Thrombospondin-2 Is Essential for Myocardial Matrix Integrity
Increased Expression Identifies Failure-Prone Cardiac Hypertrophy
Cardiac hypertrophy can lead to heart failure (HF), but it is unpredictable which hypertrophied myocardium will progress to HF. We surmised that apart from hypertrophy-related genes, failure-related genes are expressed before the onset of failure, permitting molecular prediction of HF. Hearts from hypertensive homozygous renin-overexpressing (Ren-2) rats that had progressed to early HF were compared by microarray analysis to Ren-2 rats that had remained compensated. To identify which HF-related genes preceded failure, cardiac biopsy specimens were taken during compensated hypertrophy and we then monitored whether the rat progressed to HF or remained compensated. Among 48 genes overexpressed in failing hearts, we focused on thrombospondin-2 (TSP2). TSP2 was selectively overexpressed only in biopsy specimens from rats that later progressed to HF. Moreover, expression of TSP2 was increased in human hypertrophied hearts with decreased (0.19±0.01) versus normal ejection fraction (0.11±0.03 [arbitrary units]; P<0.05). Angiotensin II induced fatal cardiac rupture in 70% of TSP2 knockout mice, with cardiac failure in the surviving mice; this was not seen in wild-type mice. In TSP2 knockout mice, angiotensin II increased matrix metalloproteinase (MMP)-2 and MMP-9 activity by 120% and 390% compared with wild-type mice (P<0.05). In conclusion, we identify TSP2 as a crucial regulator of the integrity of the cardiac matrix that is necessary for the myocardium to cope with increased loading and that may function by its regulation of MMP activity. This suggests that expression of TSP2 marks an early-stage molecular program that is activated uniquely in hypertrophied hearts that are prone to fail.
Hypertension causes cardiac hypertrophy, one of the most important risk factors for heart failure (HF). However, not all hypertrophied hearts will ultimately fail.1,2 This suggests that additional mechanisms, besides those that cause hypertrophy, are recruited during progression from compensated hypertrophy to failure. Possibly, failure-prone forms of left ventricular hypertrophy are already discernible on a molecular level at early stages, before transition toward overt HF has occurred. If failure-prone hypertrophied hearts would indeed express distinct molecular signs of their propensity to transgress to failure, this property would provide an opportunity to identify these failure-prone hearts at an early stage in the disease process.
Although recent studies have reported many molecular and cellular changes underlying cardiac hypertrophy,3,4 the additional factors that contribute to HF have remained elusive. In a hypothesis-driven search for mechanisms that characterize failing hypertrophied hearts, Boluyt et al documented the upregulation of genes encoding extracellular matrix components in spontaneously hypertensive rats with HF.5–8 However, it is not clear whether the overexpression of these genes preceded the overt clinical syndrome of HF, or whether their overexpression was a consequence of an established process of active failure.
Several unbiased approaches have also been used to identify mechanisms specific for HF.9,10 Microarray studies of human HF have generated a large body of data germane to the altered gene expression profile in failing hearts.11,12 However, these studies often compare end-stage and drug-treated myocardium with normal myocardium. Therefore, an important fraction of the differences obtained can be secondary to failure and its treatment,13 and such studies do not identify the factors that uniquely mark the hypertrophied heart that is prone to fail. We therefore compared the gene expression profiles from failing hypertrophied hearts and hypertrophied hearts that had remained compensated. We used homozygous renin transgenic TGR(mRen-2)27 (Ren-2) rats, because we had established that cardiac hypertrophy invariably develops in these hypertensive rat but that some of these outbred rats rapidly progress to HF, whereas other similarly hypertensive littermates remain compensated14 (unpublished observations).
