Decreased right as well as left ventricular function can be associated with pulmonary hypertension (PH). Numerous investigations have examined cardiac function following induction of pulmonary hypertension with monocrotaline (MCT) assuming that MCT has no direct cardiac effect. We tested this assumption by examining left ventricular function and histology of isolated and perfused hearts from MCT-treated rats. Experiments were performed on 50 male Sprague-Dawley rats [348 ± 6 g (SD)]. Thirty-seven rats received MCT (50 mg/kg sc; MCT group) while the remainder did not (Control group). Three weeks later, pulmonary artery pressure was assessed echocardiographically in 20 MCT and 8 Control rats. The hearts were then excised and perfused in the constant pressure Langendorff mode to determine peak left ventricular pressure (LVP), the peak instantaneous rate of pressure increase (+dP/dtmax) and decrease (−dP/dtmax), as well as the rate pressure product (RPP). Histological sections were subsequently examined. Pulmonary artery pressure was higher in the MCT-treated group compared with the Control group [12.9 ± 6 vs. 51 ± 35.3 mmHg (P < 0.01)]. Left ventricular systolic function and diastolic relaxation were decreased in the MCT group compared with the Control group (+dP/dtmax 4,178 ± 388 vs. 2,801 ± 503 mmHg/s, LVP 115 ± 11 vs. 83 ± 14 mmHg, RPP 33,688 ± 1,910 vs. 23,541 ± 3,858 beats·min−1·mmHg−1, −dP/dtmax −3,036 ± 247 vs. −2,091 ± 389 mmHg/s; P < 0.0001). The impairment of cardiac function was associated with myocarditis and coronary arteriolar medial thickening. Similarly depressed ventricular function and inflammatory infiltration was seen in 12 rats 7 days after MCT administration. Our findings appear unrelated to the degree of PH and indicate a direct cardiotoxic effect of MCT.
- cardiovascular physiology
- pyrrolizidine alkaloids
- coronary vessels
pulmonary hypertension represents a serious clinical situation that can occur either as a primary process or secondary to a variety of diseases, conditions, and agents (39). Regardless of etiology, all cases of pulmonary hypertension exhibit similar structural vascular remodeling resulting in elevated pulmonary vascular resistance (55). An important aspect of the management of patients with pulmonary hypertension concerns the effects of pulmonary hypertension on cardiac function. Right ventricular dysfunction in the setting of chronic pulmonary hypertension has been well described both clinically and experimentally (8, 33, 48). Decreases in left ventricular systolic and diastolic function have also been reported with pulmonary hypertension (2, 15, 37, 43). Several mechanisms have been proposed to explain such left ventricular effects, including alteration in left ventricular geometry due to leftward displacement of the interventricular septum, interstitial edema in the left ventricular wall, decrease in left ventricular preload, increase in myocardial stiffness, diastolic asynchrony of the apical and lateral walls, humoral effects originating in the right ventricle, and alterations in intracellular calcium handling (see references in 15, 43).
There have been numerous animal investigations of right and left ventricular function following induction of pulmonary hypertension with the toxin monocrotaline (MCT; 10–12, 24, 57). The changes in cardiac function following administration of MCT have been attributed to the effects of the associated pulmonary hypertension.
MCT, derived from the plant Crotalaria spectabilis, has been used extensively to produce pulmonary hypertension in rats (51). MCT is a pyrrolizidine alkaloid that is known to be pneumotoxic as well as hepatotoxic in mammalian species. A single dose of MCT in the rat results 14 days later in progressive and sustained pulmonary hypertension (41). The pathological features seen following MCT administration are similar to those evident in human primary and secondary pulmonary hypertension. They are characterized histologically by endothelial damage followed by pulmonary vascular remodeling including hypertrophy of medial smooth muscle cells in pulmonary arteries and arterioles and neomuscularization of nonmuscular distal pulmonary arteries (41, 50). In addition, diffuse lung interstitial mononuclear infiltration, hemorrhage, and edema are found in the lungs of rats treated with MCT or its bioactive derivative MCT pyrrole (6, 47). Hemodynamically there is evidence of elevated pulmonary vascular resistance, increased pulmonary artery pressure (PAP), and subsequent right ventricular hypertrophy (41, 47, 50). These changes are accompanied by increases in right ventricular systolic and diastolic pressures and ultimately by right ventricular failure (9, 19, 27, 52). The severity of MCT-induced pulmonary hypertension appears to be correlated with the extent of the structural changes of the pulmonary vasculature (41). The precise mechanism of MCT action on the pulmonary vascular bed is unknown.
