Mechanisms underlying cardiac fibrogenesis in magnesium deficiency are unclear. It was reported earlier from this laboratory that serum from magnesium-deficient rats has a more pronounced stimulatory effect on cell proliferation, net collagen production, and superoxide generation in adult rat cardiac fibroblasts than serum from rats on the control diet. The profibrotic serum factors were, however, not identified. This study tested the hypothesis that circulating angiotensin II may modulate cardiac fibroblast activity in hypomagnesemic rats. Male Sprague-Dawley rats were pair-fed a magnesium-deficient (0.0008% Mg) or -sufficient (0.05%) diet for 6 days, and the effects of serum from these rats on [3H]thymidine and [3H]proline incorporation into cardiac fibroblasts from young adult rats were evaluated in the presence of losartan, an angiotensin II type 1 (AT1) receptor antagonist, and spironolactone, an aldosterone antagonist. Losartan and spironolactone markedly attenuated the stimulatory effects in vitro of serum from the magnesium-deficient and control groups, but the inhibitory effects were considerably higher in cells exposed to serum from magnesium-deficient animals. Circulating and cardiac tissue levels of angiotensin II were significantly elevated in magnesium-deficient animals (67.6% and 93.1%, respectively, vs. control). Plasma renin activity was 61.9% higher in magnesium-deficient rats, but serum angiotensin-converting enzyme activity was comparable in the two groups. Furthermore, preliminary experiments in vivo using enalapril supported a role for angiotensin II in magnesium deficiency. There was no significant difference between the groups in serum aldosterone levels. The findings suggest that circulating angiotensin II and aldosterone may stimulate fibroblast activity and contribute to a fibrogenic response in the heart in magnesium deficiency.
- cardiac fibroblasts
- renin-angiotensin-aldosterone system
- magnesium-deficient rats
magnesium plays an important role in maintaining the structural and functional integrity of the cardiovascular system (1, 2, 19). It has been known for a long time that magnesium deficiency produces a cardiomyopathy characterized by focal myocardial necrosis and fibrosis (3, 9). Many theories have been advanced to explain the molecular basis of myocardial damage associated with magnesium deficiency (22). A large body of experimental evidence suggests that increased oxidative stress in the setting of an immunoinflammatory reaction may produce cardiovascular injury in magnesium deficiency (30). Furthermore, alterations in collagen metabolism and fibroblast proliferation rates, reported earlier from this laboratory (12), point to the activation of cardiac fibroblasts in a rodent model of acute magnesium deficiency. Whereas the prooxidant agents that may contribute to increased oxidative stress in magnesium deficiency have been investigated (30), factors that modulate cardiac fibroblast activity and promote myocardial fibrosis in magnesium deficiency have received little attention.
We had demonstrated earlier that serum from magnesium-deficient rats (MgD) has a more pronounced stimulatory effect on cell proliferation, net collagen production, and superoxide generation in adult rat cardiac fibroblasts than serum from magnesium-sufficient rats (MgS). Superoxide was found to mediate the enhanced proliferative response of the cells (14). The profibrotic serum factors were, however, not identified. The present study was undertaken to identify the circulating factors that were earlier shown to activate cardiac fibroblasts in hypomagnesemic rats.
ANG II is an important regulator of cardiac fibroblast function and collagen turnover (6, 17, 28). Moreover, ANG II can trigger most of the changes reported to be associated with magnesium deficiency, including induction of prooxidant (24, 32) and proinflammatory conditions (8, 16). It stimulates the expression of intercellular adhesion molecule-1 by human vascular endothelial cells (16) and that of E-selectin by human coronary endothelial cells to promote leukocyte adhesion to these cells (8). Furthermore, it is a very potent inducer of vasospasm (26) and an important etiological factor in hypertension (10, 18). ANG II also promotes hypertrophy and hyperplasia of vascular cells (25). Based on these observations, Shivakumar (20) published a model of cardiovascular injury in magnesium deficiency that postulated a role for the renin-angiotensin system in the pathobiology of magnesium deficiency.
