Diabetic cardiomyopathy is characterized by reduced cardiac contractility independent of vascular disease. A contributor to contractile dysfunction in the diabetic heart is impaired sarcoplasmic reticulum function with reduced sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2a) pump activity, leading to disturbed intracellular calcium handling. It is currently unclear whether increasing SERCA2a activity in hearts with existing diabetic cardiomyopathy could still improve calcium flux and contractile performance. To test this hypothesis, we generated a cardiac-specific tetracycline-inducible double transgenic mouse, which allows for doxycycline (DOX)-based inducible SERCA2a expression in which DOX exposure turns on SERCA2a expression. Isolated cardiomyocytes and Langendorff perfused hearts from streptozotocin-induced diabetic mice were studied. Our results show that total SERCA2a protein levels were decreased in the diabetic mice by 60% compared with control. SERCA2a increased above control values in the diabetic mice after DOX. Dysfunctional contractility in the diabetic cardiomyocyte was restored to normal by induction of SERCA2a expression. Calcium transients from diabetic cardiomyocytes showed a delayed rate of diastolic calcium decay of 66%, which was reverted toward normal after SERCA2a expression induced by DOX. Global cardiac function assessed in the diabetic perfused heart showed diminished left ventricular pressure, rate of contraction, and relaxation. These parameters were returned to control values by SERCA2a expression. In conclusion, we have used mice allowing for inducible expression of SERCA2a and could demonstrate that increased expression of SERCA2a leads to improved cardiac function in mice with an already established diabetic cardiomyopathy in absence of detrimental effects.
- transgenic mice
- cardiac myocytes
congestive heart failure is an important medical problem resulting in significant morbidity and mortality. Heart failure occurs at an increased incidence in patients with diabetes mellitus. In addition to an increased propensity for coronary vascular disease, resulting in ischemic heart disease, diabetic cardiomyopathy occurs in combination with or independent of coronary vascular disease (12). In the diabetic heart, abnormal Ca2+ handling during the contractile cycle results in a decreased upstroke phase of the Ca2+ transient due to diminished release of Ca2+ from the sarcoplasmic reticulum (SR) by ryanodine receptor (RyR2) (2, 17). In addition, the diastolic decline of the Ca2+ transient is diminished due to reduced activity of the sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2a) pump (5).
Most functional studies indicate decreased SERCA2a activity and protein level in diabetic cardiomyopathy (10, 18). Efforts have focused on increasing expression of SERCA2a to alleviate contractile dysfunction in animal models of diabetic cardiomyopathy. Previous studies have shown that constitutive SERCA2a overexpression, with elevated SERCA2a levels before the introduction of diabetes-induced cardiomyopathy, prevented contractile abnormalities and calcium handling abnormalities in transgenic mice (18) and rats (20). These studies show that “prophylactic” SERCA2a overexpression in a transgenic model is able to preserve cardiac function and myocyte contractile performance in diabetic animals. It remains unclear, however, whether an established diabetic cardiomyopathy with diminished contractile function and abnormal Ca2+ handling can be reverted toward normal by inducible expression of SERCA2a in cardiac myocytes. It has been reported that adenoviral gene transfer of SERCA2a to the whole heart in vivo improved left ventricular mechanical and energetic functions in a type II diabetes model of cardiomyopathy in rats (14); however, no data is available regarding Ca2+ handling or contractility at the cardiac myocyte level. The lack of information at the cellular level using an adenoviral gene transfer approach may be due to technical difficulties to assess calcium handling and cell shortening in transfected cardiac myocytes, since adenoviral transfer is inhomogeneous. In addition, adenoviral administration can lead to an inflammatory process that can jeopardize the specific effect of upregulated SERCA2a. In this work, we tested the hypothesis that conditional increased SERCA2a expression can return cardiac function, Ca2+ handling, and contractility at the cardiac myocyte level toward normal in animals with already established diabetes-induced cardiomyopathy. For this purpose, we utilized a double transgenic mouse model in which a tetracycline system leads to inducible, cardiac-myocyte-specific SERCA2a expression (tet-on SERCA2a) (16). We demonstrate that, in vivo, inducible SERCA2a expression is able to correct calcium-handling abnormalities and restores contractility of cardiac myocytes of diabetic mice toward normal and results in consequently improved global cardiac function without detrimental effects.