It has been generally assumed that MCT has no effect on the myocardium or coronary vessels per se. Of interest in this regard is seldom-cited older work describing myocarditis and diffuse coronary artery changes following MCT administration in rats (3, 7). There is limited contemporary data on the direct cardiac effects of MCT. For this purpose, we assessed the left ventricular systolic function and diastolic relaxation in MCT-treated rats in the isolated perfused heart preparation. In addition, we conducted histological studies of the ventricular myocardium and intramural coronary arteries. Our findings indicate that there are direct deleterious cardiac effects of MCT.
METHODS AND MATERIALS
Animals and Experimental Protocol
Experiments were conducted on 50 male Sprague-Dawley rats, weighing 348 ± 6 g (Charles River, Montreal, Canada). Twenty-five rats were randomly selected to receive an injection of MCT (50 mg/kg) subcutaneously over the dorsum of the neck (MCT group). Thirteen rats did not receive any MCT and acted as controls (Control group). These two sets of animals were observed daily for 3 wk after assignment to the MCT or Control group for any signs or symptoms of cutaneous, respiratory, or cardiac problems. They were weighed weekly. At 3 wk, PAP was assessed by echocardiography and left ventricular function was studied by isolated heart perfusion. Histological studies were subsequently performed on these hearts as well as on hearts that did not undergo isolated perfusion. In 12 additional rats left ventricular function and cardiac histology were examined 7 days following MCT injection. The experimental protocol was approved by the McGill University Animal Care Committee.
MCT (Sigma) was dissolved in 1N HCl, neutralized with 1N NaOH, buffered to pH 7.38, and diluted with 0.9% saline to achieve a final concentration of 24 mg/ml.
After light sedation with an intraperitoneal injection of pentobarbital sodium (20 mg/kg) 20 MCT treated and 8 control rats underwent pulsed-wave Doppler transthoracic echocardiographic interrogation in the pulmonary artery with two-dimensional guidance using a commercially available echocardiography machine (Acuson 128XP) and a 7.5 MHz short-focus, phased-array transducer. Pulmonary artery acceleration time (AT) was measured from the onset to the peak velocity of forward systolic pulmonary artery flow. Peak systolic PAP was calculated using the equation: PAP = 137.2–3.3(AT) (30).
Isolated Heart Perfusion
The hearts of the 28 animals studied echocardiographically and those of the 12 rats treated with MCT 7 days earlier were isolated and perfused in the constant pressure Langendorff mode. The rats were anesthetized (total pentobarbital dose 50 mg/kg ip), their hearts exposed, chilled with cold saline, and cannulated in situ by way of the aorta. In the rats studied echocardiographically this was done immediately following PAP determination. The hearts were perfused at a constant pressure of 100 cmH2O with modified Krebs-Ringer buffer at 37°C consisting (in mM) of 119 NaCl, 4.8 KCl, 1.3 CaCl2, 1.2 KH2PO4, 1.2 MgSO4, 25 NaHCO3 as well as 15 glucose maintained at 37°C and aerated with 95% O2 + 5% CO2. Retrograde aortic flow was measured using an ultrasonic flow probe (Transonic). Electrocardiographic recordings were made from three fine needle electrodes inserted into the right ventricle and atria (Powerlab). A small balloon was inserted in the left ventricle by way of the mitral valve. This balloon was connected to a pressure transducer, its volume adjusted to achieve a diastolic pressure of 10 mmHg at the beginning of the experiment, and the volume was kept constant for the duration of the study. The natural resonant frequency (48 Hz) and the damping ratio (0.12) of our pressure recording system were determined (22) and found to be within the range required for accurate measurement of left ventricular pressure and its first derivative using fluid-filled catheters (18). The pressure signal, retrograde aortic flow signal, and electrocardiogram were recorded digitally using Powerlab instrumentation and software. Following 10 min of stabilization heart rate (HR), peak left ventricular pressure (LVP), the maximum rate of pressure increase (+dP/dtmax), the maximum rate of pressure decrease (−dP/dtmax), and the rate-pressure product (RPP; RPP = HR·LVP) were determined over five cardiac cycles. Completion of the entire procedure required no more than 15 min.