Against this backdrop, the major goals of the present study were 1) to test whether circulating ANG II and aldosterone activate cardiac fibroblasts and 2) to examine levels of expression of the components of the renin-angiotensin-aldosterone system (RAAS) in magnesium deficiency. Furthermore, preliminary experiments on the effect of orally administered enalapril on lipid peroxidation in the heart were also performed to confirm the involvement of ANG II in vivo in magnesium deficiency. This communication presents evidence for the first time that circulating ANG II and aldosterone may play a role in activating cardiac fibroblasts in magnesium deficiency. The findings provide useful insights into mechanisms of cardiac fibrogenesis in magnesium deficiency that remain unclear.
MATERIALS AND METHODS
All chemicals were from Sigma Chemical (St. Louis, MO). [3H]thymidine [specific (Sp) activity, 18 Ci/mmol] and [3H]proline (7,500 mCi/mmol) were obtained from Bhabha Atomic Research Center. (3-[125I]iodotyrosyl4)Sar1Ile8, ANG II (Sp activity, 2,000 Ci/mmol) was purchased from Amersham Biosciences. Animal feed, based on the nutrient requirements of rats proposed by the American Institute of Nutrition, was purchased from Zeigler Brothers (Gardners, PA). The feed for the control and test groups differed only in magnesium content and contained 50% cellulose-glucose, 20% casein lactate, 15% corn starch, 5% cellulose, 5% corn oil, 1% American Institute of Nutrition Vitamin Mix, 3.5% American Institute of Nutrition Mineral Mix, 0.3% Dimethionine, and 0.2% choline bitartrate. Magnesium-sufficient and magnesium-deficient diets contained 0.0515% and 0.0008% magnesium, respectively.
Animal care and use were approved by the Institutional Animal Ethics Committee and conformed to APS guidelines. Losartan potassium was a kind gift of Merck-DuPont.
Preparation of serum.
Magnesium deficiency was induced in rats as described earlier (12). Briefly, male Sprague-Dawley rats, weighing about 65–85 g at the start of the experiment, were pair-fed a magnesium-sufficient or magnesium-deficient diet for 6 days, and deionized water was provided ad libitum. Blood was collected from the descending aorta of anesthetized rats, and serum or plasma was separated by centrifugation and stored at −20°C. For experiments with cardiac fibroblasts, serum was complement inactivated at 56°C for 30 min and filtered through 0.22-μm membrane before storage at −20°C. Cardiac tissue was washed in PBS repeatedly to remove blood, frozen in liquid nitrogen, and stored at −80°C.
Isolation and culture of cardiac fibroblasts.
Ventricular fibroblasts were isolated, as described earlier (13), by enzymatic digestion of cardiac tissue from young adult rats. Cells from passages 2 or 3 were used for the experiments, and their fibroblastic nature was confirmed by immunocytochemistry using antibodies against vimentin, factor VIII, and desmin. The cultures were free of other contaminating cell types (≥99% purity). Cells were maintained in medium 199 containing 10% FBS.
Measurement of DNA synthesis.
DNA synthesis was measured in terms of [3H]thymidine incorporation into trichloroacetic acid-insoluble material, as described earlier (14). Briefly, subconfluent cultures, serum-deprived for 24 h, were exposed to 10% MgS or MgD in medium 199 for 24 h with 2.5 μCi/ml of [3H]thymidine in the presence and absence of losartan (1 μM) or spironolactone (0.1 μM) and were processed for determination of acid-precipitable radioactivity. Serum was replenished after 12 h.
Measurement of [3H]proline incorporation.
Confluent cultures were serum-deprived for 24 h and exposed to 10% MgS or MgD in DMEM for 24 h with 2 μCi/ml [3H]proline in the presence and absence of losartan (1 μM) or spironolactone (0.1 μM) and were processed for determination of acid-precipitable radioactivity. Serum was replenished after 12 h.
