Matrix Metalloproteinase 2 Activation of Transforming Growth Factor-β1 (TGF-β1) and TGF-β1–Type II Receptor Signaling Within the Aged Arterial Wall
Objective— To study matrix metalloproteinase 2 (MMP-2) effects on transforming growth factor-β1 (TGF-β1) activation status and downstream signaling during arterial aging.
Methods and Results— Western blotting and immunostaining showed that latent and activated TGF-β1 are markedly increased within the aorta of aged Fisher 344 cross-bred Brown Norway (30 months of age) rats compared with adult (8 months of age) rats. Aortic TGF-β1–type II receptor (TβRII), its downstream molecules p-similar to mad-mother against decapentaplegic (SMAD)2/3 and SMAD4, fibronectin, and collagen also increased with age. Moreover, TGF-β1 staining is colocalized with that of activated MMP-2 within the aged arterial wall and vascular smooth muscle cell (VSMC) in vitro, and this physical association was confirmed by coimmunoprecipitation. Incubation of young aortic rings ex vivo or VSMCs in vitro with activated MMP-2 enhanced active TGF-β1, collagen, and fibronectin expression to the level of untreated old counterparts, and this effect was abolished via inhibitors of MMP-2. Interestingly, in old untreated rings or VSMCs, the increased TGF-β1, fibronectin, and collagen were also substantially reduced by inhibition of MMP-2.
Conclusions— Active TGF-β1, its receptor, and receptor-mediated signaling increase within the aortic wall with aging. TGF-β1 activation is dependent, in part at least, by a concomitant age-associated increase in MMP-2 activity. Thus, MMP-2–activated TGF-β1, and subsequently TβRII signaling, is a novel molecular mechanism for arterial aging.
Transforming growth factor β1 (TGF-β1) is a pluripotent growth factor implicated in various aspects of vascular development and structural remodeling in health and disease.1–3 Notably, TGF-β1 signaling is a key regulator of collagen and fibronectin (FN) expression that is linked to tissue fibrosis in various organs such as heart, liver, lung, and kidney.3–6 With advancing age, TGF-β1, collagen, and FN increase within the arterial wall, predominantly within the intimal and adventitial layers, and are linked to the age-associated increase in arterial stiffness.1,2,7–9 Subsequently, age-associated cross-linking of collagen attributable to nonenzymatic glycation/glycosylation reactions leads to a further loss of elasticity and increased stiffness.10 The modifications of matrix proteins alter the niche of resident vascular cells, such as vascular endothelial cells (ECs) and vascular smooth muscle cells (VSMCs), and their survival, migration, and proliferation capabilities.2
Matrix proteins like collagen and FN also sequester oxidized low-density lipoprotein, immunoglobulins, and growth factors such as TGF-β1.2 The TGF-β1 gene codes for both, the precursor TGF-β1 protein, which contains a 40-kDa N-terminal peptide known as latency-associated protein (LAP), and mature TGF-β1.11 In vitro studies indicate that TGF-β1 is secreted in a biologically inactive form in a complex with LAP.11 Within the extracellular matrix, this precursor TGF-β1 protein binds to latency TGF-binding protein (LTBP). LTBP contains a collagen-binding domain, and it is sequestered within collagen fibers after secretion from vascular cells.11 TGF-β1 secretion from cells is inefficient in the absence of LTBP.12 Although &10% of LTBP is bound to TGF-β1, the free LTBP, per se, plays an important role in the regulation of VSMC proliferation and migration.11,13 Thus, TGF-β1 is stored in a latent, macromolecular complex that is bound to the extracellular matrix. On activation by proteolytic cleavage of this protein complex, TGF-β1 binds to its cognate receptor, TGF-β1 type II receptor (TβRII), and activates the similar to mad-mother against decapentaplegic (SMAD) signaling pathway, regulating matrix-associated protein expression (ie, collagen and FN).3