Following the isolated heart perfusion experiments the hearts were placed in 10% buffered formalin. Ten days later the hearts were transversally sectioned at the mid-level of the left ventrical or papillary muscles. Sixteen 5-μm-thick sections per heart were stained with hematoxylin and eosin (H&E) or Mason's trichrome stain. The right ventricular free wall, interventricular septum, and left ventricular posterior wall were examined by light microscopy on the H&E slides for myocytes, interstitium, intramural coronary arteries, and ventricular wall thickness. Intramural coronary artery area, lumen area, wall area, and arteriolar wall thickness were measured in five coronary arterioles of the right ventricular free wall, interventricular septum, and left ventricular free wall, respectively, per section from digitized images using ImageJ software (1). Mason's trichrome-stained slides were used to evaluate the extent of collagen deposition. The hearts of five control and five rats treated with MCT 21 days earlier which did not undergo isolated heart perfusion were also fixed, stained, and examined as described above.
All data are expressed as means ± SD. Comparison of mean values between the groups was performed using an independent t-test. Correlations were analyzed using Pearson's correlation coefficient. The null hypothesis of no effect was rejected at P < 0.05.
MCT-treated rats showed significantly less weight gain than the Control group over a period of 21 days (51 ± 10 vs. 90 ± 8 g). None of the animals showed signs of illness related to right heart failure such as dyspnea or peripheral edema and none exhibited ascites on direct examination of the abdominal cavity at the time of cardiac excision. There was no mortality in either group.
MCT administration caused pulmonary hypertension. Pulmonary arterial systolic pressure estimated from AT (see methods) was significantly higher in the MCT-treated group compared with the Control group [12.9 ± 6 vs. 51 ± 35.3 mmHg (P < 0.01); see Fig. 1]. In 7 of 20 of the MCT group PAP was <20 mmHg (10 ± 7), in 5 of 20, PAP was >20 mmHg and <60 mmHg (47 ± 7), while in 8 of 20, PAP was >60 mmHg (89 ± 9).
Isolated Heart Perfusion
The hearts of 17 of 20 MCT-treated and 8 of 8 control rats that underwent echocardiography were successfully excised and perfused. We were unable to obtain perfusion data in one animal with PAP <20 mmHg, one animal with PAP between 20 and 60 mmHg, and one animal with PAP >60 mmHg. There was no difference in heart rate between the MCT and Control groups (298 ± 37 vs. 295 ± 21 beats/min). Coronary flow was significantly less in the MCT group compared with the Control group (13.2 ± 1.7 vs. 15.9 ± 1.6 ml/min). Left ventricular systolic function was decreased in the MCT group compared with the Control group with +dP/dtmax, LVP, and RPP all significantly less in the MCT group compared with the Control group. Left ventricular diastolic relaxation was also decreased in the MCT rats with −dP/dtmax significantly less than in Control group (see Fig. 2). The depressed systolic function and diastolic relaxation were unrelated to the degree of preexisting pulmonary hypertension with no statistical correlation (see methods) among +dP/dtmax, −dP/dtmax, LVP, or RPP, and PAP level (see Fig. 3).
Ventricular wall thickness.
The left ventricular posterior wall, interventricular septum, and right ventricular free wall in the hearts of the isolated and perfused MCT-treated animals were significantly thicker than in the hearts of the isolated and perfused Control group (2.15 ± 0.2 vs. 1.7 ± 0.08 mm, P < 0.0005; 1.97 ± 0.19 vs. 1.72 ± 0.05 mm, P < 0.01; 0.85 ± 0.1 vs. 0.53 ± 0.06 mm, P < 0.0001).
