Activation of matrix metalloproteinase dilates and decreases cardiac tensile strength

International Journal of Cardiology 79 (2001) 277–286 www.elsevier.com / locate / ijcard

Activation of matrix metalloproteinase dilates and decreases cardiac tensile strength Vibhas S. Mujumdar, Lane M. Smiley, Suresh C. Tyagi* Department of Physiology and Biophysics, School of Medicine, The University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216 -4505, USA Received 4 January 2001; received in revised form 3 April 2001; accepted 10 April 2001

Abstract Previous studies demonstrated that transition from compensatory pressure overload hypertrophy to decompensatory volume overload heart failure is associated with decreased cardiac tensile strength and activation of matrix metalloproteinase (MMP) in spontaneously hypertensive rat (SHR). To test the hypothesis that in the absence of nitric oxide activation of MMP during cardiac failure causes disruption in the organization of extracellular matrix (ECM) and leads to decrease systolic and diastolic cardiac tensile strength, we employed SHR of 24–32 weeks, which demonstrates significant cardiac hypertrophy and fibrosis. The normotensive Wistar rats (NWR) were used as control. To determine whether cardiac hypertrophy is associated with increased elastinolytic matrix metalloproteinase-2 (MMP-2) activity; quantitative elastin-zymography was performed on cardiac tissue homogenates. The MMP-2 activity was normalized by the levels of actin. The MMP-2 / actin ratio was 2.060.5 in left ventricle (LV) and 1.560.25 in right ventricle (RV) of SHR 32wks ; and 0.560.25 in LV and 0.2560.12 in RV of NWR 32wks (P,0.02 when SHR compared with NWR). To measure passive diastolic cardiac function, rings from LV as well as RV through transmyocardial wall from male SHR and NWR of 6–8 weeks and 24–36 weeks were prepared. The LV wall thickness from endocardium to epicardium was 3.7560.25 mm in SHR 32wks as compared to 2.2560.50 mm in NWR 32wks (P,0.01). The ring was placed in tissue myobath and length–tension relationships were assessed. The pressure–length relationship was shifted to left in SHR as compared to NWR. The amounts of cardiac elastin and collagen were determined spectrophotometrically by measuring desmosine–isodesmosine and hydroxyproline contents, respectively. A negative correlation between elastic tensile strength and elastin / collagen ratio was elucidated. To create situation analogous to heart failure and MMP activation, we treated cardiac rings with active MMP-2 and length–tension relation was measured. The relationship was shifted to right in both SHR and NWR when compared to their respective untreated groups. The results suggested that activation of MMP led to decreased cardiac tissue tensile strength and may cause systolic and diastolic dysfunction.  2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Fibrosis; Wound healing; Hypertrophy; Gelatinase; Spontaneously hypertensive rat; Remodeling; Hypertension; Elastin; Collagen; Extracellular matrix; Contractile function; Heart failure

1. Introduction In the study of left ventricle dysfunction (SOLVD), the investigators demonstrated that in normal human heart, after acute ischemic injury, left ventricle (LV) *Corresponding author. Tel.: 11-601-984-1899; fax: 11-601-9841817. E-mail address: [email protected] (S.C. Tyagi).

wall stress increases whereas wall thickness and ejection fraction decrease. To compensate for the injury response the entire myocardium undergoes compensatory hypertrophy and remodels the entire chamber. While wall thickness is increased during compensatory response, the ejection fraction and wall stress are maintained. However, in decompensatory congestive heart failure, the wall stress increases, and ejection fraction and wall thickness decrease sig-

0167-5273 / 01 / $ – see front matter  2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 01 )00449-1