MATERIALS AND METHODS
All animal protocols were approved by the University of California, San Diego, Animal Subjects Committee and conform to the Guide for the Care and Use of Laboratory Animals as outlined by the National Institutes of Health.
Generation of transgenic mice.
Transgenic mice expressing a codon-optimized reverse tetracycline transactivator driven by the cardiac specific α-myosin heavy chain promoter (MHC-rtTA) have been previously described (16, 19). The inducible SERCA2a transgene construct was created by cloning the tetracycline response element (Tet-RE) in front of a rat SERCA2a transgene described previously (7), containing the first and second introns and a COOH-terminal FLAG tag. This construct was injected into fertilized mouse eggs to generate Tet-RE-SERCA2a lines. To determine transgene integration, genomic DNA was extracted from tails of 3-wk-old mice and subjected to Southern blot analysis. The MHC-rtTA and Tet-RE-SERCA2a lines were crossed to generate compound transgenic mice (SER-TG). Animals carrying both transgenes were used for the experiments. To induce the expression of SERCA2a-FLAG, animals received doxycycline (DOX) in their drinking water (200 mg/l) for 7 days.
SER-TG mice (2 mo old) were made diabetic by a single intraperitoneal injection of 200 mg/kg streptozotocin (STZ) as described previously (18). Diabetic mice had blood glucose levels in excess of 33.3 mmol/l. Experiments were carried out after 10 wk of STZ injection.
Western blot analysis.
Isolated cardiac myocytes were homogenized in lysis buffer (20 mM Tris, pH 7.4, 20 mM NaCl, 0.1 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate, 1 mM dithiothreitol). Lysates were incubated on ice for 30 min, and cellular debris were pelleted at 10,000 g for 20 min at 4°C. Protein content was measured by the Bradford method (Bio-Rad) and adjusted for equal loading. Protein extracts from heart tissue (20 μg) were separated by a 4–12% Bis-Tris-HCl buffered polyacrylamide gel (Invitrogen, Carlsbad, CA) and subjected to Western blotting for SERCA2a. Using M2 antibody against the FLAG tag we identified transgenic SERCA2a expression. Total SERCA2a (transgenic SERCA2a-FLAG and endogenous SERCA2a) was assessed with an antibody targeted to the NH2-terminus (Santa Cruz Biotechnology). Other antibodies used were anti-phospholamban (anti-PLB; Affinity Bioreagents, Golden, CO), anti-phospho-PLB (phosphoSer16) (Upstate, Lake Placid, NY), antiRyR (Affinity Bioreagents, Golden, CO), and anti-phospho-RyR (phosphoSer2808) (Abcam, Cambridge, MA). The blot was also probed by a mouse monoclonal α-actin antibody as an internal control to ensure equivalent protein loading and protein integrity. Signals on the films were digitized on a 350 dots-per-inch scanner and analyzed using Image J software (National Institutes of Health).
Isolation of adult ventricular cardiomyocytes.
Ca2+ tolerant adult cardiomyocytes were isolated from ventricular tissue of mice by a standard enzymatic digestion procedure (9) and cultured on 24 × 50 mm, No. 1 glass coverslips treated with laminin as described previously (9, 15).
Calcium transients were recorded in freshly plated myocytes, with the use of Indo-1 or Fluo-3. The method has been described previously (6, 15, 16). Calcium transients were recorded from at least 20 cells per heart and for at least three hearts per treatment. Diastolic and systolic intracellular Ca2+ levels were inferred from the basal and maximal indo-1 ratio per cycle, respectively. Diastolic decay time (Tdecay) was calculated from the normalized Ca2+ transient. These experiments were performed at room temperature.
Measurement of myocyte contractility by edge detection.
Contractile properties of single myocytes were measured at room temperature by using edge detection, as described (16). Myocyte fractional shortening, maximal shortening rate (+dL/dt), and relengthening rate (−dL/dt) were analyzed with the use of Felix32 software (Photon Technology International, Birmingham, NJ). Data were collected from at least 20 cells per heart and three hearts per group.
