Pioglitazone Improves Aortic Wall Elasticity in a Rat Model of Elastocalcinotic Arteriosclerosis
Specific treatment of age-related aortic wall arteriosclerosis and stiffening is lacking. Because ligands for peroxisome proliferator-activated receptor γ have beneficial effects on the arterial wall in atherosclerosis, via an antiinflammatory mechanism, we investigated whether long-term pioglitazone (Pio) treatment protects against another form of vascular wall disease, arteriosclerosis. We evaluated, in a rat model of elastocalcinotic arteriosclerosis (hypervitaminosis D and nicotine [VDN]), whether Pio (3 mg · kg−1 per day for 1.5 month PO) attenuated arteriosclerosis and its consequences: aortic wall rigidity, increased aortic pulse pressure, and left ventricular hypertrophy. In VDN rats, medial calcification was associated with monocyte/macrophage infiltration and induction of tumor necrosis factor α and interleukin 1β. Pio increased nuclear peroxisome proliferator-activated receptor γ immunostaining in the aortic wall, decreased tumor necrosis factor α (P <0.05 versus VDN Pio−), tended to decrease interleukin 1β mRNA expression (P =0.08 versus VDN Pio−), blunted aortic wall calcification (271±69, P <0.05 versus VDN Pio− 562±87 μmol · g−1 dry weight) and prevented fragmentation of elastic fibers (segments per 10 000 μm2: 8.4±0.3; P <0.05 versus VDN Pio− 10.5±0.6). Pio reduced aortic wall stiffness (elastic modulus/wall stress: 4.8±0.6; P <0.05 versus VDN Pio− 10.0±1.6), aortic pulse pressure (30±2 mm Hg; P <0.05 versus VDN Pio− 39±4) and left ventricular hypertrophy (1.58±0.05 g · kg−1; P <0.05 versus VDN Pio− 1.76±0.06). In conclusion, long-term Pio treatment attenuates aortic wall elastocalcinosis and, thus, lowers aortic wall stiffness, aortic pulse pressure, and left ventricular hypertrophy.
An age-related increase in aortic wall stiffness is an independent risk factor for cardiovascular morbidity and mortality.1–3 There is no specific treatment, thus the research for drug targets is of great importance. Aortic wall stiffening arises principally from extracellular matrix remodeling of the media. Calcification and fragmentation of the elastic fiber network,4,5 and nonenzymatic cross-linking (glycation)6,7 of collagen fibers, are the most important determinants.
Several of these molecular mechanisms (calcification, glycation) have an inflammatory element. Thus, we investigated a new antiinflammatory pharmacological target, the peroxisome proliferator-activated receptor (PPAR)-γ, on the basis that thiazolidinediones (TZDs) specific PPAR-γ ligands have beneficial protective effects on the arterial wall in atherosclerosis8,9 and arteriosclerosis,10 via antiinflammatory mechanisms.
The beneficial antiproliferative and antifibrotic effects of pioglitazone (Pio) were demonstrated using a rat model of inflammation-induced wall fibrosis (long-term inhibition of endothelial NO synthesis).10 In the present study, we focused on another determinant of extracellular matrix remodeling: medial elastocalcinosis. We used an animal model of calcification and degradation of elastic fibers, the rat treated with vitamin D3 and nicotine (VDN).11–13 To our knowledge, this is the only model showing extensive aortic medial calcification, in the absence of fibrosis or any change in wall stress. In VDN rats, extracellular calcium-binding proteins and ectopic apatite deposition develop on medial elastic fibers, followed by fragmentation of the latter (elastocalcinosis).11,12 This causes wall stiffening, then elevation of pulse arterial pressure (with no change in mean arterial pressure) and compensatory left ventricular hypertrophy.12,13
Our working hypotheses were that (1) VDN-induced calcification and elastic fiber fragmentation represent, at least partially, an inflammatory response of the aortic wall; and (2) long-term treatment of VDN rats with Pio10 (Pio), by counteracting the inflammatory process,8–10 reduces calcification and elastic fiber fragmentation, therefore leading to a decrease in wall elastic modulus and the consequences of increased aortic wall rigidity, namely, increased pulse pressure and left ventricular hypertrophy. In a separate experiment, we checked whether the beneficial effect of Pio was attributable to a short-term, functional, or long-term structural effect on the aortic wall.
