IGF-I rescues diabetic heart defects and oxidative stress, although the underlying mechanism of action remains poorly understood. This study was designed to delineate the beneficial effects of IGF-I with a focus on RhoA, Akt, and eNOS coupling. Echocardiography was performed in normal or diabetic Friend Virus-B type (FVB) and IGF-I transgenic mice. Cardiomyocyte contractile properties were evaluated using peak shortening (PS), time-to-90% relengthening (TR90), and intracellular Ca2+ rise and decay. Diabetes reduced fraction shortening, PS, and intracellular Ca2+; it increased chamber size, prolonged TR90, and intracellular Ca2+ decay. Levels of RhoA mRNA, active RhoA, and O2− were elevated, whereas nitric oxide (NO) levels were reduced in diabetes. Diabetes-induced O2− accumulation was ablated by the NO synthase (NOS) inhibitor nitro-l-arginine methyl ester (l-NAME), indicating endothelial NOS (eNOS) uncoupling, all of which except heart size were negated by IGF-I. The IGF-I-elicited beneficial effects were mimicked by the Rho kinase inhibitor Y27632 and BH4. Diabetes depressed expression of Kv1.2 and dihydrofolate reductase (DHFR), increased β-myosin heavy-chain expression, stimulated p38 MAPK, and reduced levels of total Akt and phosphorylated Akt/eNOS, all of which with the exception of myosin heavy chain were attenuated by IGF-I. In addition, Y27632 and the eNOS coupler folate abrogated glucose toxicity-induced PS decline, TR90 prolongation, while it increased O2− and decreased NO and Kv1.2 levels. The DHFR inhibitor methotrexate impaired myocyte function, NO/O2− balance, and rescued Y27632-induced cardiac protection. These results revealed that IGF-I benefits diabetic hearts via Rho inhibition and antagonism of diabetes-induced decrease in pAkt, eNOS uncoupling, and K+ channel expression.
- K+ channel
- nitric oxide
- intracellular Ca2+
diabetes leads to onset of diabetic cardiomyopathy characterized by reduced contractility, prolonged contractile duration, and compromised wall compliance (8). Several rationales have been postulated for the pathogenesis of diabetic cardiomyopathy, including oxidative stress, reduced nitric oxide (NO) bioavailability, and elevated O2− production due to uncoupled endothelial NO synthase (eNOS), leading to impaired ionic conductance and aberrant Ca2+ homeostasis (10, 25, 26, 28, 43). Evidence from our laboratories and others has indicated that diabetes upregulates the Ras-related small G protein RhoA and reduces eNOS phosphorylation (8, 26, 31). In fact, RhoA was shown to mediate diabetes-induced erectile dysfunction through inhibition of eNOS (2), although a similar mechanism has not been elucidated in diabetic hearts.
IGF-I, a crucial cardiac survival factor, benefits cardiac growth, function, and energy metabolism under both normal and diabetic conditions (17, 24). Nonetheless, the precise mechanism behind its favorable effects in diabetes is still unclear. IGF-I downregulates RhoA, stimulates NO release, and regulates cardiac function, remodeling, and apoptosis (19, 31, 33, 38). We thus speculated that IGF-I benefits diabetic hearts through regulation of RhoA and eNOS-NO. Levels of RhoA (active and total), eNOS and eNOS phosphorylation (Ser1177), the eNOS cofactor BH4 salvage enzyme dihydrofolate reductase (DHFR), the upstream eNOS stimulator Akt, and its phosphorylation were evaluated. Release of NO and O2− was monitored. It has been demonstrated that oxidation and deficiency of BH4 due to decreased DHFR may render eNOS to produce O2− rather than NO, a process known as the eNOS uncoupling, in diabetic vasculature (28). However, little is known for eNOS uncoupling in diabetic hearts. In addition, voltage-gated K+ channel (Kv1.2) and MAPK were examined, given their roles as key targets for RhoA (3, 19).
MATERIALS AND METHODS
The experiments described were approved by our Institutional Animal Care and Use Committees. The work conducted here was in compliance with the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised 1996). In brief, 12-wk-old male cardiac-specific IGF-I transgenic and weight-matched Friend Virus-B type (FVB) mice were injected with streptozotocin (STZ, 200 mg/kg ip) and maintained for 4 wk with free access to standard lab chow and tap water. Mice with fasting blood glucose levels >13 mmol/l were deemed diabetic. Body weight, blood glucose, and serum IGF-I levels were measured using a lab scale, a glucose monitor, and an ELISA kit from Diagnostic Systems Laboratory (Webster, TX), respectively.