This genomic study yielded 48 differentially expressed genes, among which we focused on thrombospondin-2 (TSP2). In cardiac biopsy specimens obtained at the early compensated hypertrophy phase, TSP2 was selectively increased only in rats that later progressed to failure but not in those that remained compensated. In TSP2 knockout mice, angiotensin II induced cardiomyopathy and cardiac rupture. This demonstrates that TSP2 is required for the myocardium to cope with increased cardiac loading, and that increased TSP2 expression at an early stage identifies the hypertrophied myocardium that is prone to fail.
Materials and Methods
Transgenic Rats and Hemodynamic Studies
Thirty male homozygous Ren-2 rats and 9 age-matched Sprague-Dawley (SD) rats (Max-Delbrück-Zentrum, Berlin, Germany) were studied. Eight Ren-2 rats were euthanized at age 10 weeks and 8 were treated with 0.05 mg/kg per day candesartan, an angiotensin II receptor type I blocker, from age 7 to 13 weeks. Of the remaining 14 untreated Ren-2 rats, 6 were euthanized at 13 weeks on clinical signs of HF and designated “HF” rats. The remaining 8 Ren-2 rats were monitored and euthanized at 17 weeks when clinical signs of failure had not appeared.
All described study protocols were approved by the Animal Care and Use Committee of the Universiteit Maastricht and were performed according to the official rules formulated in the Dutch law on care and use of experimental animals. The Dutch guidelines are highly similar to those of the National Institutes of Health, so that all experiments were also performed in accordance with these guidelines.
Biopy Specimens From 10-Week-Old Ren-2 Rats
In a second group of 9 Ren-2 and 4 SD rats, a cardiac biopsy was performed at age 10 weeks. Under anesthesia and ventilation, a lateral thoracotomy was performed and a biopsy specimen from the left ventricle was taken using a custom-made 0.35-mm needle. Of the 9 Ren-2 rats, 5 progressed to rapid HF, whereas the remaining 4 rats stayed compensated until the end of the observation period at 18 weeks.
In a separate study, we assessed the functional consequences of the cardiac biopsy procedure in 4 rats after biopsy by hemodynamic measurements (dP/dt) and histological analysis after 4 weeks.
Human Myocardial Biopsies
Cardiac biopsies specimens were obtained from 25 aortic stenosis patients during valve replacement surgery, before aortic cannulation. Echocardiography was performed 5 days before surgery. The study was approved by our Institutional Ethics Committee, and written informed consent was obtained from all patients.
RNA Isolation and Reverse-Transcription
RNA was isolated with an RNeasy Mini Kit (QIAGEN, Hilden, Germany). RNA was isolated from 10-week-old rat heart biopsy specimens with the PicoPure RNA Isolation Kit (Arcturus). The RNA was transcribed into cDNA with Superscript III reverse-transcriptase, using 250 ng of random primers (Invitrogen Life Technologies).
We used the Incyte GEM-2 and GEM-3 rat cDNA libraries (8736 and 8951 reporters, respectively, with 12 336 unique genes in total). Duplicate hybridizations were performed on these glass chips with 2 SD and 6 Ren-2 rat myocardial mRNAs at 10 weeks, 13 to 15 weeks (HF), or 17 weeks of age (compensated). We only included those reporter spots in which at least 40% of pixels displayed fluorescence >2.5-times local background. The protocol for data mining and validation was adopted as detailed previously.12,15 Genes with at least 1.7-fold overexpression in failing hearts were reprinted onto a subarray for further analysis so that the genes were independently assessed 4 times. Selected clones were obtained from Incyte, resequenced, and hybridized again.
Biological function of the 48 selected genes was screened and as a result, we chose 3 candidate genes for further testing.
Primers, Probes, and Real-Time Polymerase Chain Reaction
Primers and probes against the selected genes of interest (TSP2, osteoactivin, collagen VI, and brain natriuretic peptide [BNP], as a well established control) were designed from rat and human sequences available in GenBank using Primer Express Software (PE Applied Biosystems; primer sequences available in the online data supplement available at http://circres.ahajournals.org). Amplification and detection were performed using the ABI Prism 7700 Sequence Detection System and assessed relative to the expression level of a housekeeping gene, cyclophilin A.