Widening and distension of the interstitial space was seen diffusely and to the same extent throughout the myocardium of the isolated and perfused hearts of both groups as has been reported previously in the myocardium of hearts following perfusion with electrolyte solution (58). Examination of right ventricular free wall, interventricular septum, and left ventricular posterior wall in the Control group revealed normal myocytes without evidence of inflammatory cells or collagen deposition. In contrast, in MCT-treated rats, a diffuse interstitial inflammatory infiltration was evident throughout the myocardium (Fig. 4). Cell types seen included mononuclear cells (lymphocytes, monocytes, and macrophages) as well as occasional polymorphonuclear cells. In each section, there were six to eight scattered foci of dense mononuclear and polymorphonuclear cell accumulation predominantly in the left ventricular posterior wall. Myocytes in and around these foci showed degenerative changes, fragmentation, coagulative myocytolysis, and necrosis. There was no fibrosis or collagen deposition in these foci. In some hearts, there were small areas of hemorrhage separate from these necrotic sites. In three rats, we found features of subendocardial necrosis, coagulative myocytolysis with well-developed fibrosis, and extensive collagen deposition in the left ventricular free wall and left ventricular papillary muscles. There was no correlation between the extent or severity of inflammatory infiltration, necrosis, myocytolysis, or fibrosis and the severity of pulmonary hypertension or right ventricular free wall thickness.
Intramural coronary arterioles.
The coronary vessels of the Control group appeared normal. However, in the MCT-treated rats a diffuse intramural coronary arteriolar medial thickening and decreased coronary arteriolar luminal area were seen throughout the left ventricular posterior wall, interventricular septum, and right ventricular free wall (Fig. 5). The external diameter of the intramural coronary arteries and their overall area were similar in the Control group and MCT group. However, the internal diameter in the MCT group was significantly less than in the Control group. Both arterial wall thickness and area in the MCT group were significantly greater than that in the Control group and the luminal area in the MCT group was markedly less than in the Control group (see Table 1). Examination of the media of the intramural coronary arteries revealed enlarged and hypertrophied smooth muscle cells. The nuclei of these cells were large with one to two nucleoli and extended far into the intima, sometimes extending all the way to the adventitia. In some cases there was extensive disarray of the smooth muscle cells in the media with intercellular fibrosis and collagen deposition. Similar changes were seen in small conductive arteries (>0.170 mm). There was no correlation between these changes and severity of pulmonary hypertension or right ventricular free wall thickness.
Histology without prior isolated heart perfusion.
Histological examination of the hearts of five MCT-treated rats that did not undergo isolated perfusion revealed the same inflammatory and degenerative myocardial changes as well as medial arteriolar thickening as described above without widening and distension of the interstitial space. The myocardium and coronary vessels of five control hearts that did not undergo isolated perfusion were normal.
Left Ventricular Performance and Histology 7 Days After MCT Administration
Similar changes in left ventricular function and myocardial histology were seen in the hearts of the 12 rats treated with MCT 7 days earlier compared with those treated 21 days earlier. Of note is that previous work showed that PAP does not increase until at least 10 days following MCT administration in the rat (21, 25). Left ventricular systolic function and diastolic relaxation were significantly decreased in the MCT-treated group compared with the Control group (LVP 94 ± 22 vs. 115 ± 11 mmHg, +dP/dtmax 3,044 ± 713 vs. 4,178 ± 388 mmHg/s, RPP 20,790 ± 4,621 vs. 33,688 ± 1,910 beats·min−1·mmHg−1, −dP/dtmax 1,613 ± 414 vs. 3,035 ± 247 mmHg/s, P < 0.05). Interstitial inflammatory infiltration and myocyte degeneration were seen as is described above 21 days following MCT administration (Fig. 6). In contrast to our findings 21 days following MCT administration the coronary arterioles appeared normal without evident medial thickening or decreased luminal area compared with the Control hearts (0.007 ± 0.002 vs. 0.006 ± 0.001 mm, 0.004 ± 0.002 vs. 0.005 ± 0.002 mm2).
We found an impairment of left ventricular systolic function and diastolic relaxation in isolated and perfused hearts of adult rats treated with MCT. This impairment of cardiac function is associated with a diffuse myocarditis as well as coronary arteriolar medial thickening. The effects of MCT on cardiac function as well as myocardial and coronary arteriolar histology appear to be independent of the pulmonary hypertension caused by MCT administration. Our results indicate that MCT administration has a significant direct effect on the heart.