Animals and Drug Treatment
Seven-week-old male Wistar rats (Ico: Wi, IOPS AF/Han, Charles River Laboratories, L’Arbresle, France) were kept under standard conditions for 1 week before VDN treatment.
For the long-term experiment, VDN rats (n=26) were treated on day 0 (D0) with vitamin D3 (300 000 IU · kg−1 IM) and nicotine (2×25 mg · kg−1, 5 mL · kg−1 PO)11–13; non-VDN rats (n=22) were treated with 0.15 mol/L NaCl (IM) and distilled water (PO). From D1 to D44, rats received Pio (3 mg · kg−1 per day, 6 mL · kg−1; n=17 VDN Pio+ and 13 non-VDN Pio+ rats) or carboxymethylcellulose (0.5%, 6 mL · kg−1; n=9 VDN Pio− and 9 non-VDN Pio− rats) PO at 12:00 pm. We used the same treatment schedule as Ishibashi et al, who reported that 3 mg · kg−1 per day of Pio had beneficial antiproliferative and antifibrotic effects on the coronary artery wall in a rat model of inflammation-induced wall fibrosis.10 Hemodynamic parameters were measured on D45 (24-hour drug washout) under sodium pentobarbital anesthesia (60 mg · kg−1 IP). Experiments followed the guidelines of the French Ministry of Agriculture, Paris (permit numbers IL: 54-5 and JA: 03575).
In a separate experiment, 12 VDN and 10 non-VDN rats received on D45, 1 gavage of Pio (3 mg · kg−1) or carboxymethylcellulose at 12:00 pm. Hemodynamic parameters were measured 2 hours later (time of the peak in the plasma concentration of Pio, high-performance liquid chromatography/UV, results not shown).
Aortic Blood Pressure and Pulse Wave Velocity
On D45, polyethylene cannula, connected to low-volume pressure transducers, were introduced into the descending thoracic and abdominal aorta for measurement of baseline central and peripheral blood pressures.6,12–15 An algorithm detected systolic and diastolic pressures, calculated mean pressure (waveform area), pulse pressure (systolic−diastolic), heart rate and thoraco-abdominal pulse wave transit time.6,12–15 Pulse wave velocity (PWV) was calculated as the distance between the 2 cannula tips (measured in situ following euthanasia) divided by the pulse wave transit time.
The impact of sodium pentobarbital on hemodynamic values must be considered because a decrease in sympathetic nervous system activity (1) lowers aortic smooth muscle cell tone, which could modify elasticity per se (however, we previously showed that a change in local smooth muscle cell tone is not an important determinant of aortic wall elasticity14); (2) may lower blood pressure and thus wall stress (however, in the present study, mean aortic blood pressure was similar in non-VDN and VDN rats [as we reported for awake animals13]); and (3) would lower heart rate and thus change the harmonic composition of the pulse pressure signal. This would alter PWV measured, as it did here (by the foot-to-foot method). However, changes in heart rate were relatively minor (see Results).
Plasma Glucose Concentration, Left Ventricular Weight, Thoracic Aorta Geometry and Mechanics, and Wall Elastic Network
One milliliter of arterial blood was collected for determination of plasma glucose (glucose oxidase method). After euthanasia (sodium pentobarbital overdose), the left ventricle, including septum, was dissected free and weighed; left ventricular hypertrophy was estimated as left ventricle/body weight. The apex of the ventricle (150 mg) was dehydrated at 110°C to determine the percentage of dry weight.
The first centimeter of the descending thoracic aorta (location in study by Niederhoffer et al12) was removed, fixed in formaldehyde (10% in phosphate buffered saline), and embedded in paraffin for histomorphometric analyses and evaluation of monocyte/macrophage infiltration (see above). All tests were performed by 2 independent, blinded observers. The internal diameter (Di) and medial thickness (h) were measured on 20-μm thick sections stained with hematoxylin-eosin (Saisam algorithm; Microvision Instruments, Evry, France).