Angiotensin II Infusion in TSP2 Knockout Mice
Twenty- to 30-week-old TSP2 knockout and wild-type mice on a 129SvJ EMS/Ter genetic background16 and weighing 25 to 30 grams were used. To examine if blood pressure effects of angiotensin II were independent of TSP2 genotype, dose-response curves were constructed by intravenous infusion of angiotensin II (doses of 0.5, 1.5, 5, 15, and 50 ng) in 5 wild-type and 5 TSP2 knockout mice. In a separate experiment, 0.5 μg/g per day angiotensin II was infused subcutaneously in 9 wild-type and 12 TSP2 knockout mice by an osmotic minipump (Alzet pumps) for 4 weeks. Given the high mortality in the TSP2 knockout mice treated with angiotensin II (see Results), we added 4 knockout and 4 wild-type mice, which were euthanized after 6 hours of angiotensin II infusion. At euthanization, the heart was removed and prepared for further analysis.17
Echocardiographic Analysis in Mice
Standard views were obtained by transthoracic echocardiography with a 12-MHz transducer (Hewlett Packard) on a Sonos 5500 (Hewlett Packard) echocardiograph in anesthetized mice (2% isoflurane).
Pro-Matrix Metalloproteinase-2 and Pro-Matrix Metalloproteinase-9 Zymography and Western Blotting
For gelatin zymography, pro-matrix metalloproteinase (MMP)-2 and -9 levels were measured as described.17,18 In brief, frozen cardiac tissue was incubated with 60 μL extraction buffer, and 60 mg of total protein was electrophoresed on a 10% Tris-glycine gel with 0.1% gelatin (Novex; SanverTECH). The gel was renatured for 30 minutes at room temperature, stained in 0.5% Coomassie Brilliant Blue R-250, and destained in buffer containing 45% ethanol and 10% acetic acid. The lysis of the substrate gel (area×intensity) was quantified as previously described.19 To demonstrate that the zymographic measurements were performed in the linear range of the curve and therefore can be interpreted in a quantitative manner,20 we ran a standard curve with increasing amounts of tissue extract of a sample with high zymographic activity (1.563 μg, 3.125 μg, 6.25 μg, 12.5 μg, 25 μg, 50 μg, and 100 μg of total protein).
Western blotting was performed to confirm the identity of pro-MMP-2 and -9. Extracts of cardiac tissue (50 μg) were applied to SDS-PAGE and immunoblotted with a rabbit polyclonal antibody MMP-9 (gift of H. Roger Lijnen) and MMP-2 (Biodesign International, Saco, Me) as recently described.19,21
Data are presented as mean±SEM. The data for each study group were compared using 1-way ANOVA in combination with a Dunnett post-hoc analysis to correct for multiple comparisons. Analyses were performed using the statistical package SPSS 10.0. P<0.05 was considered to be statistically significant.
Rapid Transition to Overt HF and Death in a Subset of Hypertensive Ren-2 Rats
Cardiac hypertrophy was found at 10 weeks and in untreated rats that were euthanized at later times. Six of 14 untreated Ren-2 rats rapidly transited toward HF between age 12 and 14 weeks and had decreased cardiac functional indices compared with the 8 rats that remained compensated throughout the observation period of 17 weeks (Table 1). Angiotensin II blockade prevented cardiac hypertrophy and failure (left ventricular weight/body weight percent, 2.52±0.36, dP/dtmax, 8400±202) in the treated rats euthanized at 13 weeks.
Microarray revealed 48 genes overexpressed in HF-susceptible rats. For microarray analysis, we first examined the biological variability in gene expression between failing and non-failing hearts in rats. On average, 81% of reporters per array were present in the heart lysates, of which 99.4% had similar expression levels (<1.7-fold difference) in failing and nonfailing hearts. Of a total of 12 336 genes profiled for expression, a multistep data-mining strategy revealed 48 upregulated and 14 downregulated genes in HF (data available in the online data supplement). Notably, expression of osteoactivin, TSP2, and several procollagens were increased.