278

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

nificantly [1]. The spontaneously hypertensive rat (SHR) is a laboratory model of naturally developing hypertension that appears to be similar in a number of respects to essential hypertension in humans [2,3]. Previous studies in the SHR have demonstrated that in addition to LV hypertrophy there is also an increase in interstitial fibrillar collagen [4]. Structural remodeling of the extracellular matrix (ECM) has been implicated in alterations in myocardial stiffness [5], which may contribute to both systolic and diastolic dysfunction in hypertrophied hearts. Remodeling by its very nature implies synthesis and degradation of ECM, leading to alteration in the ECM composition and concentration. The matrix metalloproteinase-2 (MMP-2) which especially degrades elastin [6] as well as interstitial fibrillar collagen [7] may induce systolic and diastolic impairment. One of the causes of myocardial wall thinning during transition from compensatory hypertrophy to decompensatory heart failure is in part due to increased MMPs activity and substantial ECM degradation [8,9]. In SHR model, persistent hypertension begins at 6–8 weeks and is followed relatively long period (24–36 weeks) of stable compensatory pressure overload hypertrophy. At |70 weeks male SHR develops volume overload with decompensatory heart failure [2] in which cardiac ECM is being degraded and MMPs are robustly activated [10] and the levels of tissue inhibitor of metalloproteinase-4 (TIMP-4) are abrogated [10]. Although MMPs are activated in end-stage heart failure and cardiac wall looses its tensile strength. It is not clear whether the activation of MMP is one of the causes of decreased tensile strength in failing heart. We hypothesize that activation of MMP during the development of cardiac failure leads to decrease systolic and diastolic cardiac tensile strength.

2. Materials and methods

2.1. Experimental model of LV hypertrophy and fibrosis Spontaneously hypertensive rats (SHR) have been shown to develop impaired myocardial function after a period (30–36 weeks) of stable hypertrophy. Trippodo and Frohlich [3] have summarized the evi-

dence supporting the use of SHR as a genetic model of hypertension and related this model to human heart hypertrophic disease. The changes in these animals are well characterized at both the functional and morphological levels [2,11]. Age matched genetically SHR and their control normotensive Wistar rat (NWR) were obtained from Charles River laboratories. This strain of SHR develop hypertension at 6 weeks of age and a steady-state elevation in blood pressure by 24–36 weeks of age; NWR remains normal throughout this time span. All studies conformed to the principles of the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the protocol was approved by our Institutional Animal Care and Use Committee.

2.2. Isolation of heart and cardiac ring preparation To determine the role of adverse ECM remodeling in cardiac dysfunction, SHR8w and SHR32w, and their normotensive control NWR8w and NWR32w, respectively, were sacrifice and heart was removed under anesthesia. SHR 32wks develops significant cardiac fibrosis and hypertrophy while their age-sex matched normotensive control has normal ECM composition. The heart weight / body weight ratio was determined. The LV and RV wall thickness and diameters were measured using a digital micrometer under 1.53 magnifying lens and using a dissecting microscope. To determine the LV and RV function separately, the left and right ventricle (LV and RV) rings were prepared from SHR and NWR of 6–8 and 24–32 weeks old. The ex vivo circumferential cardiac rings from LV and RV were sliced (in a ‘deli’ like fashion) as follows: two horizontal myocardial cross sections of 2–3 mm in thickness through the ventricular cavity were dissected from whole heart. The LV rings were obtained after removing the RV by cutting the RV wall. The RV rings were obtained after removing the LV by cutting the LV wall. Therefore, the septum was intact in both types of rings and opposite ventricle was discarded. Thereby, individual LV and RV cardiac rings were prepared [12]. To minimize the time-dependent decrease in cardiac contractility in ex vivo rings, the experiments were carried out within 10 min after removal of the heart. To keep spherical shape of the ring, a pedriatic

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

esophagus balloon was place inside the ring. To avoid hypoxia and ensure enough oxygen in the microvessels, the tissue was aerated with 95% O 2 and 5% CO 2 at 20 p.s.i. continuously. During this time no significant diminution in cardiac function (i.e. response to CaCl 2 ) was observed.

2.3. Buffer and instrumentation The endocardial rings were placed in ice-cold Krebs’ bicarbonate solution containing (mM) NaCl 131.5, KCl 0.2, NaH 2 PO 4 1.2, MgCl 2 1.2, CaCl 2 0.5, NaHCO 3 23.5 and glucose 11.2. The solution was perfused with 95% O 2 –5% CO 2 (pH 7.4) and equilibrated at 378C. To measure length tension, the rings were prepared and mounted in between the two stainless steel wires; one connected to a force transducer (World Precision Instrument, Inc., FL); the other connected to a micrometer in a tissue myobath. The signal from the ring under experimentation was digitized by on-line analysis using Pickup-95 (A software created by our electronic shop). To validate the viability of cardiac rings, the rings were contracted three times by inducing active muscle tone using 20 mM CaCl 2 , rinsed and reequilibrated before the measurements. After in vitro stabilization, each cardiac ring was systematically stretched to the optimum of its lengthactive tension. A known amount of stretch was placed on the ring, and the contraction was measured. Because CaCl 2 has been shown to induce contraction in the myocardium [13], after maximum stretch the ring was brought to resting tension (|20% of maximum) and a known amount of CaCl 2 was added. The unloaded ventricular ring diameter (mm) and initial circumference of ventricular ring was measured using a micrometer under magnifying lens, with a dissecting microscope. The cardiac ring was then stretched by twisting the knob of micrometer and tension in gram was recorded. The tension in gram was converted to dynes / cm 2 . The area for spherical cardiac ring was determined: 3.143r3r (where r is radius of the ring). The tension was converted to atmospheric pressure in mmHg. The x-axis represents the increments of stretch in millimeters (mm). The slope of tension (g) versus length (mm) was used as the measure on cardiac tensile strength.