Elastic modulus (EM=PWV2 · Di · ρ/h; Moens-Korteweg) and wall stress (WS=central aortic mean blood pressure · Di/2 h; Lamé) were calculated with ρ=blood density, 1.05 g · cm−3. EM/WS was used as an isobaric index of intrinsic aortic wall stiffness.6,15
Fragmentation of the medial elastic fiber network (excluding the external and internal laminae) was evaluated on 10-μm thick sections stained with Weigert solution by measuring the increase in the number15 and the decrease in the length of elastic lamellae.
Monocyte/Macrophage Infiltration in the Aortic Wall, Interferon γ, Tumor Necrosis Factor α, and Interleukin 1β mRNA Expression
Aortic calcium was visualized as black deposits on deparaffinized/rehydrated sections using the Von Kossa stain.11,16 Sections were then antigen-retrieved in citrate buffer (pH 6, 40 minutes, 97°C then 20 minutes, room temperature). Monocyte/macrophages were immunostained by overnight exposure (4°C) to a monoclonal mouse anti-rat ED-1 antibody (dilution 1:500; mca341r, Serotec, UK). The antibody was revealed as red deposits using an indirect streptavidin-biotin method with H2O 2/3-amino-9-ethylcarbazole as chromogen (DAKO Chemmate Detection Kit; peroxidase).
Total mRNA was extracted from the second centimeter12 of the descending thoracic aorta (6 to 9 rats per group) using Tri-Reagent (Euromedex, Souffelweyersheim, France) and RNEASY Mini Kit (Qiagen, Courtaboeuf, France). Total mRNA concentration was measured, its integrity was confirmed, and reverse transcription-polymerase chain reaction (PCR) and PCR were performed as previously described17,18 (using the housekeeping gene L27: amplimers 5′-TCCTGGCTGGACGCTACTC-3′, sense; and 5′-CCACAGAGTACCTTGTGGGC-3′, antisense; annealing temperature 62°C, PCR product 225 bp). The PCR products (separated on a 2% agarose gel containing 0.5 μg · mL−1 ethidium bromide) were quantified as density ratio (PCR product of interest/L27) by densitometry (Bio-Rad Gel Doc 1000; Bio-Rad Laboratories).
Thoracic Aorta Wall Composition
The third centimeter12 of the descending thoracic aorta was hydrolyzed in hydrochloric acid (6 mol/L, 24 hour, 105°C) and protein (molecular weight of an amino acid, 92; dinitrofluorobenzene reaction)19 and collagen (hydroxyproline content, 7.46×; chloramine T and paradimethylaminobenzaldehyde reaction)20 contents determined.
The fourth centimeter12 of the descending thoracic aorta was dehydrated at 110°C to determine the percentage of dry weight and total wall calcium content (atomic absorption spectrophotometry following mineralization and nitric acid digestion).11
Quantification and Localization of PPAR-γ in the Aortic Wall
The second centimeter12 of the descending thoracic aorta was used for Western blot analysis. Each aorta was analyzed in triplicate as previously described,21 using a polyclonal rabbit PPAR-γ anti-serum22 (dilution 1:1500) then luminol (Mouse/Rabbit chemoluminescence detection kit; Roche) and quantified by densitometry (GelDoc; Bio-Rad).
The ascending aorta (embedded in freezing medium) was used for immunostaining (3-μm thick sections; polyclonal rabbit PPAR-γ anti-serum22 [dilution 1:50]; fluorescein-conjugated goat anti-rabbit IgG [1:50]; UV microscopy).22 The analysis was performed on 3 to 4 rats per group, 3 aortic sections per rat, focusing on 30 to 40 medial smooth muscle cells per section, to evaluate the percentage of smooth muscle cells showing nuclear PPAR-γ staining.
Parametric values (means±SEM) were analyzed by a 2-way ANOVA (PVDN, PPio, PVDN×Pio) plus the Bonferroni post hoc test and nonparametric values (immunostaining) by the Mann-Whitney U test. Linear regressions were performed using standard ANOVA techniques and results expressed as slopes and intercepts.
Vitamin D3 (Duphafral D3) was purchased from Duphar B.V. (Weesp, The Netherlands), sodium pentobarbital from Sanofi Santé Nutrition Animale (Libourne, France), and other chemicals from Sigma Chemical Co (St Louis, Mo). Pio was a gift of Takeda Chemicals Industries Ltd (Osaka, Japan).