The 48 overexpressed genes were analyzed bioinformatically. Initially, we classified them functionally using the GeneFIND (Gene Family Identification Network Design) System (http://www.nbrf.georgetown.edu), which combines several search/alignment tools to classify gene families.22 This indicated that most of the overexpressed genes encode matrix-related proteins. We focused on 3 genes: osteoactivin, TSP2, and collagen VI, which have not previously been implicated in cardiac disease. The data are available in the online data supplement.
Normalization of HF Susceptibility Genes by Angiotensin II Blockade
To confirm the role of angiotensin II in this model, we reassessed the expression of the target genes on treatment with a low-dose of the angiotensin receptor blocker (ARB) candesartan (a gift from AstraZeneca NL, Zoetermeer, the Netherlands) from age 7 to 13 weeks. Candesartan improved hemodynamics and prevented the overexpression of all HF-related genes, which was confirmed for TSP2 by real-time polymerase chain reaction (3.7±0.5 for compensated; 3.8±0.5 for ARB-treated; and 5.7±0.9 for HF [arbitrary units]; P<0.01 for compensated and ARB versus HF).
TSP2 Upregulation in Rats That Later Progressed to HF
To evaluate whether cardiac expression of osteoactivin, TSP2, and collagen VI preceded progression toward HF, we obtained cardiac biopsy specimens. After biopsy, the rat was allowed to recover to determine whether it would progress to HF. To assess the effects of the biopsy procedure, we had followed-up 4 control rats by echocardiographic and evaluated by histology 4 weeks after biopsy. This revealed no important effects of the biopsy itself. Measures of contractility of 4 control rats, which had undergone a biopsy, did not differ from the contractility of 4 sham-operated rats (data not shown). Echocardiographic indices of left ventricular mass and function did not differ at age 8 weeks between rats that later progressed to HF as compared with those that remained compensated (data not shown).
TSP2 expression was significantly increased at the early hypertrophy stage (10 weeks) only in those rats in which cardiac decompensation developed within 12 to 14 weeks (Figure 1a, HF), whereas it was not upregulated at this stage in the rats that subsequently remained compensated (comp) or in nontransgenic control rat hearts (SD). Expression levels of other HF-related candidate genes, such as osteoactivin (Figure 1b), collagen VI (Figure 1c), and the widely used marker of cardiac hypertrophy and failure, BNP23 (Figure 1d), were increased in the early hypertrophy stage in the rats that later developed heart failure and in those that remained compensated, compared with controls. Therefore, these genes could not distinguish failure-prone from failure-resistant rats. In accordance with our initial microarray studies, expression of these 3 genes increased further to >2-fold their 10-week expression levels on the development of HF. Compensated rats, despite having high osteoactivin, collagen VI, and BNP at 10 weeks, had no further significant increase in the expression levels of these genes on euthanization at 17 weeks (data not shown).
TSP2 Upregulation in Human Hypertrophy With Depressed Ejection Fraction
To determine whether TSP2 expression was also increased in human hypertrophied hearts with relatively impaired left ventricle function, we evaluated TSP2 expression in 2 groups of aortic stenosis patients (Table 2). Because fully compensated concentric left ventricular hypertrophy is usually accompanied by a normal to supranormal ejection fraction,1 we considered a mild impairment of ejection fraction to indicate a subtle early-stage loss of left ventricle function. Patients with a diminished lower ejection fraction (<55%) were thus compared with those with a preserved or supranormal ejection fraction (mean ejection fraction 51.2% ± 2.5 versus 64.15±1.0; P<0.005). TSP2 mRNA expression was significantly increased in the patients with depressed ejection fraction (Figure 2).