279

2.4. Estimation of elastin and collagen The 100 mg of cardiac tissue was autoclave in double distilled water twice at 2288F for 3 h and supernatant was collected to remove collagenous material. The tissue then defatted by chloroform / methanol (2 / 1) and dried to a constant weight. The fat free and collagen-free tissue was dissolved in 0.1 mol / l NaOH at 1008C for 30 min to remove nonelastic proteins and to have maximum amount of major elastin specific cross-links, the desmosines [14]. After rehydration in Tris–Cl (pH 7.5) the samples were digested with thermolysin (10 mg / mg of tissue). The concentration of elastin was estimated by measuring desmosine using extinction coefficient at 320 nm of 7800 M 21 cm 21 [15]. Elastin content was calculated by multiplying desmosine levels by factor of 7.4, assuming that desmonsine constitute an average of 13.5% of elastin molecule [16]. Nonelastic (cartilage) tissue from same animal was used as reference. During collagen extraction samples were defatted as described above and hydrolyzed in 0.5 ml of 6 N HCl in vacuum at 1158C for 24 h. After reconstitution in 4 ml of double distilled water the resultant solution was used. The hydroxyproline was estimated by oxidative and Erlich methods by measuring absorption at 558 nm [17]. Hydroxyproline was used as the standard. Collagen content was calculated by multiplying hydroxyproline levels by the factor 7.46, assuming that hydroxyproline constitute an average of 13.4% of the collagen molecule [17].

2.5. Histology and morphometry The ventricular tissue section including endocardium were stained with van Gieson for elastin and with trichrome for collagen as described [18].

2.6. Treatment of endocardial rings by MMP-2 To determine the role of MMP-2 in cardiac dilatation and decreased tensile strength cardiac rings were incubated with 10 mM purified MMP-2 [19] for 30 min in 0.1 M phosphate buffered saline (pH 7.4) at 378C, prior to measurements. Cardiac TIMP-4 was purified as described [20]. Breakdown of elastin

280

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

fibers and collagen chain was confirmed by analysis of elastin- and collagen-degradation peptides on immuno-blot using anti-elastin and anti-collagen antibodies (Sigma, St. Louis, MO). To measure the effect of MMP-2, the rings form 32-week-old SHR were preincubated with antiMMP-2 antibody (Chemicon Corp) [1 / 200 dilution] for 30 min at 378C, prior to the addition of MMP-2. The specificity of antibody was confirmed by precipitating / inhibiting MMP-2 activity in zymographic gels. The preincubation of active MMP-2 with this antibody abolished zymographic activity. The length–tension measurements were performed on these rings.

3. Results

3.1. LV hypertrophy The heart / body ratio was: 0.003660.0002, 0.003860.0001, 0.003560.0003 and 0.007160.0001 in NWR8w, SHR8w, NWR32w and SHR32w, respectively, suggesting significant (P,0.01) LV hypertrophy in SHR at 32 weeks as compared to NWR. The LV and RV wall thickness and diameters were also increased in the SHR32wks as compared to NWR32wks (Fig. 1 and Table 1).