PPAR-γ, Monocyte/Macrophage Infiltration, Cytokine Expression, Elastic Network, Composition of the Aortic Wall, and Thoracic Aorta Geometry
VDN treatment increased PPAR-γ protein content (1.6×), but long-term treatment with Pio had no effect (Western blot, Figure 1). Medial smooth muscle cells presented positive cytoplasmic immunostaining for PPAR-γ, with no difference between non-VDN and VDN. Pio+ increased nuclear PPAR-γ staining to 95±2% (non-VDN Pio++VDN Pio+ rats, P <0.05 versus 65±10% in non-VDN Pio−+ VDN Pio− rats).
The calcium content of the aortic wall increased 43× in VDN rats (Figure 2); Pio+ partially prevented calcification (VDN Pio+: 21× versus non-VDN rats). In VDN rats, the media exhibited heavy calcium deposits along the elastic lamellae and monocyte/macrophage infiltration between calcified lamellae (Figure 2). In VDN Pio+, the occurrence of calcified lesions along the aortic segment was lower, but, when present, lesions were similar in nature to those observed in VDN Pio− and showed monocyte/macrophage infiltration.
Interferon γ mRNA content did not change in VDN rats (results not shown), but tumor necrosis factor (TNF)-α and interleukin (IL)-1β mRNA expression increased in parallel with the degree of calcification (Figure 3). There was a significant positive relationship between IL-1β expression and calcium content of the aortic wall in VDN Pio− rats (slope (×10−4)=9.8±1.7 μmol−1 · g; r =0.9035; P <0.05; intercept=0.25±0.11; n=9) but not between TNF-α and calcium content. Pio+ decreased TNF-α (P <0.05 versus VDN Pio−) and tended to decrease IL-1β (P =0.08) expression by 38% and 25%, respectively. In VDN Pio+, the correlation between IL-1β and calcium content of the wall was not statistically significant.
The medial elastic network showed severe fragmentation and disorganization in VDN rats (Figure 4), with an increase in the number (+31%) and a decrease in the length (−36%) of lamellae (Table 1). Pio+ blunted elastic network fragmentation in VDN rats.
The dry weight and collagen content of the aortic wall were similar in all groups (Table 1). Protein content decreased in VDN rats (−17%); Pio+ had no effect.
Internal diameter increased by 22% (P <0.05) in VDN rats, with no change in medial thickness (Table 1). Chronic Pio treatment prevented dilatation in VDN rats but had no effect on medial thickness.
Aortic Blood Pressure and Wall Mechanics
PWV (×1.5, Table 2) and EM/WS (×2.8, Figure 5) increased in VDN rats, with no change in wall stress (Table 2). Pio+ decreased PWV and EM/WS in VDN rats only (PVDN×Pio<0.05) to values that were not significantly different from those of non-VDN Pio− rats (PWV, P =0.150; EM/WS, P =0.189).
There was a significant positive relationship between EM/WS and calcium content of the aortic wall in VDN rats (slope=0.0074±0.0015 μmol−1 · g; r =0.719; P <0.05; intercept=3.44±0.77; n=26) but not in non-VDN rats. The linear regression slope for VDN Pio+ rats (0.0061±0.0017 μmol−1 · g; r =0.697; P <0.05; intercept=3.65±0.73; n=17) was lower (P =0.0018) than that for VDN Pio− rats (0.0093±0.0041 μmol−1 · g; r =0.678; P =0.05; intercept=2.69±2.59; n=9).
Central aortic mean blood pressure was similar in all groups (Table 2); pulse pressure increased by 56% in VDN rats (Figure 4). Pio lowered aortic pulse pressure (Figure 4) to a value not significantly different from that of non-VDN Pio− rats (P =0.083).
In the short-term experiment, VDN Pio− rats showed calcification and stiffening (calcium content 46×, PWV 1.5× non-VDN) of the aortic wall; pulse pressure increased by 61% versus non-VDN, with no change in mean aortic blood pressure or heart rate. Acute Pio administration had no effect on aortic wall calcium content, PWV or pulse pressure; it slightly increased mean aortic pressure (+4%) and decreased heart rate (−10%) both in VDN and non-VDN rats (P <0.05).