Angiotensin II Induces Cardiomyopathy and Cardiac Rupture in TSP2 Knockout Mice
To address the biological role of TSP2 in a model similar to the Ren2 rat (ie, hypertension with high angiotensin II levels), we infused angiotensin II (0.5 μg/g per day) in TSP2 knockout and wild-type mice. TSP2 knockout mice did not tolerate angiotensin II infusion, because 70% (7 of 10) of them died suddenly because of pericardial hemorrhagic tamponade within 4 days of infusion (4 mice died within 1 day) (Figure 3a). All wild-type mice, followed-up for up to 4 weeks, survived angiotensin II infusion. Measurement of the blood pressure response to increasing doses of angiotensin II did not show a significant difference in blood pressure response in wild-type as compared with TSP2 knockout mice (maximal increase in mean arterial pressure, 36.5±5.4 mm Hg and 30.0±4.9 mm Hg, respectively; P=NS) (Figure 3b). Left ventricular function at baseline was similar in wild-type as compared with TSP2 knockout mice (Table 3).
Surviving TSP2 knockout mice appeared ill and tachypneic. Therefore, TSP2 knockout (n=3) and a subgroup of wild-type (n=8) mice were euthanized within 6 days of angiotensin II infusion. Echocardiography revealed significantly decreased fractional shortening in TSP2 knockout as compared with wild-type mice (13.7±2.9% versus 24.1±1.6%; P=0.007) and increased end-diastolic dimensions within 6 days of angiotensin II infusion (Table 3).
Histology of ruptured TSP2-deficient hearts showed transmural myocardial hemorrhages with ruptured cardiac tissue (Figure 4A, a through c). Ultrastructural analysis revealed edema and disruption of the extracellular matrix with extravasation of red blood cells into the interstitial area of TSP2-deficient hearts at 6 hours of angiotensin II infusion (Figure 4B). Pronounced mitochondrial swelling and vacuolization in TSP2 knockout mice, as compared with mild mitochondrial changes in wild-type mice, indicated increased myocardial distress in the absence of TSP2. Interstitial swelling and myocardial damage dramatically increased in the absence of TSP2 at 6 hours of angiotensin II infusion. In contrast, we did not find abnormalities in the cardiac matrix in TSP2 knockout mice at baseline by either light or electron microscopy (Figure 4A, d, and Figure 4B, a). Interstitial fibrosis as quantified by Sirius red staining did not show differences at baseline between TSP2 knockout and wild-type mice (2.3±0.4% and 2.9±0.7%; n=4 and n=3, respectively) (Figure 4C, a). However, after 6 days of angiotensin II treatment, surviving TSP2 knockout mice showed less fibrosis compared with wild-type mice (4.2±0.5% versus 24.5±7.8%; n=5 and n=4, respectively; P<0.05) (Figure 4C, a through c).
Pro-MMP-2 and Pro-MMP-9 Zymography and Western Blotting
Semi-quantitative analysis of cardiac extracts on gelatin-containing gels20 revealed increased pro-MMP-2 and -9 enzyme activity in TSP2 knockout as compared with wild-type mice after 6 days of angiotensin II infusion (in arbitrary units of lysis, Spec_OD/A·U·mm2: pro-MMP-9 [94 kDA], 15.8±2.6 in TSP2 knockout versus 3.3±0.9 in wild-type mice, n=5, P=0.02; pro-MMP-2 [65 kDA], 16.5±2.5 in TSP2 knockout versus 11.2±1.1 in wild-type mice, n=5, P=0.04) (Figure 5). In vivo, these pro-enzymes serve as a constant source for the active enzymes.24 Baseline pro-MMP-2 or -9 levels did not differ significantly between TSP2 knockout and wild-type mice (arbitrary units of lysis, Spec_OD/A·U·mm2: pro-MMP-9, 3.7±0.9 in TSP2 knockout versus 2.2±0.2 in wild-type mice, n=4, P=NS; pro-MMP-2, 7.5±0.3 in TSP2 knockout versus 8.7±0.7 in wild-type mice, n=4, P=NS). A standard curve with decreasing amounts of tissue extract showed that our measurements have been performed in the exponential phase of the curve; therefore, our results can be interpreted in a semi-quantitative manner.20