3.2. MMP-2 activity 2.7. Electrophoresis, zymography and immuno-blot analysis of ventricular tissue homogenates Cardiac tissue homogenates were prepared [21]. A Bio-Rad dye-binding assay was applied to estimate total protein concentration in the tissue extracts according to the method of Bradford [22]. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) was performed with and without the reduction following the method of Laemmli [23]. To normalize the intensity of MMP-2 activity with bactin, immuno-blot analysis of b-actin was performed [10]. The gelatinase A (MMP-2) activity was measured using quantitative elastin zymography [18]. The 50-mg (wet) tissue was analyzed for MMP-2 and actin Western blots. In which a linear standard plot between standard MMP-2 and lytic activity was generated. The amount of MMP-2 in samples was estimated from the standard plot.

To determine whether LV hypertrophy is associated with increased MMP activity, the levels of MMP-2 activity in hearts from SHR and NWR were measured by elastin zymography on cardiac tissue homogenates (Fig. 2). The MMP-2 activity was normalized with total actin. The MMP-2 activity / actin ratio was 2.060.5 in LV and 1.560.25 in RV of SHR; and 0.560.25 in LV and 0.2560.12 in RV of NWR (P,0.02 between SHR and NWR).

3.3. Cardiac rings To determine LV and RV length–tension relationship, separate LV and RV rings were prepared. The size of rings was 2.560.5 mm (Fig. 1). The septum was kept common in both types of rings. This created a reproducible model in which entire transmyocardial wall contributes to cardiac stretch and tension.

3.4. Cardiac distensibilty and response to CaCl2 2.8. Statistical analysis Animals used in each groups were: NWR8w56; SHR8w56; NWR32w56; SHR32w56. Two LV and one RV rings were prepared from each heart. All reported values were means from all animals in each group. Student’s t-test was employed to compare the differences between NWR and SHR of same age. ANOVA with post-hoc multiple comparison tests was used for comparing four groups. A value of P,0.05 was considered significant.

The tension was directly proportional to the stretch generated on the ring. At the resting tension the viability of the ring was determine by its response to CaCl 2 . The results suggested that stretched rings were quite active as they produced tension in response to CaCl 2 (Fig. 3).

3.5. Passive diastolic length–tension relation The segmental increase in ventricular ring produced linear increase in tension developed in the

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

281

Fig. 1. Cardiac ring preparation: Cross-sections through both ventricular wall and septum were prepared. Hearts were isolated from NWR and SHR. The left ventricular (LV) ring was prepared by removing the right ventricular wall. The right ventricular (RV) ring was prepared by removing the left ventricular wall. The septum was common in both the ventricles. The LV wall is thick in SHR 32w (A) as compared to NWR 32w (C). (B), RV from SHR 32w and (D), RV from NWR 32w .

cardiac wall (Fig. 4). The slope of pressure developed (mmHg) versus stretch (mm) was much greater in the ventricles of SHR32w as compared to age–sex matched control NWR32w. The slope was greater in the LV than RV. These data suggested that hearts of SHR32w were less elastic and therefore less resistance to stretch as compared to NWR32w. The RV was more elastic than LV.

To determine whether cardiac function depends on the elastin / collagen ratio, we analyzed elastin / collagen content (Table 2) of the same ring, which was used to measure tensile function. To create situation analogous to heart failure and to determine whether further activation of MMP-2 in SHR32w and NWR32w decreases cardiac tensile strength, we treated cardiac rings with purified MMP-2 and mea-

Table 1 Heart and body weight (g), LV and RV wall thickness (mm), LV and RV diameters (LVD, RVD): SHR and their NWR control of age and sex matched control were sacrifice under anesthesia Animal

n

Body weight (g)

Heart (g)

LV wall (mm)

RV wall (mm)

LVD (mm)

RVD (mm)

NWR 8w SHR 8w NWR 32w SHR 32w

6 6 6 6

260620 275615 366610 40668

0.9560.25 1.0560.25 1.2860.15 1.8660.10

2.060.5 2.1560.25 2.2560.5 3.7560.25 a

0.9060.25 1.060.25 1.4560.15 1.8560.10

1062 1464 1362 1663 a

661 1165 662 962

The heart was removed and weighed. The LV and RV rings were prepared and wall thickness and LV and RV diameters were measured using a digital micrometer under 1.53 magnifying dissecting microscope. Mean6S.D. are reported. a P,0.01 when compared to NWR.

282

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

Fig. 2. Zymographic analysis of elastinolytic gelatinase (MMP-2) activity in LV (lane 1) and RV (lane 2) of SHR 32 weeks; and LV (lane 3) and RV (lane 4) of NWR 32 weeks: Rings after contractile measurements were homogenized and tissue extracts were prepared. The elestinolytic activity of MMP-2 was determined by elastin zymography. Elastin gels were prepared as described in Materials and methods. Same tissue was analyzed for b-actin by Western blot. The scanned ratio of MMP-2 /Actin is reported at the top of the gel.