Cardiac geometry and function were evaluated in anesthetized (Avertin 2.5%; 10 μl/g body wt ip) mice using a two-dimensional guided M-mode echocardiography (Sonos 5500) equipped with a 15-6-MHz linear transducer. Diastolic and systolic left ventricular (LV) dimensions were recorded from M-mode images using method adopted by the American Society of Echocardiography (12). Fractional shortening was calculated from end-diastolic diameter (EDD) and end-systolic diameter (ESD) using the equation of (EDD − ESD)/EDD. Estimated echocardiographic LV mass was calculated as [(LVEDD + septal wall thickness + posterior wall thickness)3 − LVEDD3] × 1.055, where 1.055 (mg/mm3) is the density of myocardium (12).
Cell isolation and culture.
Hearts were perfused with collagenase II using a Langendorff system as described (43). Myocyte yield was ∼70%, which was not affected by either diabetes or IGF-I transgene. For studies with the Rho inhibitor Y27632 (10 μmol/l) (21), the eNOS cofactor BH4 (10 μmol/l) (9), the eNOS coupler folate (50 μmol/l) (1) and the DHFR inhibitor methotrexate (1 μmol/l) (37). Cardiomyocytes were maintained in medium 199 containing normal glucose (NG; 5.5 mmol/l) or high glucose (HG; 25.5 mmol/l) for 24 h (unless otherwise specified) at 37°C in an incubator with 100% humidity and 5% CO2 (8, 24, 25).
Cell mechanics and intracellular Ca2+ transients.
Mechanical properties of cardiomyocytes were assessed using a MyoCam system (30). In brief, myocytes were placed onto the stage of an inverted microscope and were field stimulated at 0.5 Hz. A SoftEdge software was used to capture changes in cell length. Cell shortening and relengthening were assessed, including peak shortening (PS) [peak contractility correlating with fraction shortening, time-to-PS (TPS)], contraction duration, time-to-90% relengthening (TR90) [relaxation duration and maximal velocities of shortening/relengthening (±dL/dt)] maximal pressure development, and decline. Intracellular Ca2+ fluorescence was measured in myocytes loaded with fura-2 (0.5 μmol/l). Fluorescence emissions were detected between 480 and 520 nm. Qualitative changes in intracellular fura-2 fluorescence intensity (ΔFFI) were inferred from FFI ratio at two wavelengths (360/380).
RhoA and ROCKI/II mRNA RT-PCR.
Reverse transcription was performed according to the instruction of Invitrogen. The thermal cycle profile was denaturing for 60 s at 92°C, annealing for 60 s at 56°C, extending for 30 s at 72°C, and 28 cycles of amplification. The following primer designs were used as previously reported (8): RhoA forward primer, 5′-ACC AGT TCC CAG AGG TTT ATG T-3′; RhoA reverse primer, 5′-TTT GGT CTT TGC TGA ACA CT-3′; GAPDH forward primer, 5′-TCC CTC AAG ATT GTC AGC AA-3′; and GAPDH reverse primer, 5′-AGA TCC ACA ACG GAT ACA TT-3′. The primer designs for ROCK I and ROCK II were based on the published sequences (22). PCR products were confirmed by sequencing and run on 2% agarose gel. The gel was stained with ethidium bromide and photographed. Data were expressed as a ratio of RhoA or ROCKI/II to GAPDH using UN-SCAN-IT gel quantitation software.
Western blot analysis and determination of NO and O2−.
Proteins were extracted from cardiomyocytes as described previously (8, 30). Membrane proteins (50 μg/lane) were separated on 10–15% SDS-polyacrylamide gels in a minigel apparatus and transferred to polyvinylidene difluoride membranes. The membranes were blocked with 5% milk and incubated with respective primary and secondary antibodies. The film was scanned, and the intensity of immunoblot bands was normalized to the loading control β-actin. NO levels were determined colorimetrically after mixing 0.5-ml cell lysates and Griess reagent [0.1% N-(1-naphthyl)ethylenediamine in water and 1% sulfanilamide in 5% phosphoric acid]. Concentrations of nitrite were estimated by comparing absorbance readings at 550 nm against standard solutions of NaNO2 (also used as a positive control). NO2-free H2O was used as negative control (27). Intracellular O2− was monitored by changes in dihydroethidium (5 μmol/l) fluorescence intensity from intracellular probe oxidation (7).