There have been previous reports concerning cardiac histological changes following prolonged ingestion of Crotalaria spectabilis or its derivative MCT by rats. Blaustein et al. (7) noted that prolonged ingestion of MCT by adult Wistar rats resulted in coronary angiopathy and myocarditis. These authors describe lipid infiltration of the wall of right and left ventricular coronary arteries with some animals exhibiting subintimal or intimal coronary arterial hyperplasia. In addition, they show a histological figure demonstrating extensive myocardial destruction with infiltration of monocytes and lymphocytes. It is not clear whether the latter section was taken from the right or left ventricle. The presence of such myocardial changes in the right ventricle could be explained as a consequence of myocardial injury resulting from pulmonary hypertension. It is more difficult to attribute such changes in the left ventricle to pulmonary hypertension. Allen and Carstens (3) describe the effects of prolonged ingestion by adult Sprague-Dawley rats of Crotalaria spectabilis on various organs as well as on multiple levels of the systemic and pulmonary vasculatures. In particular, they describe myocardial inflammation, degeneration, and fibrosis as well as coronary arterial endothelial fragmentation and medial disruption “particularly in the right ventricle.” More recent investigations in the rat noted right ventricular myocardial inflammation, degeneration, and fibrosis between 2 and 5 wk following a single dose of subcutaneous MCT has been attributed to myocardial injury consequent to pulmonary hypertension (13, 26, 27, 44). In these newer reports, coronary vessel morphology is commented on only in the paper of Hirokawa et al. (26) and left ventricular myocardial histology is specifically commented on only in the work of Chen et al. (13). In contrast to earlier work and our own findings, Hirokawa et al. (26) found no noticeable changes in the intramural coronary arteries of MCT-treated rats while Chen et al. (13) specifically state that no left ventricular changes were seen after MCT administration. The reason for these discrepancies is unclear.
The cardiac histological changes that we have seen 21 days following a single dose of MCT share features with those that have been described in the lung. The latter have been extensively reviewed by Schultze and Roth (51). MCT administration causes pulmonary parenchymal edema and inflammatory cell infiltration. In addition, a diffuse pulmonary arterial angiopathy is particularly evident in smaller vessels following MCT administration. An important feature of this angiopathy is medial hypertrophy, hyperplasia, and collagen deposition. Of interest is that the increased pulmonary arterial medial wall thickness is not evident in the first week following MCT administration as we have observed here for the coronary arterioles. The precise mechanisms producing these changes are not known.
We found increased thickness of the right ventricular free wall in rats treated with MCT. This is expected given the pulmonary hypertension produced by such treatment and is consistent with previous investigations (50). In addition we found increased interventricular septal and left ventricular posterior wall thickness in the MCT-treated rats. Of note is that several prior animal studies demonstrated septal as well as left ventricular posterior wall hypertrophy associated with right ventricular hypertension produced by pulmonary arterial banding or emphysema (4, 5, 23, 34, 35, 54).
Coronary arteriolar medial muscle thickness was increased and luminal area was decreased on histological examination of the myocardium of the rats treated with MCT 21 days earlier. These structural effects as well as MCT-induced depression of endothelial-dependent relaxation (45) may have increased coronary arteriolar vasoreactivity and vascular resistance. It is difficult to precisely correlate coronary artery flow with coronary arteriolar cross-sectional area in fixed histological specimens. However, in our experiments, coronary flow at the same perfusion pressure was decreased in the isolated hearts of the MCT-treated animals compared with the Control group. This effect may have been more marked in vivo and have produced myocardial ischemia leading to some of the myocardial fibrosis observed.
As noted in the Introduction, decreases in left ventricular function have been described in association with pulmonary hypertension. However, it appears that the decrease in left ventricular diastolic relaxation and systolic function that we observed was independent of increases in PAP and was more likely due to a direct effect of MCT reflected in the myocardial and coronary artery changes evident histologically. Several findings support this contention. The decrease in left ventricular function in the hearts of rats treated with MCT 21 days earlier was independent of the preexisting pulmonary arterial pressure and was evident even in the hearts of animals without pulmonary hypertension or an increase in RV wall thickness. In addition, we observed a decrease in left ventricular function and left ventricular inflamatory changes as early as 7 days following MCT administration. In this regard previous investigators showed that pulmonary hypertension does not develop until after 10 days following MCT administration to rats (21, 25). Finally, the possibility of a mechanical interaction on the left ventricle of a hypertensive right ventricle was avoided as left ventricular function was measured in an isolated heart model with an empty right ventricle.