Transthoracic echocardiography was performed as previously described (16). For acquisition of in vivo cardiac functional data we used an Apogee CX (ATL Interspec) echocardiography system. For image acquisition, mice were anesthetized with Avertin 2.5% (10 μl/g body wt). The 2D parasternal short-axis view was used as a guide and a left ventricle (LV) M-mode tracing was obtained close to the papillary muscle level with a sweep speed of 100 mm/s. Pulsed Doppler tracings of the estimated LV outflow tract velocity were obtained in a modified parasternal long-axis view at a sweep speed of 100 mm/s. M-mode and Doppler tracings were recorded on a video tape for offline analysis on an Agilent Sonos 5500 system. After calibration of this system, left ventricular end-diastolic and end-systolic internal diameter (LVEDD and LVESD, respectively) were measured. LV fractional shortening was calculated as FS (%) = (LVEDD − LVESD)/LVEDD × 100. Using the mean aortic ejection time (ET) from three consecutive heart cycles obtained from the Doppler tracings of the LV outflow tract, we calculated the velocity of circumferential fiber shortening (Vcf) as Vcf (circ/s) = (πLVEDD − πLVESD)/(ET × πLVEDD). In addition, heart rate was also measured. Blood pressure (systolic and diastolic) was measured inserting a Millar catheter directly in the abdominal aorta.
Isolated perfused hearts.
Hearts were isolated and transferred to a miniaturized Langendorff set-up for contractile studies. A constant flow Langendorff perfusion was initiated with Krebs Henseleit buffer at a flow rate of 3 ml/min. A small fluid filled balloon was inserted into the left ventricle (LV) and inflated to an end-diastolic pressure of 10 mmHg. Pressure development was recorded digitally by connecting the intraventricular balloon to a pressure transducer. The hearts were paced at 400 beats/min and the resulting pressure waves were analyzed for pressure derivatives (+dP/dt, −dP/dt) and peak systolic pressure.
Results were expressed as means ± SE. The data among groups were compared using one-way ANOVA followed by Bonferroni's post hoc test for multiple comparisons. A P < 0.05 was considered to be statistically significant.
STZ-injected diabetic animals showed a significant decrease (∼30%) in body weight (Table 1). Serum glucose levels were increased in STZ-treated mice, and DOX treatment did not affect glucose levels (Table 1). All studied animal groups presented similar blood pressure and heart rate values (Table 1).
We tested the hypothesis that timed inducible increased expression of SERCA2a transgene may alleviate abnormalities observed in calcium handling and cardiac contractility of diabetic mice. Therefore, SER-TG mice, which were diabetic for 10 wk were studied. A group of diabetic TG mice received DOX in their drinking water during 7 days to induce SERCA2a expression. DOX treatment did not affect the studied parameters in wild-type (WT) mice (not shown). We did not observe a change in mortality in the diabetic SER-TG mice with DOX compared with SER-TG diabetic that did not receive DOX (not shown).
Induced SERCA2a expression in the diabetic heart.
To confirm increased SERCA2a expression, protein from myocytes was subjected to SDS-PAGE. Western blot analysis revealed that STZ-induced diabetes reduced SERCA2a protein levels by 60% in SER-TG mice compared with control SER-TG (Fig. 1, A and B). Induction of SERCA2a expression increased SERCA2a levels in control SER-TG by 38% after DOX treatment compared with control SER-TG in absence of DOX. Surprisingly, SER-TG diabetic mice presented a higher SERCA2a protein levels after DOX induction compared with controls and diabetic mice (Fig. 1, A and B). Expression of inducible SERCA2a protein was observed only in the animals that received DOX as detected by the anti-FLAG antibody. In addition, the expression of the inducible SERCA2a was increased in the diabetic mice (Fig. 1, A and B). Basal SERCA2a expression was not different between control SER-TG and WT mice (not shown).
Influence of increased SERCA2a expression on the expression of RyR and PLB.