Fig. 3. Protocol of cardiac function: representative tracing of the length– tension recordings of ventricular rings: The LV rings were prepared from NWR. After stable tension (g), the rings were stretched by a micrometer. After maximum stretch and stable tension, the ring was relaxed to resting tension. To validate the viability of cardiac rings, CaCl 2 (20 mM) was added (arrow).

Fig. 4. Passive diastolic function: LV (*) and RV (m) from SHR; LV (j) and RV (d) from NWR: After in vitro stabilization, each cardiac ring was systematically stretched to the optimum of its length–active tension relation. A known amount of stretch (mm) was placed on the ring, and the tension was measured. The tension in gram was converted to dynes normalized with area of the ring and converted to mmHg. Each point is an average of six rings from six different animals.

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286 Table 2 Cardiac collagen and elastin content in SHR and NWR: SHR and their normotensive control rats of same age and sex were sacrifice under anesthesia Animal

NWR 8w SHR 8w NWR 32w SHR 32w

n

6 6 6 6

Collagen (mg / mg of tissue)

Elastin (mg / mg of tissue)

LV

RV

LV

RV

4.1060.23 5.0460.30 3.4060.15 7.0360.25 a

4.9060.20 5.3660.32 6.0060.23 3.1660.17

0.4360.04 0.5360.03 0.4160.05 0.2260.03 a

0.5260.03 0.5660.04 0.7360.02 0.1560.02

The collagen and elastin contents were measured biochemically as described in the Materials and methods section. The values are reported mean6S.D. a P,0.01 when compared with NWR.

sured cardiac tensile strength. Although MMP-2 treated rings demonstrated negative correlation between elastin / collagen ratio and cardiac tensile strength, the magnitude of negative correlation was lower in the MMP-2 treated group than untreated group (Fig. 5). The results suggested that the data lie along two lines, one with and one without MMP treatment. Apparently, the slope of the line from MMP treated rings was lower than untreated rings (Fig. 5).

3.6. Attenuation of MMP-2 driven cardiac dysfunction by anti-MMP-2 antibody

283

determine whether the activation of MMP-2 decreases cardiac function, we treated cardiac rings with anti-MMP-2 antibody prior to the treatment of MMP-2. Fig. 6 demonstrates significant decrease in tensile strength by MMP-2. Incubation with antiMMP-2 antibody prior to MMP-2 treatment was associated with significant improvement in cardiac tensile function versus untreated group. The results suggested that further MMP-2 treatment deteriorated the cardiac tensile strength and inhibition of MMP-2 by anti-MMP-2 antibody reverses the decrease in tensile strength (Fig. 6). Similar results were obtained when rings were treated with purified cardiac TIMP-4 in place of anti-MMP-2 antibody.

3.7. Histology Representative histological sections of the left ventricular interstitium of NWR32w and SHR32w were strained with van Gieson for elastin and trichrome for collagen (Fig. 7). The results suggested that NWR32w have more interstitial elastin (i.e. bluish–black staining) than SHR32w. The collagen staining suggests that SHR32w contains more collagen and myocytes are hypertrophied in SHR32w than NWR32w. These data are in agreement with our biochemical data on elastin and collagen content. The

To establish a cause and effect relationship and to

Fig. 5. Relationship between total elastin / collagen ratio and the slope of the length–tension curve generated for each ring to the resulting data. The x-axis is multiplied by 10. The tension (g) normalized with length (mm) represents the myocardial tensile strength of the ventricle from SHR 6 – 8w , SHR 32w ,NWR 6 – 8w , NWR 32w and rings treated with active purified MMP2. Data are mean6S.D.; six to eight cardiac rings per group were used. The treatment with MMP-2 decreases the gram / mm (tensile strength) of the ring in both SHR and NWR.