The activation status of arterial wall TGF-β1 with aging is unknown, and age-associated changes in its underlying activation mechanisms have not been reported. Previous studies in several mammalian cell types have shown that the TGF-β1 complex is activated in vitro by matrix metalloproteinase 2 (MMP-2) and by activators of MMP-2 (ie, membrane type-1 MMP [MT1-MMP] and plasmin).6,14,15 An age-associated increase in activity of MMP-2 is a potential mechanism that could lead to an increase in TGF-β1 activation with aging because MT1-MMP, MMP-2, and plasmin activators such as tissue-type plasminogen activator (tPA) and urokinase-type plasminogen activator (uPA) increase in parallel with TGF-β1 within the arterial wall with aging, particularly within the thickened intima of old arteries.8,9,16
In the present study, we measured the levels of LTBP, LAP, activated TGF-β1, MMP-2 activity, TβRII, and its downstream molecules SMAD2/3, SMAD4, and SMAD7 in the aorta of young (8 months of age) and old (30 months of age) Fisher 344 cross-bred Brown Norway rats. In arterial rings ex vivo and early-passage VSMCs in vitro, we further characterized the role of MMP-2 activity in TGF-β1 activation and the production of the matrix molecules FN and collagen. Our results demonstrate novel links between MMP-2 activity, TGF-β1 activation, and collagen production in the context of arterial matrix remodeling with advancing age.
Materials and Methods
For details on materials and methods (including animals, organ culture, VSMC isolation and culture, RT-PCR, immunohistochemistry, immunofluorescence, immunoprecipitation, coimmunoprecipitation assays, Western blotting, in situ and PAGE zymography, in vitro activation assay of LAP-TGF-β1 and statistical analysis), please see the online supplement, available at http://atvb.ahajournals.org.
Aortic Wall TGF-β1 Transcript and Protein Expression Increase With Aging
Real-time quantitative (Q)-PCR showed that aortic TGF-β1 mRNA was increased 5.8-fold within the aorta of old rats compared with that of young rats (Figure 1A and 1B). Three major bands were present in Western blots of protein homogenates from rat aorta, corresponding to the molecular weights of activated TGF-β1 (&20 kDa), LAP-bound TGF-β1 (&75 kDa), and the LTBP-bound to precursor TGF-β1 (190 to 250 kDa; Figure 1C). Densitometric analysis showed that aortic TGF-β1 was mainly present (98%) in the latent form, bound to LTBP and LAP, and that all bands, including that of the active form of TGF-β1, increased with aging. On average, the precursor TGF-β1 bound to LTBP was increased by 3.2-fold, that bound to LAP by 3.7-fold, and active TGF-β1 was increased by 18-fold in the aortae of old versus young rats (Figure 1D).
Immunostaining with antibodies against the 3 different protein forms further confirmed that the abundance of TGF-β1, LAP, and LTBP proteins increased within the aged aortic wall, particularly within the thickened intima (Figure 1E, right panels). TGF-β1 staining within the aortic walls of aged rats was dramatically increased in both intracellular and extracellular regions. Interestingly, the stronger immunostaining signal for TGF-β1 protein was present within the nuclei or the perinuclear area of vascular cells (Figure 1E, right bottom panel, star), suggesting that the de novo synthesis of cellular TGF-β1 protein within the aged arterial wall was increased.5 Because the internal elastin lamina has a potent autofluorescence trait, the continuous sections of aorta were also stained with an LTBP antibody and detected by diaminobenzidine color reaction, confirming that the fluorescence signal was derived from the stained targets rather than from the internal elastin lamina (supplemental Figure I, available online at http://atvb.ahajournals.org).
Colocalization of TGF-β1 and Endothelial and VSMC Markers Within the Aged Aortic Wall
Double immunofluoresence indicated that TGF-β1 staining was colocalized with that of CD31, a marker for ECs, and with that of α-smooth muscle actin, a marker of VSMCs (supplemental Figure II, right panels). Neither lymphocyte (CD45 staining) nor macrophage (CD68 staining) was detected within the intimal and medial arterial layers with aging (supplemental Figure III, top panels). However, within the adventitia of the same arteries, both lymphocytes and macrophages were occasionally observed (supplemental Figure III, middle panels). As an additional positive control, both lymphocytes and macrophages in liver from the same strain of rats were detected with these antibodies (supplemental Figure III, bottom panels).