Hearts from SER-TG mice presented a 40% reduction in RyR protein levels after SERCA2a induction with DOX compared with control SER-TG (Fig. 1, A and C). RyR protein levels were reduced by 70% in SER-TG diabetic hearts and 66% after SERCA2a induction with DOX (Fig. 1, A and C). Phosphorylated RyR (p-RyR) presented the same pattern described for nonphosphorylated RyR (Fig. 1, A and C). The p-RyR-to-RyR ratio diminished by SERCA2a expression in the normal mouse heart (Fig. 1, A and C). PLB and phospho-PLB were decreased in all groups compared with control SER-TG mouse hearts (Fig. 1, A and D). The p-PLB-to-PLB ratio was also decreased by ∼40% in all groups compared with SER-TG hearts.
Influence of increased SERCA2a expression on contractile function and calcium transients of diabetic cardiac myocytes.
Contractility in the cardiac myocyte was assessed by edge detection. Cell shortening was diminished in diabetic cardiac myocytes by 30% (Fig. 2). Rate of contraction (+dL/dt) and relaxation(−dL/dt) was also decreased in diabetic myocytes (Fig. 2). SERCA2a induction by DOX returned these parameters to control values. We investigated whether these changes in contractility were associated with corresponding alterations of calcium handling. Analysis of calcium transients showed that the rate of diastolic decay of Ca2+ (Tdecay) was delayed in diabetic cardiac myocytes by 66% (Fig. 3). Induction of SERCA2a expression accelerated Ca2+ decline in the diabetic myocyte above control values (Fig. 3). Systolic Ca2+ and calcium released (ΔCa2+) were decreased in the diabetic myocytes and increased toward control values by SERCA2a expression (Table 1). Diastolic calcium levels and time to reach peak systolic Ca2+ (Rsys and Tmax, respectively) did not change (Table 1).
Effect of SERCA2a expression on global contractile function of diabetic hearts.
Diminished fractional shortening and the mean rate of circumferential shortening measured in vivo by echocardiography in SER-TG diabetic mice was returned to normal by SERCA2a induction with DOX (Table 1). Cardiac function in ex vivo Langendorff-perfused hearts was measured using a fluid-filled intraventricular balloon. Developed left ventricular pressure was 30% lower in the diabetic heart compared with control (Fig. 4). It was associated with slower rates of contraction (+dP/dt) and relaxation (−dP/dt). SERCA2a overexpression was able to return these parameters to control values (Fig. 4).
Diabetic cardiomyopathy results in significant contractile and Ca2+ handling abnormalities, in part, mediated by decreased SERCA2a expression. It is currently unclear whether alleviating decreased SERCA2a expression in hearts with established diabetic cardiomyopathy could still return the contractile function of the diabetic heart toward normal.
Our results show that increasing SERCA2a protein levels in mouse hearts with diabetic cardiomyopathy resulted in a dramatic improvement of contractile function and calcium handling in the myocyte. In addition, the global cardiac function of the diabetic heart was normalized.
Diabetic cardiomyopathy is characterized by reduced cardiac contractility due to direct changes in heart muscle function independent of vascular disease (1, 13). Abnormal calcium handling with diminished Ca2+ entry into the cytoplasm during systole and delayed lowering of diastolic Ca2+ levels is an important contributor to contractile dysfunction in the diabetic state (2). Decreased numbers of RyRs, as well as diminished SERCA2a activity and expression, are important contributors to the abnormal Ca2+ handling in diabetes (17). Efforts have focused on increasing SERCA2a activity or protein levels. Increased SERCA2a activity would result in acceleration of Ca2+ decline and increased SR calcium load (8, 16). These effects may improve cardiac contraction and relaxation (4, 11). Constitutive SERCA2a overexpression has a “prophylactic” protective effect against diabetic cardiomyopathy. However, whether SERCA2a overexpression can revert an established diabetic cardiomyopathy was unclear. In the present study, we have demonstrated that SERCA2a induction in a chronic (10-wk) type-I diabetic mouse with severe cardiomyocyte dysfunction was able to return contractility and calcium handling abnormalities toward normal.