Measurement of plasma and cardiac tissue ANG II levels.
Levels of ANG II in rat plasma were determined, after extraction with ethanol, by radioimmunoassay using commercial kits (Euro-Diagnostica), as per the manufacturer's protocol. Cardiac tissue was homogenized in 0.25 M sucrose solution using an Ultra Turrax T8 homogenizer. ANG II was extracted using absolute ethanol and was dried under a stream of nitrogen. The extract was solubilized in the buffer provided with the kit and was used for the assay. The interassay variation was within acceptable limits.
Measurement of serum and cardiac tissue aldosterone levels.
Levels of aldosterone in rat serum and cardiac tissue homogenates were measured using kits from DSL, as per the manufacturer's protocol. The interassay variation was within acceptable limits.
Measurement of plasma renin activity.
Plasma renin activity was determined using kits from DiaSorin, by measuring ANG I generated from maleate by radioimmunoassay, as per the manufacturer's protocol. In this assay, the conversion of ANG I to ANG II was blocked by PMSF. The interassay variation was within acceptable limits.
Measurement of angiotensin-converting enzyme activity.
Angiotensin-converting enzyme (ACE) activity in rat serum was measured after conversion of its substrate Hip-His-Leu to His-Leu, which combines with the fluorescent dye, o-pthaldialdehyde, and fluorescence was measured using an excitation wavelength of 360 nm and an emission wavelength of 500 nm.
125I-labeled-ANG binding to cardiac fibroblasts.
The 125I-labeled ANG (125I-ANG) binding assay was carried out as described earlier (27). Briefly, confluent cultures of adult rat cardiac fibroblasts were exposed to 10% MgS or MgD for 24 h, washed with PBS twice, and then incubated with (3-[125I]iodotyrosyl4)Sar1Ile8, ANG II at 0.2 nM for 90 min in Tris·HCl binding buffer (pH 7.2) containing 0.25% BSA. Nonspecific binding was determined in the presence of a large excess of nonradioactive ligand and was subtracted from the total binding to obtain specific binding.
Measurement of levels of thiobarbituric acid-reactive substances in hearts.
Male rats weighing 65–85 g were divided into four groups. Two groups were pair-fed magnesium-sufficient and two groups magnesium-deficient diets for 28 days. One group each on the control and deficient diets was orally administered enalapril through drinking water at 10 mg/kg body wt throughout the duration of the experiment. After rats were euthanized, cardiac tissue was collected, homogenized in ice-cold physiological saline, and used for determination of thiobarbituric acid-reactive substances levels as described earlier (12).
All data are expressed as means (SD). To evaluate the effects of antagonists on MgS- and MgD-treated cells, pairwise comparisons were made by Student's t-test, and significance was determined at P ≤ 0.05. Data were also analyzed by two-way ANOVA with MgD/MgS, antagonist, and MgD/MgS × antagonist interaction terms, and P ≤ 0.05 was considered as statistically significant.
Pursuing the earlier observation that serum factors exert profibrotic effects in the rodent model of acute magnesium deficiency (14), we designed experiments to ascertain whether ANG II is the serum factor that enhances cardiac fibroblast proliferation. The effect of MgD and MgS on cardiac fibroblast proliferation was evaluated in the presence and absence of losartan (1 μM), the ANG II type 1 (AT1) receptor antagonist. Consistent with the earlier observation, MgD produced a significantly higher (>2-fold, P < 0.001) incorporation of [3H]thymidine into DNA than MgS (Fig. 1). Losartan decreased the mitogenic effect of MgS and MgD by ∼22% (P = 0.001) and 35% (P = 0.001), respectively.