Body Weight, Plasma Glucose Concentration, and Cardiac Parameters
VDN rats lost weight between D0 and D6 (−18%, results not shown), then recovered normal growth. At D45, body weight was 5% lower in VDN than in non-VDN rats (Table 3). Pio+ increased body weight in both VDN and non-VDN rats (+10%). Plasma glucose concentrations were similar in all groups (Table 3).
Heart rate, left ventricle weight, and the percentage of myocardium dry weight were similar in all groups (Table 3). Left ventricle/body weight increased by 16% in VDN rats (Figure 5); Pio lowered this ratio to a value not significantly different from that of non-VDN Pio− rats (P =0.150).
VDN treatment reveals an inflammatory response of the aortic wall (monocyte/macrophage infiltration and induction of proinflammatory cytokines) that accompanies calcification associated with medial elastic fiber fragmentation and wall stiffening. Treatment of VDN rats with Pio, which purportedly acts as an antiinflammatory agent in the arterial wall,8–10,23 blunts elastocalcinosis and its structural and hemodynamic consequences. Whether this is attributable to an antiinflammatory effect is discussed below.
The beneficial effect of Pio on vascular structure and hemodynamics was attributable not to a short-term but to a long-term effect; with short-term administration, Pio had no effect on aortic wall calcium content, PWV, or pulse pressure.
The anticalcinotic effect of Pio in VDN is more likely to be caused by activation of PPAR-γ rather than to a change in PPAR-γ expression. Pio treatment did not induce any change in the PPAR-γ protein content in VDN, and the decrease in aortic wall calcium content did not correlate with changes in PPAR-γ levels. VDN treatment itself causes a significant increase in PPAR-γ (30% to 40%). Thus, although Pio tended to increase PPAR-γ in non-VDN rats, it does not produce any further significant increase in VDN rats. This does not exclude that nuclear activation of PPAR-γ by Pio may induce expression of target genes involved in the anticalcinotic effect. Another possibility is that PPAR-γ expression increases at an early stage of Pio treatment in VDN rats (and produces an anticalcinotic effect) and that later (D45) the anticalcinotic effect remains, but the difference in PPAR-γ expression is masked.
An antiinflammatory action of Pio (witnessed by the decreased expression in TNF-α and IL-1β) is suggested in our VDN model but may be less important or more subtle than in other articles.8–10,23 In VDN rats, aortic wall calcification was strongly associated with monocyte/macrophage infiltration and increased expression of TNF-α and IL-1β; IFN-γ mRNA, which is produced by lymphocytes and/or natural killer cells, did not change. Pio did not modify the strong association between calcium deposits and monocyte/macrophage infiltration because, when present, aortic calcified lesions were similar in nature to those observed in VDN Pio−; however, the frequency of calcified lesions along the aortic segment was lower in VDN Pio−. Moreover, the lack of correlation between IL-1β mRNA and calcium contents in VDN Pio+ suggests that the antiinflammatory effect of Pio is not the only mechanism responsible for the decrease in aortic wall calcification in the VDN model.