Western blotting confirmed the identity of the pro-form of MMP-2 and MMP-9, and semi-quantitative analysis confirmed enhanced levels of pro-MMP-2 and -9 protein in TSP2 knockout as compared with wild-type mice at 6 days of angiotensin II infusion (in arbitrary units: pro-MMP-9, 41.2±13.2 in TSP2 knockout versus 7.8. ± 2.3 in wild-type mice, n=5, P=0.02; pro-MMP-2, 17.5±1.5 in TSP2 knockout versus 11.5±1.3 in wild-type mice, n=5, P=0.04). Baseline pro-MMP2 or -9 protein levels did not differ significantly between TSP2 knockout and wild-type mice (arbitrary units of lysis: pro-MMP-9, 11±1.3 in TSP2 knockout versus 7.8±1.4 in wild-type mice, n=4, P=NS; pro-MMP-2: 6.0±0.4 in TSP2 knockout versus 6.8±0.3 in wild-type mice, n=4, P=NS).
We demonstrate in this study of hypertensive renin-overexpressing rats that increased cardiac expression of TSP2 identifies those hearts that are prone to progress to HF. Pressure overload by angiotensin II infusion in mice deficient in TSP2 resulted in a severe dilated cardiomyopathy and cardiac rupture, whereas cardiac tissue of TSP2 knockout mice appeared and functioned normally under basal conditions. These results clearly indicate that TSP2 is essential for the maintenance of cardiac matrix integrity during cardiac overload. We suggest that the increased expression of cardiac TSP2 in the hypertrophic heart that is prone to failure reflects an unsuccessful adaptive attempt to avert imminent failure by recruiting a protein that either stabilizes the matrix or reduces its degradation. Other well-known adaptive responses to hypertrophy and HF include increased expression of BNP. However, we found BNP expression to be invariably increased in all forms of cardiac hypertrophy, thus this response could not distinguish between failure-prone and failure-resistant forms of hypertrophy. Therefore, the increase in TSP2 levels seems to reflect a mechanism that differs from those represented by known molecular markers of hypertrophy and HF.
We extend these findings to humans, because we also show that the expression of TSP2 is increased in human hypertrophied hearts in which dysfunction has developed, as compared with well-compensated hypertrophic hearts. This finding suggests that also in human cardiac hypertrophy, increased expression of TSP2 characterizes the dysfunctional forms of cardiac hypertrophy.
Although the family of thrombospondins has been studied extensively in vascular and thrombotic diseases, there are to date no reports to our knowledge that substantiate an important role for thrombospondins in HF. TSP2 is a secreted matricellular glycoprotein whose functions are diverse and incompletely understood.25,26 Because no close orthologues of TSP2 were found in the genomes of Caenorhabditis elegans or Drosophila, it appears that this protein has evolved to cope with the increased complexity of cell-matrix interaction in vertebrates.27 As evidence for a role of TSP2 in the organization of the extracellular matrix, previous studies in TSP2 knockout mice have shown that loss of TSP2 expression results in abnormally large collagen fibrils with irregular contours. Furthermore, the skin of TSP2 knockout mice is fragile and has reduced tensile strength.16 TSP2 knockout skin fibroblasts are defective in their attachment to a substratum and have increased levels of MMP-2 in their culture medium.28 It was subsequently shown that TSP2 is capable of binding both pro-MMP-2 and MMP-2, and that the resulting complexes are bound to the scavenger receptor, LRP1, and endocytosed.29 Thus, in the absence of TSP2, MMP-2 levels can be expected to increase in the pericellular environment.