Aortic and VSMC TGF-β1 Signaling Increases With Age
Activated TGF-β1 exerts its biological effects by binding its TβRII receptor.3,12 TβRII mRNA, determined by real-time Q-PCR, was increased by 2.6-fold in the aortae of old rats compared with young rats (supplemental Figure IVA and IVB). Western blot analysis showed that TβRII protein also increased by 2.6-fold with aging (supplemental Figure IVC and IVD). Immunofluorescence staining showed that the increase in TβRII was widely distributed throughout all compartments within the wall of the aged aorta (Figure 2A, middle panel).
We next determined whether the increase in active TGF-β1/TβRII within the aged aortic wall results in the activation of the SMAD signaling pathways. Western blot analysis showed that the SMAD proteins, phosphorylated p-SMAD2/3 and SMAD4, were increased by 1.9- and 1.4-fold, respectively, within the aged aortic wall, whereas the level of the inhibitory SMAD7 protein decreased by 20% with age within the arterial media (supplemental Figure IVE and IVF).
Immunofluorescence staining showed increased SMAD2, SMAD3, and SMAD4 protein levels (red color) with aging in the rat aortae (Figure 2B). Conversely, SMAD7 staining (green color) was decreased in the aortic media of old rats but did not significantly differ with age within the intima (Figure 2B, bottom panels).
Because most of the intimal cells within normal arterial walls are VSMCs, and those play a key role in the course of arterial aging,7,16 we determined whether SMAD signaling is also increased in VSMCs from young and old rat aorta with aging. Both p-SMAD2/3 and SMAD4 protein staining were located in the perinuclear and nuclear areas of VSMCs and were increased within VSMCs from old rats (Figure 2C, bottom panels, left and middle). However, the density of SMAD7 staining within old VSMCs was slightly decreased compared with that of young VSMCs (Figure 2C, bottom right panel).
Interactions of TGF-β1 and MMP-2 In Vivo and In Vitro
We previously demonstrated via in situ zymography that activated MMP-2 is increased within the aged rat and nonhuman primate aorta.7,9,16 Figure 3A showed that in situ MMP-2 activation increase within the aged rat arterial wall. To determine whether TGF-β1 and activated MMP-2 colocalize within the aged arterial wall, we performed in situ MMP-2 zymography followed by immunofluorescence staining for TGF-β1 (Figure 3B). Figure 3B suggested a close association between activated MMP-2 (green color) and TGF-β1 aortic staining (red color), predominantly in the thickened intima.
Next, we determined whether MMP-2/TGF-β1 associations also exist within VSMCs with aging. Figure 3C showed that MMP-2 and TGF-β1 protein staining was increased within old VSMCs compared with young. Notably, the overlapping area of MMP-2 and TGF-β1 staining (yellow color) was substantially increased within the old VSMCs (Figure 3C, right bottom panels).
To validate the physical interaction between TGF-β1 and MMP-2 in vivo within the aortic wall with aging, the lysate from rat aorta was immunoprecipitated with an antibody against TGF-β1, followed by Western blotting for MMP-2. This analysis showed that MMP-2 physically interacts with TGF-β1 in the aortic wall from both young and old rats (Figure 3D). Interestingly, aging increased the interaction between MMP-2 and TGF-β1 by 1.6-fold (P<0.05; Figure 3D, right panel).
To demonstrate the functional interaction between TGF-β1 and MMP-2, an in vitro activation assay of LAP-TGF-β1 by MMP-2 was performed. The LAP–TGF-β1 (5 ng/mL) was significantly converted by MMP-2 (100 ng/mL) into the active TGF-β1. TGF-β1 activation was inhibited up to 95% by recombinant human tissue type inhibitor of MMP-2 (rhTIMP-2; 500 ng/mL), a specific tissue inhibitor (Figure 3E).