Fig. 6. Effect of MMP-2 and MMP-2 inhibition by antibody on cardiac ring passive diastolic length–tension relationship: The rings from 32week-old SHR were prepared. Two groups of rings were pretreated with antiMMP-2 antibody (1 / 200 dilution) for 30 min. Control, no treatment; control1anti, ring treated with antibody; MMP-2, in which MMP-2 was added prior to the measurement; MMP-21anti, the rings were incubated with MMP-2 antibody prior to the addition of MMP-2. The tension (gram / mm) versus the treatment of rings is plotted. Data was mean6S.D.; n56 cardiac LV rings per group. * P,0.01 when compared to control1anti; ** P,0.02 when compared to MMP-2 treatment.

284

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

Fig. 7. Morphometric analysis of cross-sections from LV obtained from SHR 32-week-old (B, E); NWR 32-week-old (A, D); and SHR 32-week-old treated with MMP-2 (C, F): Trichrome stain was used for collagen labelling and van Gieson stain was used for elastin labeling.

treatment with MMP-2 may degrade elastin and collagen and decrease cardiac tensile strength.

4. Discussion The mechanism of decreased cardiac tensile strength during the development of heart failure is not well understood. We demonstrated that increased MMP-2 activity and decreased elastin / collagen ratio is one of the causes of decreased ventricular tensile strength during the transition from compensatory hypertrophy to decompensatory heart failure. Similar changes were observed in RV and LV of SHR, but because the hypertrophy is primarily associated with LV dysfunction, the levels of changes in LV were much greater than RV. The magnitude of tensile strength was much less in NWR than in SHR. The association between LV hypertrophy (Fig. 1) and increased MMP-2 activity (Fig. 2) suggests that cardiac muscle remodels surrounding ECM by in-

creasing MMP-2 activity during the development of hypertrophy. Previous studies have demonstrated that drugs (captopril, lisinopril and quinapril) that reduce cardiac hypertrophy also decrease MMP activity [24]. Nitric oxide (NO), that inhibits cell proliferation [25] and reduces blood pressure, dwindled in end-stage heart failure [26]. In vivo decreased NO production is associated with increased MMP activity [27]. In vivo treatment of SHR with L-arginine (a substrate for NO production) reduces cardiac mass but not blood pressure [28]. Collectively these studies suggested a role of MMP activity in LV hypertrophy in SHR. In SHR and NWR, Bing et al. [11] demonstrated using isolated perfused heart (Langendorff preparation) that at a given ventricle volume, peak systolic pressure was greater in the hearts of SHR than age–sex matched control NWR. Also, the slope of the pressure–volume relationship was greater in SHR than NWR [29]. This suggested that there is reduced elastic compliance in SHR heart than NWR. These authors also demonstrated that the pressure–volume

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

slope was significantly decreased below NWR in SHR with heart failure [11,29]. We have demonstrated that in SHR with heart failure there is robust increase in MMP-2 activity [10]. In the studies of whole heart preparation, however, it is difficult to know the differences between SHR and NWR are not in fact caused by a deficit in oxygen diffusion which may be caused by the thickness of the wall. Also, to determine what happens in the entire transmural myocardial wall as well as to separate the effect of LV from RV, we developed a new method in which entire transmural myocardial wall contributes to cardiac contraction and the effect of LV can be separated from RV. In this preparation the continuous oxygenation prevented anoxic injury to the muscle [12]. The passive diastolic function of cardiac muscle is dependent upon the composition and concentration of ECM surrounding the cardiomyocyte [30]. We have demonstrated that in SHR having developed hypertension and fibrosis, LV is less resistance to stretch than RV (Fig. 4). Previous studies have demonstrated increased myocardial stiffness in the aging SHR, which was associated with decreased shortening velocity in the isolated myocytes and in the papillary muscle preparation [31]. A similar relationship between stiffness and shortening velocity with myocardial hypertrophy was observed by Mirsky et al. [32]. Recently, we have demonstrated that elastin / collagen ratio was linearly related to cardiac performance, (2dP/ dt) / mean arterial pressure [10]. Here, we demonstrate that the treatment of cardiac rings with MMP-2 in SHR decreases pressure–length slope (Fig. 5) analogous to what has been observed by Bing et al. [11] in SHR with heart failure. Our results demonstrated a relationship between the cardiac elastin / collagen ratio and myocardial tensile function (Fig. 5). The degradative changes in the ECM (ie. alteration in the composition and concentration of ECM) were associated with altered pressure–volume relationship in the myocardium [33]. Our experiments support that breakdown of active (i.e. ultrastructural, non-oxidative) collagen and elastin fibers by MMP-2 leads to reduced tensile strength (Figs. 5 and 6) which was holding the cardiac muscle in proper orientation and function. This may also suggest that disruption of cardiac wall connective tissue network lead to disorientation of the