In line with a central role for TSP2 in matrix assembly during injury, the present study documents loss of integrity of the cardiac matrix after angiotensin II-induced hypertension in mice lacking TSP2. As a consequence, these mice are prone to dilatation, intramyocardial bleeding, and fatal cardiac rupture. In contrast, cardiac abnormalities were not seen at baseline in TSP2 knockout mice. In our studies, cardiac rupture and dilatation after angiotensin II infusion were associated with increased MMP-2 and MMP-9 activity. Although these gelatinases have previously been implicated in cardiac rupture17,30 and dilatation30,31 after myocardial infarction, to our knowledge, such a dramatic phenotype induced by angiotensin II has never been reported in mice. High levels of MMP-2 and MMP-9 may therefore contribute to partial disruption of the myocardial matrix in TSP2 knockout mice after angiotensin II treatment, as evidenced by disorganization of the matrix and extravasation of erythrocytes from capillaries. Given that an increase in blood pressure is not transmitted to capillaries, and that we did not see arteriolar ruptures, it seems that these changes weaken the capillary wall sufficiently to cause it to rupture. In addition to this mechanism, deficient platelet function has been described previously in TSP2 knockout mice,32 suggesting that on rupture of supportive structures, extravasation of erythrocytes would be less efficiently halted in these mice.
Taken together, our data suggest that TSP2 is not essential to the myocardium under normal circumstances, but that the protein is needed to sustain even mild stresses. Parallel findings were recently reported in studies of cervical softening in TSP2 knockout mice.33 It was shown that application of a circumferential load that simulated the pressure of a birthing pup to a pregnant cervix from a TSP2 knockout mouse led to cervical rupture, whereas the cervix of a control mouse tolerated the stress without tissue damage.
This study was supported by a vidi grant (016.036.346) from the Netherlands Organization for Scientific Research (NWO) to Y.M.P., grant 2003T036 of the Netherlands Heart Foundation to S.H., a National Institutes of Health grant (AR45418) to P.B., and a Wellcome Trust Cardiovascular Genomics grant to J.D.
↵*These authors contributed equally to this work.
Original received July 28, 2003; resubmission received April 20, 2004; revised resubmission received July 21, 2004; accepted July 21, 2004.
Lorell BH, Carabello BA. Left ventricular hypertrophy: pathogenesis, detection, and prognosis. Circulation. 2000; 102: 470–479.
Manabe I, Shindo T, Nagai R. Gene expression in fibroblasts and fibrosis: involvement in cardiac hypertrophy. Circ Res. 2002; 91: 1103–1113.
Boluyt MO, Bing OH. Matrix gene expression and decompensated heart failure: the aged SHR model. Cardiovasc Res. 2000; 46: 239–249.
Brooks WW, Bing OH, Conrad CH, O’Neill L, Crow MT, Lakatta EG, Dostal DE, Baker KM, Boluyt MO. Captopril modifies gene expression in hypertrophied and failing hearts of aged spontaneously hypertensive rats. Hypertension. 1997; 30: 1362–1368.
Boluyt MO, Bing OH, Lakatta EG. The aging spontaneously hypertensive rat as a model of the transition from stable compensated hypertrophy to heart failure. Eur Heart J. 1995; 16: 19–30.
Kostin S, Pool L, Elsasser A, Hein S, Drexler HC, Arnon E, Hayakawa Y, Zimmermann R, Bauer E, Klovekorn WP, Schaper J. Myocytes die by multiple mechanisms in failing human hearts. Circ Res. 2003; 92: 715–724.
Hein S, Arnon E, Kostin S, Schonburg M, Elsasser A, Polyakova V, Bauer EP, Klovekorn WP, Schaper J. Progression from compensated hypertrophy to failure in the pressure-overloaded human heart: structural deterioration and compensatory mechanisms. Circulation. 2003; 107: 984–991.
Tan FL, Moravec CS, Li J, Apperson-Hansen C, McCarthy PM, Young JB, Bond M. The gene expression fingerprint of human heart failure. Proc Natl Acad Sci U S A. 2002; 99: 11387–11392.