As aldosterone is an important modulator of cardiac fibroblast function and its association with magnesium, and a proinflammatory and profibrotic cardiac phenotype, has been reported recently (7), the proliferation assay was repeated as described above in the presence and absence of 0.1 μM spironolactone, an aldosterone antagonist (Fig. 1). Spironolactone diminished the mitogenic effect of MgS and MgD by ∼18% (P = 0.016) and 48% (P = 0.001), respectively.
In another set of experiments, losartan was found to decrease the effects of MgS and MgD on [3H]proline incorporation by ∼10% (P = 0.078, not significant) and 21% (P = 0.018), respectively (Fig. 2). Spironolactone attenuated the effects of MgS and MgD on [3H]proline incorporation by ∼21% (P = 0.016) and 26% (P = 0.005), respectively (Fig. 2). Collagenase-sensitive counts were not determined in these experiments because MgD was earlier reported to augment collagen production to a greater extent than MgS (14) and because the primary objective of this study was to identify the serum factor. Moreover, because proline is more abundant in collagen than in noncollagen proteins, an increase in proline incorporation would reflect increased collagen synthesis.
Additional experiments examined whether significant amounts of ANG II in serum might be lost through degradation under culture conditions over extended periods of incubation so that the reported serum effects may not be due to ANG II. Such a possibility was, however, excluded by the observation that 24-h incubation of cells with ANG II (150 pM) in MgS produced a 37% increase (P = 0.003) in [3H]proline incorporation.
After the experiments with losartan and spironolactone, plasma/serum and cardiac tissue levels of ANG II and aldosterone were determined by radioimmunoassay. Plasma ANG II levels were 67.6% higher (P < 0.001) in the magnesium-deficient group (Table 1). There was a nearly twofold increase (P < 0.005) in the cardiac tissue levels of ANG II in the deficient group (Table 1).
As circulating ANG II levels were elevated in magnesium deficiency and because ANG II is known to upregulate the AT1 receptor (23), we looked for a possible increase in AT1 receptor levels in cardiac fibroblasts exposed to MgD. Figure 3A shows the kinetics of the binding of 125I-ANG II to cardiac fibroblasts. A small but statistically insignificant decrease of ∼24% in 125I-ANG II binding to cardiac fibroblasts exposed to MgD was observed (Fig. 3B), possibly due to the higher levels of ANG II in MgD competing with 125I-ANG II for the binding sites.
Serum and cardiac tissue levels of aldosterone were comparable in the two groups (Table 1). Plasma renin activity was 61.9% higher (P < 0.01) in magnesium-deficient rats (Table 1). There was no significant difference between the two groups in serum ACE activity (Table 1).
Further support for the role of ANG II in magnesium deficiency was obtained from preliminary in vivo experiments that showed that orally administered ACE inhibitor enalapril attenuates the increase in lipid peroxidation in the heart in magnesium deficiency (Fig. 4).
Several lines of evidence, from our laboratory and elsewhere (5, 11, 12, 21, 29–31), support the view that oxidative injury may contribute to the cardiac and vascular lesions of magnesium deficiency. The importance of prooxidant and proinflammatory events in the pathobiology of magnesium deficiency has been stressed by several laboratories based on investigations using the rodent model of acute magnesium deficiency. Marked elevations in the circulating levels of prooxidant and inflammatory mediators, such as IL-1, IL-6, and TNF-α, observed at the end of 2 wk on the magnesium-deficient diet, correlated temporally with the increasing occurrence of cardiomyopathic changes (30). Furthermore, there are conflicting reports (4, 30) on elevations in the circulating and cardiac tissue levels of the neuropeptides, substance P and calcitonin gene-related peptide, in the first week of acute magnesium deficiency that could play a role in cardiac lesion formation via a neurogenic inflammation mechanism (30).
Our studies have focused on mechanisms underlying cardiac fibrogenesis in magnesium deficiency. We had reported earlier that substance P might play a limited role in modulating cardiac fibroblast activity in magnesium deficiency. Involvement of factors other than substance P remained to be examined. The components of the RAAS were an obvious choice because they modulate cardiac fibroblast activity in pathological conditions (28). Furthermore, ANG II is known for its prooxidant (24, 32) and proinflammatory actions (8, 16). It seemed reasonable to hypothesize that it may mediate many of the changes reported to be associated with magnesium deficiency.
