Distinct Upregulation of Extracellular Matrix Genes in Transition From Hypertrophy to Hypertensive Heart Failure
Cardiac hypertrophy in response to pressure overload is initially beneficial but eventually leads to heart failure, a major cause of morbidity and mortality in the Western countries. Although abnormalities in left ventricular (LV) diastolic filling are early features associated with pressure overload-induced LV hypertrophy, the molecular mechanisms regulating transition to diastolic heart failure are poorly understood. We analyzed global changes in gene expression in 12-, 16-, and 20-month-old spontaneously hypertensive rats (SHR) and their age-matched controls, Wistar Kyoto rats, using DNA microarrays. In SHR, a progressive LV hypertrophy was associated with increased expression of hypertrophy-associated genes including contractile protein and natriuretic peptide genes. Echocardiography indicated that 16-month-old SHR had features of diastolic dysfunction leading to diastolic failure at age 20 months without significant changes in LV systolic function. Comparison analysis revealed that the extracellular matrix genes strikingly dominated the list of altered genes after transition to the heart failure, whereas there was no major shift in gene expression patterns involved in calcium homeostasis and neurohumoral activation, as well as myofilament contractile and cytoskeletal proteins. The microarray analysis also revealed differential gene expression of several novel factors, such as thrombospondin-4 and matrix Gla protein, as well as unknown expressed sequence tags. Our data show that transition from LV hypertrophy to diastolic hypertensive heart failure is almost exclusively associated with progressive remodeling of the extracellular matrix and provide new insights into the pathogenesis of hypertrophy by suggesting existence of novel regulators of LV remodeling.
Heart failure is a complex clinical syndrome that can result from any structural or functional cardiac disorder that impairs the ability of the ventricle to fill or eject blood.1 The principal hallmark of patients with predominant systolic dysfunction is a depressed left ventricular ejection fraction (LVEF). When the symptoms and signs of heart failure are accompanied by a preserved LVEF and a predominant or isolated abnormality in diastolic function, this clinical syndrome is called diastolic heart failure.1–3 Clinically, patients with diastolic heart failure are elderly, more likely to be women, and often have hypertension associated with left ventricular hypertrophy (LVH).3 Diastolic heart failure occurs when the ventricular chamber is unable to accept an adequate volume of blood during diastole, at normal diastolic pressures, and at volumes sufficient to maintain an appropriate stroke volume. These abnormalities are caused by a decrease in ventricular relaxation (caused by an increase in afterload) and/or an increase in ventricular stiffness (caused by fibrosis).1–5
The mechanisms that cause abnormalities in diastolic function and lead to the development of diastolic heart failure can be divided into factors intrinsic to the myocardium itself and factors that are extrinsic to the myocardium.4,5 Myocardial factors include changes in calcium homeostasis (abnormalities in the sarcolemmal channels, sarcoplasmic reticulum calcium reuptake and in the phosphorylation state of the proteins that modify sarcoplasmic reticulum calcium ATPase function), the myofilament contractile proteins (tropomyosin and troponin T, C, and I), and cytoskeletal proteins (eg, desmin, actin, and titin).4,6 Also, changes in interstitial fibrosis (fibrillar collagens, proteoglycans, basement membrane proteins) and the extracellular matrix (ECM), as well as neurohumoral (renin-angiotensin-aldosterone system, sympathetic nervous system) and cardiac endothelial activation and/or inhibition, may contribute to alterations in active relaxation and increased myocardial stiffness and thus lead to the development of diastolic dysfunction.4 However, although abnormalities in LV diastolic filling are early features associated with hypertension-induced hypertrophy,7 the precise molecular mechanisms and the relative role of myocardial and extramyocardial mechanisms, cellular and extracellular mechanisms, and neurohumoral activation in transition from LVH to diastolic heart failure are poorly understood.
In the present study, we used DNA microarrays to define comprehensive gene expression profiles in an experimental model of genetic hypertension, spontaneous hypertension in rats, during the development of LVH and in transition to diastolic heart failure. Spontaneously hypertensive rats (SHR) have alterations in active relaxation and passive compliance characterizing the diastolic dysfunction and therefore provides a useful model for genomic studies of diastolic heart failure.8,9
Male 12-, 16-, and 20-month-old SHR of the Okamoto-Aoki strain and their age-matched Wistar-Kyoto (WKY) rats from the colony of the Centre of Experimental Animals at the University of Oulu, Finland, were used (n=4 to 11 per group). SHR strain was originally obtained from Mollegaards Avslaboratorium, Skensved, Denmark. The experimental design was approved by the Animal Care and Use Committee of the University of Oulu, Finland.