Steenman M, Chen YW, Le Cunff M, Lamirault G, Varro A, Hoffman E, Leger JJ. Transcriptomal analysis of failing and nonfailing human hearts. Physiol Genomics. 2003; 12: 97–112.
Pinto YM, Pinto-Sietsma SJ, Philipp T, Engler S, Kossamehl P, Hocher B, Marquardt H, Sethmann S, Lauster R, Merker HJ, Paul M. Reduction in left ventricular messenger RNA for transforming growth factor beta(1) attenuates left ventricular fibrosis and improves survival without lowering blood pressure in the hypertensive TGR(mRen2)27 Rat. Hypertension. 2000; 36: 747–754.
Kyriakides TR, Zhu YH, Smith LT, Bain SD, Yang Z, Lin MT, Danielson KG, Iozzo RV, LaMarca M, McKinney CE, Ginns EI, Bornstein P. Mice that lack thrombospondin 2 display connective tissue abnormalities that are associated with disordered collagen fibrillogenesis, an increased vascular density, and a bleeding diathesis. J Cell Biol. 1998; 140: 419–430.
Heymans S, Luttun A, Nuyens D, Theilmeier G, Creemers E, Moons L, Dyspersin GD, Cleutjens JP, Shipley M, Angellilo A, Levi M, Nube O, Baker A, Keshet E, Lupu F, Herbert JM, Smits JF, Shapiro SD, Baes M, Borgers M, Collen D, Daemen MJ, Carmeliet P. Inhibition of plasminogen activators or matrix metalloproteinases prevents cardiac rupture but impairs therapeutic angiogenesis and causes cardiac failure. Nat Med. 1999; 5: 1135–1142.
Lehoux S, Lemarie CA, Esposito B, Lijnen HR, Tedgui A. Pressure-induced matrix metalloproteinase-9 contributes to early hypertensive remodeling. Circulation. 2004; 109: 1041–1047.
Wu CH, Shivakumar S, Shivakumar CV, Chen SC. GeneFIND web server for protein family identification and information retrieval. Bioinformatics. 1998; 14: 223–224.
Yang Z, Kyriakides TR, Bornstein P. Matricellular proteins as modulators of cell-matrix interactions: adhesive defect in thrombospondin 2-null fibroblasts is a consequence of increased levels of matrix metalloproteinase-2. Mol Biol Cell. 2000; 11: 3353–3364.
Yang Z, Strickland DK, Bornstein P. Extracellular matrix metalloproteinase 2 levels are regulated by the low density lipoprotein-related scavenger receptor and thrombospondin 2. J Biol Chem. 2001; 276: 8403–8408.
Hayashidani S, Tsutsui H, Ikeuchi M, Shiomi T, Matsusaka H, Kubota T, Imanaka-Yoshida K, Itoh T, Takeshita A. Targeted deletion of MMP-2 attenuates early LV rupture and late remodeling after experimental myocardial infarction. Am J Physiol Heart Circ Physiol. 2003; 285: H1229–H1235.
Ducharme A, Frantz S, Aikawa M, Rabkin E, Lindsey M, Rohde LE, Schoen FJ, Kelly RA, Werb Z, Libby P, Lee RT. Targeted deletion of matrix metalloproteinase-9 attenuates left ventricular enlargement and collagen accumulation after experimental myocardial infarction. J Clin Invest. 2000; 106: 55–62.
Kyriakides TR, Rojnuckarin P, Reidy MA, Hankenson KD, Papayannopoulou T, Kaushansky K, Bornstein P. Megakaryocytes require thrombospondin-2 for normal platelet formation and function. Blood. 2003; 101: 3915–3923.
Kokenyesi R, Armstrong LC, Agah A, Artal R, Bornstein P. Thrombospondin 2 Deficiency in Pregnant Mice Results in Premature Softening of the Uterine Cervix. Biol Reprod. 2004; 70: 385–390.