MMP-2 Increases TGF-β1 Activity in Aortic Rings and VSMCs from Young Rats Simulating Untreated Counterparts From Old Rats
Based on the physical interaction of activated MMP-2 and TGF-β1 within the aortic wall, and on known MMP-2 effects on hemodynamics in vivo17 and TGF-β1 activation in vitro, we hypothesized that MMP-2 per se could play a role in the activation of arterial latent TGF-β1. Therefore, we exposed explanted aortic rings from young rats, in absence of intramural pressure and flow effects, to exogenous MMP-2 and compared TGF-β1 protein production with that of untreated rings from young and old rats (Figure 4A). Interestingly, an exposure of aortic rings from young rats to MMP-2 at a concentration of 100 ng/mL increased ring-associated MMP-2 activity, assessed by gelatin zymography, up to the level of old untreated rings (supplemental Figure V). Densitometric analysis of Western blots showed that an exposure to MMP-2 for 72 hours at different doses increased active TGF-β1 protein expression in young aortic rings by a maximum of 7.4-fold (100 ng/mL), similar to the level of untreated aortic rings from old rats (Figure 4A, top right panels). In addition, the increase in production of activated TGF-β1 was time dependent, evident at 24 hours, and peaking at 72 hours of MMP-2 exposure (100 ng/mL; Figure 4A, bottom right panel).
To confirm the role of activated MMP-2 in the activation of TGF-β1 within the aged aortic rings, we exposed them to an MMP inhibitor, GM6001, or to an rhTIMP-2. Figure 4B showed that GM6001 treatment markedly inhibited MMP-2 activity from old rat aortic rings (top left panel) and, in parallel, also reduced TGF-β1 production from old rings to &50% of control in a dose-dependent manner (left, middle, and bottom panels). Figure 4B, right panel, shows that rhTIMP-2, at low doses (100 and 500 ng/mL), slightly increased MMP-2 activation and markedly enhanced TGF-β1 activity; however, at a high dose (1000 ng/mL), it reduced MMP-2 and correspondingly significantly decreased activated TGF-β1 (94.3±9.2 versus 35.3±0.2 OD; P<0.05).
Next, we also determined whether MMP-2 per se also plays a causal role in the activation of latent TGF-β1 in VSMCs as it did in aortic rings. Western blot analysis showed that TGF-β1 increased by 3-fold in old compared with young VSMCs (Figure 4C). Interestingly, the exposure of young VSMCs to MMP-2 (100 ng/mL) markedly increased the activated TGF-β1 up to the level of old VSMCs (Figure 4C). This effect was substantially reduced by rhTIMP-2 or GM 6001. Similar to the ex vivo study, the MMP-2 and TGF-β1 activity from old VSMCs was substantially reduced by GM6001 but not by rhTIMP2 (500 ng/mL; Figure 4C). However, rhTIMP2 at a higher concentration (1000 ng/mL) substantially reduced activated TGF-β1 production in old VSMCs (0.56±0.16 versus 0.17±0.09 arbitrary units; P<0.01), a result very similar to that in aortic rings (Figure 4B).
MMP-2 Activates SMAD Signaling and the Production of Collagen and FN in Aortic Rings and VSMCs
Next, we determined whether, as in the aortic wall in vivo, increased TGF-β1 activation in aortic rings ex vivo was accompanied by an activation of SMAD signaling and downstream production of collagen and FN expression. Western blot analysis demonstrated that, as in the aorta in vivo, activated SMAD2/3 protein (phosphorylated form, p-SMAD2/3) was increased in the untreated rings from old aortae compared with rings from young aortae (supplemental Figure VIA). Immunoblotting showed that in old versus young untreated aortic rings, collagen types I and III and FN protein expression increased 1.7-, 2.7-, and 4.6-fold, respectively (supplemental Figure VIA). In young aortic rings treated with MMP-2 for 72 hours, the p-SMAD2/3 protein also increased in a dose-dependent manner, achieving that level of old untreated arterial rings (supplemental Figure VIA). This was accompanied by dose-dependent increases in collagen types I and III and FN protein production of 42%, 57%, and 80% above control, respectively (supplemental Figure VIA). In addition, we also examined the effects of rhTIMP-2 on the FN and collagen production within the aged aortic rings. Low-dose rhTIMP-2 increased collagen and FN production, but high-dose rhTIMP-2 (1000 ng/mL) substantially reduced the production of these matrices within the old aortic rings (supplemental Figure VIA, right panels).