We found a dramatic decrease in SERCA2a protein levels in our diabetic model. Induction of SERCA2a by administration of DOX increased SERCA2a levels in the diabetic heart even higher than control values (Fig. 1, A and B). Different results were observed in previous studies in a model of pressure overload-induced heart failure (16). In the pressure overload model, SERCA2a levels reached only control values after SERCA2a induction in the failing heart. In this work, SERCA2a overexpression also influenced the endogenous expression of RyR and PLB. These two calcium-related proteins were downregulated in the diabetic heart (Fig. 1, A, C, and D), which is in agreement with other reports (10). Surprisingly, SERCA2a induction decreased RyR and PLB in the normal mouse heart. Similar results were found in a transgenic mouse that overexpresses SERCA1a in the SR (8). This effect of SERCA overexpression indicates a compensatory downregulation of the RyR protein expression to regulate the amount of Ca2+ released during each contraction to counteract the increased SR Ca2+ load caused by increased SERCA2a activity (11, 16). Furthermore, this data strongly suggest that the improvement in cardiac contraction and relaxation is mostly due to SERCA2a overexpression. In this situation, effects related to decreased RyR and PLB expression or the phosphorylation level of these proteins would be largely nullified by the effects of SERCA2a overexpression. This idea is supported by the observation that RyR, PLB, and the phosphorylated forms of these proteins appeared decreased after SERCA2a induction in control mice and decreased further in diabetic mice with no improvement after SERCA2a induction. However, contractile and calcium flux parameters were dramatically changed.
Increase in SERCA2a levels in the diabetic heart was able to return myocyte contractile parameters to the normal range. Cell shortening, rate of contraction, and relaxation were higher in diabetic myocytes with increased SERCA2a expression than in control myocytes, and these results are in line with the observed increased SERCA2a expression. Cytosolic calcium decline (Tdecay) in the mouse myocyte is responsible for myocyte relaxation and is mainly due to SERCA2a activity. Tdecay was markedly prolonged in the diabetic myocyte, which is in agreement with previous studies (18). SERCA2a expression accelerated calcium decline in the diabetic myocytes even faster than in control myocytes, a finding consistent with higher SERCA2a protein levels. The beneficial effect on SERCA2a overexpression on contraction is the result of the increased SR calcium load produced by increased SERCA2a activity. The experiments at the cellular level showed a supranormal recovery of the diabetic myocyte after SERCA2a overexpression that was not observed in the global cardiac function. Other adaptive mechanisms (cardiac remodeling, increased cell death), which could counteract the effects at the cellular level, could be responsible for this discrepancy. SERCA2a overexpression may have detrimental effects on cardiac global function due to higher ATP consumption by the SERCA2a pump. Such an effect could worsen the diabetic energy imbalance resulting in deteriorated cardiac function. However, our results showed the opposite. Determining contractile function in isolated perfused hearts or in vivo by echocardiography, we found that SERCA2a expression returned cardiac contractile function to control values. It is interesting to note that we have demonstrated, in previous studies, that SERCA2a overexpression increased mitochondrial calcium levels, which was associated with an increase in the active form of the pyruvate dehydrogenase complex (3). Furthermore, it has been demonstrated that adenoviral gene transfer of SERCA2a improved energetic parameters in a type II diabetic model transforming inefficient energy utilization into a more efficient state (14).
Perspectives and Significance
Diabetes mellitus is a growing public heath concern, affecting 170 million individuals worldwide. Cardiovascular disease is the leading cause of death in diabetes. Diabetes predisposes patients to the development of a specific cardiomyopathy that increases cardiovascular risk. In the present study, we presented data demonstrating that expression of SERCA2a protein during a well-established diabetic state is able to alleviate cardiac dysfunction of the diabetic cardiomyopathy. Further experiments can be performed to develop potential therapeutic approaches using gene therapy to induce SERCA2a expression in nontransgenic diabetic models that could be applied in humans.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-66917 (to W. H. Dillmann) and was partially supported by Grant P60-MD-00220, from the San Diego EXPORT Center, National Center of Minority Health and Health Disparities, National Institutes of Health.
The contents of this study are solely the responsibility of the authors and do not necessarily represent the official views of National Institutes of Health.
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 © 2008 the American Physiological Society