Echocardiography and RNA Analysis
The methods for echocardiography and RNA analysis are described in an online Methods supplement available at http://hyper.ahajournals.org.
Microarray Analysis and Data Analysis
Expression profiling was performed with Rat Genome U34A GeneChips (Affymetrix). Total LV RNA from 12-, 16-, and 20-month-old SHR and WKY rats (n=4, except n=3 at 20 months old for WKY) was used. GeneChip experiments were performed according to Affymetrix’s protocols at the Turku Centre for Biotechnology by National Microarray service. Raw data analysis was performed using Affymetrix software. Normalization and filtering of data were performed with the GeneSpring software (Silicon Genetics). Genes were defined as differentially expressed if the fold change was at least 1.5-fold and statistically significant (Welch t test; P<0.05). The expanded methods are provided in the online Methods supplement at http://hyper.ahajournals.org.
Blood Pressure and LVH in SHR
SHR are a useful experimental model of essential hypertension.8,9 Mean arterial pressure is significantly higher already in young animals and continues to increase with aging, as we and others have previously reported.10,11 The increase in pressure overload was associated with a progressive LVH, as reflected by increased LV weight-to-body weight ratio (online Figure I, see http://hyper.ahajournals.org).
Expression of LVH-Associated Genes
At the age of 16 and 20 months, Northern blot analysis demonstrated increased LV atrial natriuretic peptide gene expression in SHR, which is a known feature of hypertrophied heart12 (Figure I). To further characterize the experimental model, we measured mRNA levels of different isoforms of contractile protein genes in the left ventricle. A progressive increase was seen in β-myosin heavy chain/α- myosin heavy chain ratio and skeletal α-actin/cardiac α-actin ratio in SHR with aging (Figure I). Moreover, collagen IIIα1 mRNA levels were 2.2-fold (P<0.05) higher at 16 months and 3.4-fold (P<0.01) higher at 20 months (Figure I); also, collagen Iα1 mRNA levels were significantly increased in 20-month-old SHR when compared with 12-month-old SHR (4.7-fold; P<0.01) (Figure I), whereas no change in collagen mRNA levels was seen in WKY rats. To evaluate changes in gene expression of factors promoting growth and fibrosis, we measured mRNA levels of endothelin-1 and transforming growth factor (TGF)-β1. Gene expression of endothelin-1 and TGF-β1 was significantly increased (P<0.05) in 20-month-old SHR (Figure I).
Echocardiography revealed increased thickness of LV wall and features of diastolic dysfunction (Figure 1 and Figure II) characterized by decreased ratio of peak flow velocity of the early rapid diastolic filling wave to peak flow velocity of the late diastolic filling wave (E/A ratio) (2.4±0.2 at 12 months versus 1.8±0.2 at 16 months) (P<0.05) and prolonged LV isovolumic relaxation time in 16-month-old SHR (Figure 1). At age 20 months, E/A ratio was significantly increased above normal (4.8±0.6) (P<0.001), suggesting development of diastolic heart failure. In contrast, no significant change in LV systolic function was seen (Figure 1). Thus, old SHR had heart failure with a preserved LVEF, abnormal LV diastolic properties, and marked LVH.
DNA Array Analysis
To identify genes that are associated with transition of LVH to diastolic heart failure, the LV gene expression profiles from 12-, 16-, and 20-month-old SHR were compared with profiles seen in WKY rats by screening Affymetrix U34A arrays. Age-matched WKY controls were used to exclude gene expression changes related to aging. Venn diagrams (Figure 2A) show the number of genes that were upregulated or downregulated at least 1.5-fold in the left ventricle among the SHR age groups. Genes were further organized into groups representing their known biological functions, including cell division, cell signaling/communication, cell structure/motility, cell/organism defense, gene expression, protein expression, metabolism, genes of unknown function, and expressed sequence tags (ESTs) (Figure 2B to 2D). All genes in the cell structure/motility group were upregulated and predominantly encode ECM proteins, whereas in the cell signaling, cell defense, and metabolism groups, there were enhanced and repressed genes. However, very few genes changed in categories of cell division and gene and protein expression during development of hypertensive heart disease. A number of upregulated genes observed in the present study have been previously described to be elevated during cardiac hypertrophy and failure in SHR, including atrial natriuretic peptide, collagen I, collagen III, fibronectin, and osteopontin.8–9 Several downregulated genes encode proteins involved in fatty acid and energy metabolism, among others enoyl-coenzyme A (CoA) isomerase and acyl-CoA dehydrogenase.