Other hypotheses can be evoked. Pio may attenuate the recruitment of cells producing proteins involved in arterial wall calcification. Firstly, in the VDN model, arterial wall calcification is initiated by recruitment of S-100+ cells and deposition of extracellular S-100 calcium-binding proteins.12 Secondly, Pio and vitamin D are ligands for nuclear receptors, which may interfere with each other,24 and those modify bone calcium metabolism and at a later stage arterial wall calcification. In VDN, part of the arterial ectopic apatite deposited is of bone origin. The calcium content of the femoral bone decreases during the first 2 weeks following VDN treatment,25 inhibition of osteoclasts prevents arterial calcification in vitamin D-treated rats,26 and TZDs inhibit bone resorption induced by vitamin D.27 Thirdly, part of the calcium deposited on the wall elastic fibers is released from smooth muscle cells following calcium overload-induced necrosis4,28 produced by a direct or indirect (sympathomimetic) action28 of nicotine. Calcium channel blockers such as isradipine reduce aortic wall calcification in the VDN model,4 and TZDs block smooth muscle cell calcium channels.29
Treatment with Pio halved calcium deposition, which remained, none-the-less, elevated (21-fold above non-VDN rats). In spite of this, Pio normalized EM/WS (Figure 4) in VDN rats. This suggests that the reduction of calcification is not the only way by which Pio improves wall elasticity in the VDN model and that calcification is not the only factor responsible for elastic fiber degradation. This is backed up by the observation that the regression analysis of EM/WS versus wall calcium content shows a lower slope for VDN rats treated with Pio. The normalization of EM/WS cannot be explained by a change in aortic geometry because, although Pio does reduce VDN-induced dilatation, it does not modify global wall stress. It is possible that more subtle changes in the distribution of tension throughout the wall (eg, less recruitment of stiff collagen fibers because of the less pronounced enlargement of the aorta) may account for the changes in wall integrity/component and, thus, for the normalization of the elastic properties. However, global changes in wall composition such as fibrosis (Pio has no effect on collagen content) or edema (Pio has no effect on dry weight percentage) are seemingly not involved. It is possible that Pio protects the elastic fiber network from degradation by an antiinflammatory action (decreased expression of proinflammatory cytokines), leading to a decrease in metalloprotease activity.8,30,31 Other potential mechanisms could include antioxidant effects and protection of endothelial function32 or some effect at the level of the angiotensin II type 1 receptor.33 Whatever the mechanism, protection of the elastic fiber network from degradation maintained aortic wall elastic properties and structural integrity, thus preventing aortic aneurysm. It is interesting to note that dilatation is not accompanied by wall thickening in VDN.
Finally, it should be noted that the beneficial Pio effect on pulse pressure is attributable to attenuation of the fall in diastolic pressure in VDN. This would maintain coronary perfusion and may be relevant in the elderly. Pio significantly lowered pulse pressure, and it could be argued that the subsequent reduction in pulsatile wall stress could somehow attenuate the pressure-induced amplification of the physicochemical reaction of elastocalcinosis. There are arguments for and against this. Megnien et al34 showed that hypertension (ie, an increase in both mean and pulse pressures) amplifies arterial calcification. However, we have shown in the VDN model that long-term treatment with the calcium entry blocker, isradipine, lowers mesenteric artery calcification but has no effect on blood pressure (or on aortic calcification).35 This has been discussed previously5 and the question is still unanswered as to whether increased pressure (and, thus, increased wall stress) amplifies wall elastocalcinosis.
In conclusion, we have shown that long-term Pio treatment protects against calcification and degradation of aortic wall elastic fibers via, at least partially, an antiinflammatory mechanism. This improves aortic wall elasticity in a rat model of elastocalcinotic arteriosclerosis, thus preventing elevation of pulse pressure and left ventricular hypertrophy. Our results may be clinically relevant in elderly patients experiencing aortic wall stiffening.
This work was funded by grants from the French Education, Research and Technology Ministry; the Regional Development Committee (Metz, France); the Greater Nancy Urban Council (Nancy, France); the Pharmacolor Association (Nancy, France); the “Fondation de la Recherche Médicale” (Lorraine Committee, France); the “Ligue contre le Cancer” (54 Committee, France); the “Fondation de France,” Paris; and the “Association de la Recherche contre le Cancer” (ARC No. 5637; Paris). We thank Patrick Liminana (Pharmacology Laboratory), Aurélie Beltz (EA3446, Nancy, France), and Annie Artuso (EA3127, Montpellier, France) for help with histomorphometry, Western blot analysis, and ED-1 immunostaining.
- Received October 26, 2004.
- Revision received November 16, 2004.
- Accepted May 11, 2005.
Lehmann ED, Hopkins KD, Rawesh A, Joseph RC, Kongola K, Coppack SW, Gosling RG. Relation between number of cardiovascular risk factor/events and noninvasive Doppler ultrasound assessments of aortic compliance. Hypertension. 1998; 32: 565–569.
Blacher J, Asmar R, Djane S, London GM, Safar ME. Aortic pulse wave velocity as a marker of cardiovascular risk in hypertensive patients. Hypertension. 1999; 33: 1111–1117.
Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aortic pulse wave velocity predicts cardiovascular mortality in subjects >70 years of age. Arterioscler Thromb Vasc Biol. 2001; 21: 2046–2050.
