Parks WC, Thompson RW. Doxycycline inhibition of aneurysmal degeneration in an elastase-induced rat model of abdominal aortic aneurysm: preservation of aortic elastin associated with suppressed production of 92 kD gelatinase. J Vasc Surg. 1996; 23: 336-346.

Mason DP, Kenagy RD, Hasenstab D, Bowen-Pope DF, Seifert RA, Coats S, Hawkins SM, Clowes AW. Matrix metalloproteinase-9 overexpression enhances vascular smooth muscle cell migration and alters remodeling in the injured rat carotid artery. Circ Res. 1999; 85: 1179-1185.

Bendeck MP, Conte M, Zhang M, Nili N, Strauss BH, Farwell SM. Doxycycline modulates smooth muscle cell growth, migration, and matrix remodeling after arterial injury. Am J Pathol. 2002; 160: 1089-1095.

Basalyga DM, Simionescu DT, Xiong W, Baxter BT, Starcher BC, Vyavahare NR. Elastin degradation and calcification in an abdominal aorta injury model: role of matrix metalloproteinases. Circulation. 2004; 110: 3480-3487.

Lee JS, Basalyga DM, Simionescu A, Isenburg JC, Simionescu DT, Vyavahare NR. Elastin calcification in the rat subdermal model is accompanied by upregulation of degradative and osteogenic cellular responses. Am J Pathol. 2006; 168: 490-498.

Lomashvili KA, Cobbs S, Hennigar RA, Hardcastle KI, O?Neill WC. Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin. J Am Soc Nephrol. 2004; 15: 1392-1401.

Bonucci E, Sadun R. Dihydrotachysterol-induced aortic calcification. A histochemical and ultrastructural investigation. Clin Orthop Relat Res. 1975; 283-294.

Tanimura A, Cho T, Tanaka S. Aortic changes induced by hypercholesterolemia and hypercalcemia in rats. Exp Mol Pathol. 1986; 44: 297-306.

Cannon EP, Williams BJ. Raised vascular calcium in an animal model: effects on aortic function. Cardiovasc Res. 1990; 24: 47-52.

Price PA, Williamson MK, Nguyen TM, Than TN. Serum levels of the fetuin-mineral complex correlate with artery calcification in the rat. J Biol Chem. 2004; 279: 1594-1600.

Price PA, Faus SA, Williamson MK. Bisphosphonates alendronate and ibandronate inhibit artery calcification at doses comparable to those that inhibit bone resorption. Arterioscler Thromb Vasc Biol. 2001; 21: 817-824.

Price PA, June HH, Buckley JR, Williamson MK. Osteoprotegerin inhibits artery calcification induced by warfarin and by vitamin D. Arterioscler Thromb Vasc Biol. 2001; 21: 1610-1616.

Price PA, Omid N, Than TN, Williamson MK. The amino bisphosphonate ibandronate prevents calciphylaxis in the rat at doses that inhibit bone resorption. Calcif Tissue Int. 2002; 71: 356-363.

Tartaglia LA. The leptin receptor. J Biol Chem. 1997; 272: 6093-6096.

Parhami F, Tintut Y, Ballard A, Fogelman AM, Demer LL. Leptin enhances the calcification of vascular cells: artery wall as a target of leptin. Circ Res. 2001; 88: 954-960.

Doherty TM, Asotra K, Fitzpatrick LA, Qiao J-H, Wilkin DJ, Detrano RC, Dunstan CR, Shah PK, Rajavashisth TB. Calcification in atherosclerosis: Bone biology and chronic inflammation at the arterial crossroads. Proc Natl Acad Sci U S A. 2003; 100: 11201-11206.

Liu J, Xiong W, Baca-Regen L, Nagase H, Baxter BT. Mechanism of inhibition of matrix metalloproteinase-2 expression by doxycycline in human aortic smooth muscle cells. J Vasc Surg. 2003; 38: 1376-1383.

作者单位：Department of Surgery, Division of Vascular Surgery, Vanderbilt University Medical Center (X.Q., M.A.C., R.J.G.), Nashville, Tenn; Guangxi Medical University First Affiliated Hospital, China (X.Q.); and the Departments of Cell and Developmental Biology (R.J.G.) and Cancer Biology (L.M.M.), Vanderbil