When we examined the entire time course of the progression of LVH and diastolic heart failure, analysis identified 127 transcripts that showed differential expression. Comparison of LV RNA profiles from 20- and 12-month-old SHR identified 61 known genes and 20 ESTs, whose expression was upregulated >1.5-fold, and 31 known genes and 15 ESTs, whose expression was downregulated >1.5-fold (Table 1⇓). Comparison of LV RNA profiles during development of LVH (16-month-old versus 12-month-old SHR) showed that the expression of 13 known genes and 6 ESTs was elevated >1.5-fold and expression of 12 genes (9 known genes and 3 ESTs) was downregulated >1.5-fold (Table 2). Most of the enhanced genes encoded ECM proteins, whereas the majority of the repressed genes were encoding metabolic enzymes (Figure 2C). Importantly, after development of diastolic dysfunction at age 16 months (20-month-old versus 16-month-old SHR), only 9 known genes and 5 ESTs were upregulated and 2 genes were downregulated (1 known gene and 1 EST) >1.5-fold (Table 3). The majority of upregulated genes encode ECM proteins and, interestingly, many ESTs were upregulated (Figure 2D). The differentially expressed genes in WKY with aging are shown in Tables II, III, and IV (http://hyper.ahajournals.org).
We confirmed selected microarray results by comparison with mRNA levels obtained by Northern blot analysis or quantitative real-time reverse-transcription polymerase chain reaction. As shown in Table V, we observed similar fold changes in mRNA levels in SHR as measured by both microarray and Northern/reverse-transcription polymerase chain reaction.
The gene expression patterns of matrix Gla protein, TGF-β-stimulated clone-22 (TSC-22), thrombospondin-4, and EST-sequence AA800844 (similar to mouse lysyl oxidase-like protein) define these genes as potential novel modulators of LV remodeling in diastolic heart failure, as shown in Figure 3.
Congestive heart failure caused by a predominant abnormality in diastolic function is common and causes significant morbidity and mortality.2,6 However, mechanisms that cause diastolic heart failure, ie, abnormalities in the passive elastic properties of the left ventricle and/or the process of active ventricular relaxation, are not fully understood. In the present study, we used DNA microarrays to identify genes that may play a role in pathophysiologic changes of cardiac structure and function in an experimental model of diastolic hypertensive heart failure.
Consistent with previous reports,13 a progressive LVH and thickening of LV walls was seen in SHR, leading to diastolic heart failure with preserved systolic performance. We identified 92 known genes and 35 ESTs that were differentially expressed in the left ventricle of 20-month-old SHR compared with 12-month-old SHR. To rule out age-related changes, we excluded the genes that showed altered expression both in SHR and WKY in DNA microarray analysis. The relatively small number of altered genes agrees with previous observations showing that more genes alter in response to acute than to chronic overload.14–15 The majority of upregulated genes encode cell structure and signaling proteins, whereas most of the downregulated genes encode proteins involved in fatty acid and energy metabolism. Many of these genes are known contributors of LVH and heart failure in SHR.8,9
A key finding of the present study is that there was no extensive shift in gene expression patterns when diastolic dysfunction was progressing to diastolic heart failure. Instead, significant changes in gene expression developed over time associating with hypertrophic process and development of heart failure. Moreover, our results show that the transition to heart failure is mainly a consequence of increased ECM composition, leading to myocardial stiffness and abnormal relaxation, because after development of diastolic dysfunction at the age of 16 months, almost all of changes were seen in genes encoding ECM proteins. In addition, upregulation of several inhibitors of proteolytic enzymes (tissue inhibitor of metalloproteinase-1, plasminogen activator inhibitor-1) and cysteine endoproteases (cathepsin K and cathepsin S) suggests dynamic regulation of matrix degradation and deposition during the LV remodeling process in hypertrophic heart failure.