285

cardiomyocyte during systole. It is known, however, that proteinases do disrupt intracellular myofibrillar structures [34] and the treatment of cardiac tissue by collagenase (MMP-2) leads to myocyte dispersion, cell swelling [35], and apoptosis [20]. Consistent with our biochemical and functional data, the morphometric analysis of the myocardium demonstrated increased fibrosis in SHR (i.e. decreased elastin content and increased collagen) [Fig. 7]. The degradation of ECM by MMP-2 is in part responsible for reduced systolic function in congestive heart failure.

4.1. Perspective In conclusion, this study demonstrates that activation of elastinolytic MMP-2 might be one of the causes of reduced elastic compliance in the hypertrophic heart. We have developed an ex vivo method in which separate myocardial LV and RV contractile function can be evaluated. This method can be applied to other studies, which sought to evaluate separate LV and RV function, as well as distinct sites in the same heart such as myocardial infarction (MI) and non-MI regions.

Acknowledgements The authors greatly appreciate the generous help of Dr Indria Rao and Ms Yolanda Smith in histological staining. This work was supported in part by NIH grants GM-48595, HL-51971, and AHA Mississippi Affiliate, and Kidney Care Foundation of Mississippi.

References [1] Pouleur HG, Konstam MA, Udelson JE, Rouseau MF. Changes in ventricular volume, wall thickness and wall stress during progression of left ventricular dysfunction (SOLVD). J Am Coll Cardiol 1993;22:43A–8A. [2] Conrad CH, Brooks WW, Robinson KG, Bing OHL. Impaired myocardial function in the SHR with heart failure. Am J Physiol 1991;260:H136–45. [3] Trippodo NC, Frohlich ED. Similarities of genetic (spontaneous) hypertension: man and rat. Circ Res 1981;48:309–19. [4] Weber KT, Sun Y, Tyagi SC, Cleutjens JPM. Collagen network of the myocardium: functional, structural remodeling and regulatory mechanisms. J Mol Cell Cardiol 1994;26:279–92.

286

V.S. Mujumdar et al. / International Journal of Cardiology 79 (2001) 277 – 286

[5] Conrad CH, Brooks WW, Hayes JA, Sen S, Robinson KG, Bing OHL. Myocardial fibrosis and stiffness with hypertrophy and heart failure in spontaneously hypertensive rat. Circulation 1995;91:161– 70. [6] Senior RM, Griffin GL, Eliszar CJ, Shapiro SD, Goldberg GI, Welgus HG. Human 92- and 72-kilodalton type IV collagenases are elatases. J Biol Chem 1991;266:7870–4. [7] Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase. J Biol Chem 1995;270:5872–6. [8] Tyagi SC, Haas SJ, Kumar SG et al. Post-trancriptional regulation of extracellular matrix metalloproteinse in human heart end-stage failure secondary to ischemic cardiomyopathy. J Mol Cell Cardiol 1996;28:1415–28. [9] Tyagi SC. Dynamic role of extracellular matrix metalloproteinases in heart failure. Cardiovasc Pathol 1998;7:153–9. [10] Mujumdar VS, Tyagi SC. Temporal regulation of extracellular matrix components in transition from compensatory hypertrophy to decompensatory heart failure. J Hypertens 1999;17:261–70. [11] Bing OH, Brooks WW, Robinson KG et al. The SHR as a model of the transition from compensated left ventricular hypertrophy to failure. J Mol Cell Cardiol 1995;27:383–96. [12] Tyagi SC, Smiley LM, Mujumdar VS. Homocyst(e)ine impairs endocardial endothelial function. Can J Physiol Pharmacol 1999;77:950–7. [13] Tsukube T, McCully JD, Metz KR, Cook CU, Levitsky S. Amelioration of ischemic calcium overload correlates with high-energy phosphate in sebescent myocardium. Am J Physiol 1997;273:H418– 25. [14] Hodgkin DD, Gilbert RD, Roos PJ, Sandberg LB, Boucek RJ. Dietary lipid modulation of connective tissue matrix in rat abdominal aorta. Am J Physiol 1992;262:R389–94. [15] Tyagi SC, Simon SR. Hydrophobic binding sites of elastin-derived peptide on neutrophil elastase. Biochem Cell Biol 1994;72:419–27. [16] Paz MA, Keith DA, Gallop PM. Elastin isolation and crosslinking. Methods Enzymol 1982;82:571–87. [17] Chiariello M, Ambrosio G, Cappelli-Bigazzi M. A biochemical method for the quantitation of myocardial scarring after experimental coronary artery occlusion. J Mol Cell Cardiol 1986;18:283–90. [18] Tyagi SC, Meyer L, Schmaltz RA, Reddy HK, Voelker DJ. Proteinases and restenosis in human coronary artery: extracellular matrix production exceeds the expression of proteolytic activity. Atherosclerosis 1995;116:43–57. [19] Tyagi SC, Cleutjens JPM. Myocardial collagenase: purification and structural characterization. Can J Cardiol 1995;12:165–71. [20] Tummalapalli CM, Heath BJ, Tyagi SC. Tissue inhibitor of metalloproteinase-4 instigates apoptosis in transformed cardiac fibroblasts. J Cell Biochem 2001;80:512–21. [21] Tyagi SC, Campbell SE, Reddy HK, Tjahja E, Voelker DJ. Matrix metalloproteinase activity expression in infarcted, noninfarcted and dilated cardiomyopathic human hearts. Mol Cell Biochem 1996;155:13–21.