Using losartan and spironolactone, we demonstrated in this study the involvement of ANG II and aldosterone in mediating the profibrotic effects of MgD on cardiac fibroblasts. Losartan decreased [3H]thymidine incorporation in MgS- and MgD-treated cells by 22% and 35%, respectively (Fig. 1). Spironolactone, however, reduced [3H]thymidine incorporation in MgS- and MgD-treated cells by 18% and 48%, respectively (Fig. 1), which reflected a 2.7-fold increment in the percent inhibition. The data suggested that aldosterone may play a more marked role than ANG II in enhancing a fibroproliferative response in magnesium deficiency.
Losartan inhibited [3H]proline incorporation in MgS- and MgD-treated cells by 10% and 21%, respectively (Fig. 1). On the other hand, spironolactone diminished [3H]proline incorporation in MgS- and MgD-treated cells by 21% and 26%, respectively (Fig. 1). Thus, although the effect of aldosterone on [3H]proline incorporation appeared greater than that of ANG II in both MgS- and MgD-treated cells, the contribution of ANG II was enhanced to a greater extent in MgD-treated cells (21% in MgD vs. 10% in MgS). The 125I-ANG II binding data (Fig. 3) suggested that the more marked effects of MgD versus MgS on cells did not relate to alterations in ANG II receptor density.
Consistent with these observations, plasma and cardiac tissue levels of ANG II were significantly higher in magnesium-deficient animals (Table 1). However, the possibility that tissue ANG II levels may reflect the ANG II sequestered from blood cannot be ruled out. Confirmatory evidence of intracardiac activation of the renin-angiotensin system in magnesium deficiency would require measurement of angiotensinogen mRNA levels. As circulating levels of aldosterone were not elevated in magnesium deficiency (Table 1), the observed effects of spironolactone on cardiac fibroblasts (Table 1) may reflect synergistic effects of circulating aldosterone with other factors in the serum from the deficient group, possibly even ANG II. It may, however, be noted that a significant elevation in rat plasma aldosterone levels was reported by Laurent et al. (15) at 2 wk of severe dietary magnesium deficiency (0.096% Mg in the control vs. 0.008% Mg in the deficient group). We did not ascertain whether the circulating aldosterone titer increases as magnesium deficiency advances.
The significant increase in plasma renin activity in magnesium-deficient rats, with no increment in serum ACE activity, suggested that the elevation in ANG II levels in MgD may be due to increased angiotensinogen expression and/or increased renin activity and not due to increased conversion of ANG I to ANG II. Preliminary in vivo experiments demonstrating the myocardial effects of ACE inhibition in magnesium deficiency (Fig. 4) provided further evidence of the involvement of ANG II and strengthened the findings of this study.
Together, the findings presented here suggest that dietary deficiency of magnesium enhances ANG II levels and that circulating ANG II and aldosterone may stimulate cardiac fibroblast activity and contribute to a fibrogenic response in the heart. Future studies should clarify whether the alterations in the renin-ANG II system are a direct consequence of magnesium deficiency or are secondary to changes in other parameters, such as blood pressure and electrolyte balance.
This work was supported by a research grant (to K. Shivakumar) from the Indian Council of Medical Research. S. Sapna and S. K. Ranjith gratefully acknowledge financial support from the Department of Science and Technology and the Indian Council of Medical Research.
We thank Dr. P. Sankara Sarma for statistical evaluation of data, Dr. A. C. Fernandez for help in organizing the animal experiments with the help of S. Shaji and Abhilash Rajan, and Merck-DuPont for providing losartan potassium.
↵* S. Sapna and S. K. Ranjith contributed equally to this work.
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 © 2006 by the American Physiological Society