Our finding of decreased left ventricular systolic function and relaxation following MCT administration is supported by the recently published work of Lourenco et al. (36). These authors noted lower left ventricular systolic pressure and dP/dt as well as a longer time constant of isovolumetric relaxation in anesthetized rats 6 wk after MCT administration compared with untreated animals. Although the latter findings could be attributed to the severe pulmonary hypertension in their animals, they also demonstrated a negative force-frequency relationship of isolated left ventricular muscle strips, indicating an inherently reduced contractile reserve. In contrast to our own work, these authors observed no inflammatory changes or increased fibrosis of the left ventricular myocardium. This discrepancy may be due to the small number of MCT-treated hearts examined histologically (n = 4) by these authors as well as the survival of only less effected rats to 6 wk in their work. It is also possible that the absence of inflammatory changes in their work reflects the longer postexposure period of 6 wk that could have allowed for resolution of such changes as is seen in viral myocarditis (20). These authors also found an increase in endothelin-1 mRNA in the left ventricular myocardium of MCT-treated rats with an improvement of the force-frequency relationship of in vitro muscle strips treated with an endothelin blocker. This is not contradictory to our own work, as increased endothelin-1 expression as well as improved contractile function with endothelin antagonism have been demonstrated in the failing left ventricle (42).
The mechanisms of MCT action are not entirely clear. It is known that MCT is metabolized in the liver to bioactive pyrrollic derivatives, including MCT pyrrole by P-450 3A (32). Following a single dose of subcutaneous MCT, significant concentrations of MCT and/or its derivatives are found in various tissues including the liver, lungs, heart, and kidneys (17). MCT pyrrole has been shown to interfere with normal DNA and protein synthesis (28). Whether these are the cause of the endothelial and inflammatory effects of MCT is not known. Investigation of the time course over which MCT produces pulmonary hypertension indicates that both in anesthetized (21) and awake rats (25), PAP remains unchanged until at least 10 days following MCT administration with a gradual increase thereafter. Direct measurement of PAP in anesthetized rats 21 days following MCT administration by various investigators (14, 25, 40) has shown an elevation in systolic PAP to a similar degree and with a comparable variability as we found echocardiographically in lightly sedated rats at the same interval following MCT administration.
As outlined in the Introduction, numerous studies used the rat MCT model to study cardiac function in the setting of pulmonary hypertension. Our work indicates that this model may not be suited to such investigation. While alternative animal models exist they each have their own limitations and disadvantages. A number of investigators have created right ventricular hypertension by placement of a pulmonary artery band in pigs, dogs, sheep, rabbits, and rats (16, 29, 35, 46, 59). However, this method requires operative intervention, and the degree of right ventricular hypertension can be variable. Chronic hypoxia can be used to cause pulmonary hypertension in various mammals (53). However, chronic hypoxia also has effects on cardiac function. For instance, chronic hypoxia has been shown to decrease left ventricular adrenergic receptor content (31, 38) while we recently showed that transient hypoxic exposure in early life decreases the left ventricular response to inotropic stimulation at maturity (49). Rodman and coworkers (56) reported the development of a transgenic mouse in which the postnatal activation of a bone morphogenic protein receptor (BMPRII) mutation results in increased pulmonary artery muscularization and right ventricular hypertrophy. Although this transgenic model is of interest, it remains to be established that there are no associated cardiac effects of the postnatal activation of this mutation.
In conclusion, we found an impairment of left ventricular systolic function and diastolic relaxation associated with a diffuse myocarditis as well as coronary arteriolar medial thickening that are independent of the severity of pulmonary hypertension and that likely results from a direct toxic effect of MCT on the myocardium. Our results call into question the use of the MCT-induced pulmonary hypertension model to study the cardiac consequences of pulmonary hypertension.
This work was supported by an operating grant from the Canadian Institutes for Health Research (to C. V. Rohlicek) and a fellowship from the Montreal Children's Hospital Research Institute Fellowship (to E. L. Seifert).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society