Fleckenstein A, Frey M, Fleckenstein-Grun G. Protection by calcium antagonists against experimental arterial calcinosis. Secondary Prevention of Coronary Heart Disease, Workshop of the International Society and Federation of Cardiology. New York: Pyörälä K, Georg Thieme Verlag; 1983: 109–122.
Cantini C, Kieffer P, Corman B, Liminana P, Atkinson J, Lartaud-Idjouadiene I. Aminoguanidine and aortic wall mechanics, structure, and composition in aged rats. Hypertension. 2001; 38: 943–948.
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.
Law RE, Goetze S, Xi XP, Jackson S, Kawano Y, Demer L, Fishbein MC, Meehan WP, Hsueh WA. Expression and function of PPARgamma in rat and human vascular smooth muscle cells. Circulation. 2000; 101: 1311–1318.
Ishibashi M, Egashira K, Hiasa K, Inoue S, Ni W, Zhao Q, Usui M, Kitamoto S, Ichiki T, Takeshita A. Antiinflammatory and antiarteriosclerotic effects of pioglitazone. Hypertension. 2002; 40: 687–693.
Niederhoffer N, Marque V, Lartaud-Idjouadiene I, Duvivier C, Peslin R, Atkinson J. Vasodilators, aortic elasticity, and ventricular end-systolic stress in nonanesthetized unrestrained rats. Hypertension. 1997; 30: 1169–1174.
Marque V, Kieffer P, Gayraud B, Lartaud-Idjouadiene I, Ramirez F, Atkinson J. Aortic wall mechanics and composition in a transgenic mouse model of Marfan syndrome. Arterioscler Thromb Vasc Biol. 2001; 21: 1184–1189.
Von Kossa J. Uber die im organismus künstlich erzeugboren Verkalkungen. Beitr Path Anat. 1901; 29: 163–202.
Devaux Y, Grosjean S, Seguin C, David C, Dousset B, Zannad F, Meistelman C, De Talance N, Mertes PM, Ungureanu-Longrois D. Retinoic acid and host-pathogen interactions: effects on the inducible nitric oxide synthase in vivo. Am J Physiol. 2000; 279: E1045–E1053.
Seguin-Devaux C, Devaux Y, Latger-Cannard V, Grosjean S, Rochette-Egly C, Meistelman C, Zannad F, Mertes PM, Longrois D. Enhancement of the inducible NO synthase activation by all-trans retinoic acid is mimicked by RARα agonist in vivo. Am J Physiol. 2002; 283: E525–E535.
Ghuysen JM, Strominger JL. Structure of the cell wall of Staphylococcus aureus, strain Copenhagen. I. Preparation of fragments by enzymatic hydrolysis. Biochemistry. 1963; 5: 1110–1119.
Neuman RE, Logan MA. The determination of collagen and elastin in tissues. J Biol Chem. 1950; 184: 549–556.
Huin C, Corriveau L, Bianchi A, Keller JM, Collet P, Kremarik-Bouillaud P, Domenjoud L, Becuwe P, Schohn H, Menard D, Dauca M. Differential expression of peroxisome proliferator-activated receptors (PPARs) in the developing human fetal digestive tract. J Histochem Cytochem. 2000; 48: 603–611.
Brice PA, June HH, Buckley JR, Williamson MK. SB 242784, a selective inhibitor of the osteoclastic V-H+-ATPase, inhibits arterial calcification in the rat. Circ Res. 2002; 91: 547–552.
Marx N, Schonbeck U, Lazar MA, Libby P, Plutzky J. Peroxisome proliferator-activated receptor gamma activators inhibit gene expression and migration in human vascular smooth muscle cells. Circ Res. 1998; 83: 1097–1103.
Megnien JL, Simon A, Lemariey M, Plainfosse MC, Levenson J. Hypertens promotes coronary calcium deposit in asymptomatic men. Hypertension. 1996; 27: 949–954.
Henrion D, Chillon JM, Capdeville-Atkinson C, Atkinson J. Effect of long-term treatment with the calcium entry blocker, isradipine, on vascular calcium overload produced by vitamin D3 and nicotine in rats. J Pharmacol Exp Ther. 1992; 260: 1–8.