We also detected whether increased TGF-β1 activation in VSMCs in vitro was accompanied by an increase of downstream collagen and FN production. Western blot analysis showed the exposure of young VSMCs to MMP-2 (100 ng/mL) substantially increased collagen type I by 2.6-fold and collagen type III by 2.8-fold compared with untreated young VSMCs, and this effect was reduced by GM6001 or rhTIMP2 (500 ng/mL; supplemental Figure VIB). Similar to the ex vivo study, the collagen and FN production from old VSMCs was substantially reduced by GM6001 but not by rhTIMP2 (500 ng/mL; supplemental Figure VIB).
In addition, because a plasmin inhibitor, plasminogen activator inhibitor-1 (PAI-1), is known to decrease within the aortic media with aging9 and plasmin is an activator of MMP-2,19 we here analyzed whether PAI-1 plays an inhibitory role in MMP-2–dependent TGF-β1 activation, contributing to age-associated arterial fibrosis. In the present study, Western blots further showed that PAI-1 protein was decreased &4-fold within old compared with young aortae (P<0.05; supplemental Figure VIIA) and also showed that the exposure of young VSMCs to PAI-1 reduced the activated MMP-2 and TGF-β1 in a dose-dependent manner (supplemental Figure VIIB).
The present study, for the first time, demonstrates that all the components of the TGF-β1 complex, LTBP, LAP, and active TGF-β1 are increased within ECs, VSMCs, and matrix of the aged aorta. In addition, mRNA for TGF-β1 and TβRII also increase with aging, suggesting a parallel increase in the transcription of LAP, mature TGF-β1, and TβRII. Additional novel findings of our study are that activated MMP-2 within the aged arterial wall or VSMCs in situ colocalizes with TGF-β1, that MMP-2 is involved in TGF-β1 activation, and that the consequent activation of TβRII via SMAD signaling is linked to increased FN and collagen expression, markers of age-associated arterial remodeling.
Although MMP-2 activity in the arterial wall in rodents and nonhuman primates has been demonstrated to increase with aging,7–9,16 its functional effects are not fully understood. In the present study, we show that a physical association of MMP-2 with TGF-β1 is present within the aortic wall and is enhanced with aging. Our result showing that MMP-2 activates aortic TGF-β1 is consistent with previous observations in several cell types in vitro. In osteoblasts, LAP is cleaved by MMP-2 to release activated TGF-β1.14 Similarly, MMP-2 also activates TGF-β1 in TA3 murine mammary carcinoma cells.18 Moreover, MT1-MMP, an MMP-2 activator that we have shown previously to increase within the arterial wall with aging,8,9,16 induces TGF-β1 activation in osteocytes, whereas an MMP inhibitor, GM6001, blocks TGF-β1 activity in the osteoblast.14 Interestingly, previous studies have shown in vitro that tPA and uPA, which convert plasminogen to plasmin, another MMP-2 activator, both increase in the arterial wall with aging.9 These findings raise the possibility that TGF-β1 activation by plasmin in airway VSMCs in vitro6 may be via plasmin-induced activation of MMP-2.19 Indeed, the present results demonstrate, for the first time, that an exposure of old VSMCs to the PAI-1, a physiological inhibitor of uPA or tPA, significantly inhibits MMP-2 and TGF-β1 activation. These novel findings further support this concept.