[22] Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248–54. [23] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. [24] Tyagi SC, Hayden MR, Hall JE. Role of angiotensin in angiogenesis and cardiac fibrosis in heart failure. In: Dhalla NS, Zahradka P, Dixon I, Beamish R, editors, Progress in Experimental Cardiology: Angiotensin II Receptor Blockade: Physiological and Clinical Implications, vol. 2, Boston, MA: Kluwer Academic Publishers, 1998, pp. 537–49. [25] Sharma RV, Tan E, Fang S, Gurjar MV, Bhalla RC. NOS gene transfer inhibits expression of cell cycle regulatory molecules in vascular smooth muscle cells. Am J Physiol 1999;276:H1450–1459. [26] Malmsjo M, Bergdahl A, Zhao XH et al. Endothelial dysfunction in chronic myocardial infarction despite increased vascular eNOS and soluble guanylate cyclase expression:role of enhanced vascular superoxide production. Circulation 1999;100:292–8. [27] Radomski A, Sawicki G, Olson DM, Radomski MW. The role of nitric oxide and metalloproteinases in the pathogenesis of hyperoxiainduced lung injury in newborn rats. Br J Pharmacol 1998;125:1455–62. [28] Matsuoka H, Nakat M, Keisuke K et al. Chronic L-arginine administration attenuates cardiac hypertrophy in SHR. Hypertension 1996;27:14–8. [29] Schraeger JA, Canby CA, Rongish BJ, Kawai M, Tomanek RJ. Normal LV diastolic compliance after regression of hypertrophy. J Cardiovasc Pharmacol 1994;23:349–57. [30] Burgess ML, Buggy J, Price RL, Abel FL, Terracio L, Samarel AM, Borg TK. Exercise- and hypertension-induced collagen changes are related to left ventricular function in rat heart. Am J Physiol 1996;270:H151–9. [31] Brooks WW, Bing OHL, Robinson KG, Slawsky MT, Chaletsky DM, Conrad CH. Effect of ACE-inhibition on myocardial fibrosis and function in hypertrophied and failing myocardium from SHR. Circulation 1997;96:4002–10. [32] Mirsky I, Pfeffer JM, Pfeffer MA. Mechanical properties of normal and hypertrophied myocardium: is there a relationship between diastolic and systolic function? Cardiovasc Res 1983;7:39–55. [33] O’Brien LJ, Moore CM. Connective tissue degradation and distensibility characteristics of the non-living heart. Experientia 1966;22(12):845–7. [34] Taylor RG, Tassy C, Briand M, Robert N, Briand Y, Ouali A. Proteolytic activity of proteasome on myofibrillar structures. Mol Biol Rep 1995;21:71–3. [35] Flegler-Balon C, Behrendt H. Effects of calcium (21) deficiency, collagenase, and mechanical dispersion on the ultrastructure of cardiac myocyes of adult rats. Eur J Cell Biol 1982;27:262–9.