We identified a number of genes not previously associated with the development of hypertensive cardiac hypertrophy or diastolic heart failure. Even though gene expression information alone is not enough to define the role of these genes, several new genes need a particular note. Increased thrombospondin-4 mRNA levels have been previously reported in patients with end-stage dilated cardiomyopathy16 and after myocardial infarction in rats,17 suggesting that thrombospondin-4 plays an important role in mediating remodeling process in heart failure. Increased expression of matrix Gla protein, an alleged calcification inhibitor, is reported in human hypertrophic heart18 and failing mouse hearts after chronic pressure overload.14 Elevated mRNA levels of TSC-22, a TGF-β–inducible repressor of transcription, has been reported after myocardial infarction in rats.19 This may indicate that TSC-22 has a role in controlling the transcriptional response of cardiac remodeling. However, it should be noted that TSC-22 mRNA levels increased slightly with aging also in WKY rats, suggesting that other factors that diastolic dysfunction may regulate TSC-22 gene expression.
Most of the downregulated genes are involved in fatty acid and energy metabolism like enoyl-CoA isomerase and acyl-CoA dehydrogenase. Repression of metabolic genes indicates a long-term adaptation process in myocardial bioenergetics, consistent with previous findings.20 Interestingly, the expression of a Ca2+-sensing protein S100A1 was repressed with aging in SHR. S100A1 protein levels are reduced in human end-stage heart failure;21 in transgenic mice, overexpression of S100A1 leads to increased myocardial performance.22 As a regulator of calcium homeostasis and cytoskeleton dynamics,21–22 S100A1 may play a role in causative mechanisms of diastolic heart failure.
In addition to known genes, we identified differential expression of several ESTs. Because many ESTs have limited sequences, it is not possible to assign them to structural protein families. One of the EST sequences (AA800844) shows similarity to mouse lysyl oxidase. Lysyl oxidases are extracellular copper enzymes, and recent report shows that inactivation of the mouse lysyl oxidase gene leads to cardiovascular dysfunction.23 Another EST sequence (AA859885) shows similarity to mouse TGF-β–inducible protein 36 (TSC-36). TSC-36 is a TGF-β–inducible follistatin-related protein, and increased TSC-36 expression has been reported in transgenic mouse hearts overexpressing calsequestrin.24
To our knowledge, this is the first comprehensive transcription profiling study of diastolic hypertensive heart disease. In the ending phase of diastolic heart failure, changes were seen almost exclusively in genes encoding ECM proteins. Recent human gene expression profiling studies have examined failing hearts with different diagnoses of end-stage cardiomyopathies.16,25 Although some similarity between this and human microarray studies was noted (eg, upregulation of atrial natriuretic peptide, collagen I, and thrombospondin-4), comparison is complex because of human failing hearts have abnormalities also in systolic function. An obvious difference can be seen also between the present study and DNA microarray analysis of myocardial infarction with systolic dysfunction17,19,26 or progression of heart failure in Dahl salt-sensitive rats.27 Although many genes encoding ECM proteins were altered, there were also other equally important functional groups of differentially expressed genes. In our study, the ECM and structural genes strikingly dominated the list of altered genes after transition to the heart failure, suggesting that this could be a unique genetic pattern for the diastolic heart failure. Changes in calcium homeostasis, myofilament contractile proteins, cardiomyocyte cytoskeleton proteins, and extramyocardial factors (neurohumoral activation) seem not to play a significant role in the development of diastolic heart failure caused by pressure-overload hypertrophy.
We evaluated the molecular basis of diastolic heart failure by transcription profiling. The present study shows a long-term adaptive changes in transcription of genes related to cellular and extracellular architecture, myocardial energetics, and signaling molecules mediating remodeling process. Transition from LVH to diastolic hypertensive heart failure was almost exclusively associated with changes in genes encoding ECM proteins and their regulatory processes. Moreover, the identification of a variety of new matrix-related genes showed a prevalence of progressive ECM remodeling that will predispose to failure. The results also provide new insights into the pathogenesis of LVH and diastolic heart failure by suggesting existence of novel modulators of LV remodeling.
This study was supported by the Academy of Finland, Aarne Koskelo Foundation, the Finnish Foundation for Cardiovascular Research, National Technology Agency of Finland, and the Sigfrid Jusélius Foundation. We thank Marja Arbelius, Sirpa Rutanen, and Kati Viitala for expert technical assistance.
- Received January 19, 2005.
- Revision received January 19, 2005.
- Accepted February 28, 2005.