The functional interaction of MMP-2 with TGF-β1 affects age-associated matrix remodeling in the aortic rings or in VSMCs. MMP-2 activation of TGF-β1, leading to enhanced TGF-β1 signaling, is an important mechanism for the arterial matrix remodeling and stiffening that accompanies aging.1–3 Treatment of young rings or VSMCs with MMP-2 increases activated TGF-β1 production up to the level of old. Notably, these effects are abolished by GM6001 or rhTIMP-2. Interestingly the MMP inhibitor GM6001 reverses the age-associated increase in activated TGF-β1 levels in old rat aortic rings ex vivo and VSMCs in vitro. The addition of rhTIMP2 reproduces most of the findings of GM6001. These findings suggest that MMP-2 is an important but not necessarily an exclusive activator of latent TGF-β1 within the old arterial wall or VSMCs in rats. Interestingly, a lower concentration of rhTIMP (500 ng/mL) did not inhibit TGF-β1 activity; but at a high concentration (1000 ng/mL), rhTIMP-2 substantially reduced TGF-β1 activation within old arterial rings or VSMCs. Several factors may contribute to the inconsistency of the rhTIMP2 and GM6001 effect in the activation of latent TGF-β1 in the old artery or VSMCs. First, in vivo TIMP-2 inhibition of MMP-2 is dependent on a balance of TIMP-2/MT1-MMP.20,21 Second, MT1-MMP, as well as plasmin, is increased within the arterial wall with aging,9 and both have capabilities to activate latent TGF-β1. Third, GM6001 not only inhibits MMP-2 but also MT1-MMP activity.14
Our results also show that MMP-2 activation of TGF-β1 in aortic rings and VSMCs engages the SMAD signaling pathway, resulting in the production of the matrix molecules, FN, and collagen. After MMP-2 incubation of aortic rings or VSMCs from young rats, the levels of collagen and FN deposition are quantitatively similar to those that are present within rings of old untreated rats. The age-associated increase in aortic FN and collagen in vivo likely results from the activation of SMAD signaling with age. In particular, in addition to active TGF-β1, TβRII, SMAD2/3, and SMAD4 are also increased within the arterial wall and VSMCs of older rats. In contrast, the inhibitor SMAD7 decreases with age, particularly in the arterial media.
Data from present and previous studies from our group point at a central role for angiotensin II (Ang II) and associated signaling molecules in arterial aging.2 TGF-β1, MMP-2, monocyte chemoattractant protein-1 (MCP-1), collagen, and FN production are all downstream events of Ang II–Ang II type 1 (AT1) receptor signaling.8,22 The age-associated increases of these related molecules within the remodeled aorta of older rats implicate local arterial Ang II signaling.8,16 This idea is consistent with previous studies indicating that angiotensinogen expression and AT1 are both upregulated in the arterial wall of aged rats.23,24 The chronic infusion of Ang II increases arterial MCP-1 transcription and translation, both of which are upregulated within the aged aortic wall.25 Moreover, in older nonhuman primates and rodents, arterial wall, angiotensin-converting enzyme, and Ang II are increased.8,16 Furthermore, in rodents, chronic angiotensin-converting enzyme inhibition dramatically retards multiple features of age-associated aortic remodeling.26 Chronic infusion of Ang II to young rats, on the other hand, induces MMP-2 activation via AT1 signaling and generates molecular and structural alterations similar to arterial aging, such as collagen deposition and intima and media thickening.8
In summary, the present study demonstrates that aortic TGF-β1 expression and TβRII increase with aging; TGF-β1 activation is also increased and is dependent on the concomitant age-associated increase in MMP-2 activity; TGF-β1 signaling via the TβRII activated SMAD pathway leads to an increase in FN and collagen expression. Thus, the age-associated increase in MMP-2 activation not only regulates proteolysis of matrix proteins, such as collagen and elastin, favorably enhancing invasion potential of VSMCs within the aged arterial wall,8 but it also regulates the activity of the growth factor TGF-β1, which promotes fibrosis. Both enhanced TGF-β1 and MMP-2 expression and activation and downstream events of Ang II signaling (ie, increased MCP-1 within the arterial wall remodeled by aging) point to Ang II signaling pathways as targets for prevention or amelioration of arterial remodeling and its attendant risk for the incidence and severity of arterial diseases that increase exponentially with advancing age.1,2
Sources of Funding
This research was supported by the Intramural Research Program of the National Institutes of Health, National Institute on Aging.