Vasan RS, Levy D. Defining diastolic heart failure: a call for standardized diagnostic criteria. Circulation. 2000; 101: 2118–2121.
Zile MR, Brutsaert DL. New concepts in diastolic dysfunction and diastolic heart failure: Part II: causal mechanisms and treatment. Circulation. 2002; 105: 1503–1508.
Zile MR, Brutsaert DL. New concepts in diastolic dysfunction and diastolic heart failure: Part I: diagnosis, prognosis, and measurements of diastolic function. Circulation. 2002; 105: 1387–1393.
Kass DA, Bronzwaer JG, Paulus WJ. What mechanisms underlie diastolic dysfunction in heart failure? Circ Res. 2004; 94: 1533–1542.
Mandinov L, Eberli FR, Seiler C, Hess OM. Diastolic heart failure. Cardiovasc Res. 2000; 45: 813–825.
Boluyt MO, Bing OH. Matrix gene expression and decompensated heart failure: the aged SHR model. Cardiovasc Res. 2000; 46: 239–249.
Kinnunen P, Vuolteenaho O, Uusimaa P, Ruskoaho H. Passive mechanical stretch releases atrial natriuretic peptide from rat ventricular myocardium. Circ Res. 1992; 70: 1244–1253.
Kuoppala A, Shiota N, Lindstedt KA, Rysä J, Leskinen HK, Luodonpää M, Liesmaa I, Ruskoaho H, Kaaja R, Kovanen PT, Kokkonen JO. Expression of bradykinin receptors in the left ventricles of rats with pressure overload hypertrophy and heart failure. J Hypertens. 2003; 21: 1729–1736.
Cingolani OH, Yang XP, Cavasin MA, Carretero OA. Increased systolic performance with diastolic dysfunction in adult spontaneously hypertensive rats. Hypertension. 2003; 41: 249–254.
Weinberg EO, Mirotsou M, Gannon J, Dzau VJ, Lee RT, Pratt RE. Sex dependence and temporal dependence of the left ventricular genomic response to pressure overload. Physiol Genomics. 2003; 12: 113–127.
Larkin JE, Frank BC, Gaspard RM, Duka I, Gavras H, Quackenbush J. Cardiac transcriptional response to acute and chronic angiotensin II treatments. Physiol Genomics. 2004; 18: 152–166.
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.
Jin H, Yang R, Awad TA, Wang F, Li W, Williams SP, Ogasawara A, Shimada B, Williams PM, de Feo G, Paoni NF. Effects of early angiotensin-converting enzyme inhibition on cardiac gene expression after acute myocardial infarction. Circulation. 2001; 103: 736–742.
Stanton LW, Garrard LJ, Damm D, Garrick BL, Lam A, Kapoun AM, Zheng Q, Protter AA, Schreiner GF, White RT. Altered patterns of gene expression in response to myocardial infarction. Circ Res. 2000; 86: 939–945.
Sack MN, Rader TA, Park S, Bastin J, McCune SA, Kelly DP. Fatty acid oxidation enzyme gene expression is downregulated in the failing heart. Circulation. 1996; 94: 2837–2842.
Most P, Remppis A, Pleger ST, Loffler E, Ehlermann P, Bernotat J, Kleuss C, Heierhorst J, Ruiz P, Witt H, Karczewski P, Mao L, Rockman HA, Duncan SJ, Katus HA, Koch WJ. Transgenic overexpression of the Ca2+-binding protein S100A1 in the heart leads to increased in vivo myocardial contractile performance. J Biol Chem. 2003; 278: 33809–33817.
Mäki JM, Räsänen J, Tikkanen H, Sormunen R, Mäkikallio K, Kivirikko KI, Soininen R. Inactivation of the lysyl oxidase gene Lox leads to aortic aneurysms, cardiovascular dysfunction, and perinatal death in mice. Circulation. 2002; 106: 2503–2509.
Steenbergen C, Afshari CA, Petranka JG, Collins J, Martin K, Bennett L, Haugen A, Bushel P, Murphy E. Alterations in apoptotic signaling in human idiopathic cardiomyopathic hearts in failure. Am J Physiol Heart Circ Physiol. 2003; 284: H268–H276.
Sehl PD, Tai JT, Hillan KJ, Brown LA, Goddard A, Yang R, Jin H, Lowe DG. Application of cDNA microarrays in determining molecular phenotype in cardiac growth, development, and response to injury. Circulation. 2000; 101: 1990–1999.