M.W. and D.Z. contributed equally to this study.
Original received October 12, 2005; final version accepted April 21, 2006.
Lakatta EG. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: part III: cellular and molecular clues to heart and arterial aging. Circulation. 2003; 107: 490–497.
Wang M, Lakatta EG. Chapter 9. Central arterial aging: humans to molecules. Hypertension in the elderly. In Safar ME, O’Rourke, ME, eds. Handbook of Hypertension. The Netherlands: Elsevier; 2006: 137–160.
Leask A, Abraham DJ. TGF-beta signaling and the fibrotic response. FASEB J. 2004; 18: 816–827.
Shivakumar K, Dostal DE, Boheler K, Baker KM, Lakatta EG. Differential response of cardiac fibroblasts from young adult and senescent rats to Ang II. Am J Physiol Heart Circ Physiol. 2003; 284: H1454–H1459.
Coutts A, Chen G, Stephens N, Hirst S, Douglas D, Eichholtz T, Khalil N. Release of biologically active TGF-beta from airway smooth muscle cells induces autocrine synthesis of collagen. Am J Physiol Lung Cell Mol Physiol. 2001; 280: L999–L1008.
Li Z, Froehlich J, Galis ZS, Lakatta EG. Increased expression of matrix metalloproteinase-2 in the thickened intima of aged rats. Hypertension. 1999; 33: 116–123.
Wang M, Lakatta EG. Altered regulation of matrix metalloproteinase-2 in aortic remodeling during aging. Hypertension. 2002; 39: 865–873.
Kass DA, Shapiro EP, Kawaguchi M, Capriotti AR, Scuteri A, deGroof RC, Lakatta EG. Improved arterial compliance by a novel advanced glycation end-product crosslink breaker. Circulation. 2001; 104: 1464–1470.
Sinha S, Heagerty AM, Shuttleworth CA, Kielty CM. Expression of latent TGF-beta binding proteins and association with TGF-beta 1 and fibrillin-1 following arterial injury. Cardiovasc Res. 2002; 53: 971–983.
Karsdal MA, Larsen L, Engsig MT, Lou H, Ferreras M, Lochter A, Delaisse JM, Foged NT. Matrix metalloproteinase-dependent activation of latent transforming growth factor-beta controls the conversion of osteoblasts into osteocytes by blocking osteoblast apoptosis. J Biol Chem. 2002; 277: 44061–44067.
Wang M, Takagi G, Asai K, Resuello RG, Natividad FF, Vatner DE, Vatner SF, Lakatta EG. Aging increases aortic MMP-2 activity and angiotensin II in nonhuman primates. Hypertension. 2003; 41: 1308–1316.
Yu Q, Stamenkovic I. Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev. 2000; 14: 163–176.
Spinetti G, Wang M, Monticone R, Zhang J, Zhao D, Lakatta EG. Rat aortic MCP-1 and its receptor CCR2 increase with age and alter vascular smooth muscle cell function. Arterioscler Thromb Vasc Biol. 2004; 24: 1397–1402.
Eggena P, Morin AM, Barrett JD, Krall JF. The influence of aging on angiotensinogen production by rat vascular smooth muscle cells in vitro. Clin Exp Hypertens. 1988; 10: 597–603.
Touyz RM, Endemann D, He G, Li JS, Schiffrin EL. Role of AT2 receptors in angiotensin II-stimulated contraction of small mesenteric arteries in young SHR. Hypertension. 1999; 33: 366–372.
Ni W, Kitamoto S, Ishibashi M, Usui M, Inoue S, Hiasa KI, Zhao Q, Nishida KI, Takeshita A, Egashira K. Monocyte chemoattractant protein-1 is an essential inflammatory mediator in angiotensin II-induced progression of established atherosclerosis in hypercholesterolemic mice. Arterioscler Thromb Vasc Biol. 2004; 24: 534–539.