Determination of active RhoA.
RhoA expression was measured using the EZ-Detect Rho activity kit. After incubation of heart lysates with SwellGel Immobilized Glutathione Disc and GST-Rhotekin-RBD for 1 h at 4°C, the resin containing active Rho was carefully washed three times. Protein was recovered from resin, and Western blot analysis was performed using 12% SDS-PAGE separating gel. The nitrocellulose membrane was blocked in 3% BSA and incubated with an anti-Rho primary antibody (1:500), followed by a horseradish peroxidase-conjugated goat anti-mouse IgG secondary antibody. Conventional Western blot analysis was performed with heart lysates to detect total RhoA (42).
Myosin heavy chain isoform analysis.
Each sample (20 to 30 mg of heart tissue) was placed in a microcentrifuge tube, and 30 μl of sample buffer was added per milligram of tissue. The sample was homogenized for ∼10 s, heated for 2 min at 95°C, and chilled on ice for 5 min. Three microliters of samples was loaded for electrophoresis. Gel electrophoresis was performed at 8°C in a Bio-Rad PROTEAN II unit. Stacking gels consisted of 4% acrylamide (acrylamide:bis = 50:1) and 5% glycerol (vol/vol, pH 8.8). Gels were run at a constant voltage of 200 V for 30 h, fixed for a minimum of 2 h in 5% glutaraldehyde before being silver-stained and scanned to determine the amount of α-myosin heavy chain (MHC) and β-MHC (42).
Values are presented as means ± SE. Statistical significance (P < 0.05) was estimated by ANOVA followed by Dunnett's post hoc analysis using the SPSS software (α = 0.5).
Morphometric characteristics, echocardiographic and cardiomyocyte mechanical parameters.
STZ significantly elevated blood glucose levels, lowered body weight and plasma IGF-I levels, increased ESD, heart weight and LV mass normalized to body weight, as well as decreased fractional shortening without affecting absolute heart weight and echocardiographic LV mass. Heart rate and EDD were comparable among all groups. IGF-I masked STZ-induced changes of plasma IGF-I levels, ESD, and fractional shortening but not blood glucose, body weight, and hypertrophied heart. IGF-I itself increased both absolute and normalized weights of heart and LV without altering myocardial geometry and fractional shortening (Table 1). IGF-I enhanced myocyte cross-sectional area in both groups. Diabetes significantly reduced cardiomyocyte PS and ±dL/dt) in FVB mice. Diabetic myocytes also demonstrated significantly prolonged TR90 but normal TPS, consistent with our previous findings (8, 43). Importantly, IGF-I abolished these diabetes-induced mechanical abnormalities without any overt effect by itself (Fig. 1). Consistent with data from a 2-wk STZ treatment study (24), cardiomyocytes from 4-wk STZ mice displayed a significantly reduced rate of intracellular Ca2+ clearing, peak, and rise in intracellular Ca2+ (ΔFFI), which were alleviated by IGF-I. Baseline FFI was not affected by either diabetes or IGF-I (Fig. 2). Our further study revealed that a 4- to 6-h short-term incubation of Y27632 or BH4 cancelled out STZ-induced contractile and intracellular Ca2+ defects. Neither Y27632 nor BH4 exerted any effect on FVB cardiomyocyte function (Figs. 1 and 2) or IGF-I-elicited beneficial cardiac contractile effects against diabetes (data not shown).
Protein and mRNA levels of RhoA, ROCKI/II, as well as levels of NO and O2−.
Diabetes is associated with elevated cardiac RhoA signaling, reduced NO bioavailability, and accumulation of free radicals (8, 43). To examine whether IGF-I-elicited cardiac protection is related to these signaling cascades, levels of RhoA, ROCK (ROCKI/II), NO, and O2− were evaluated. Our data revealed that diabetes significantly enhanced RhoA mRNA, active (but not total) RhoA protein and O2− production associated with reduced NO levels. While IGF-I itself did not affect levels of RhoA, NO, and O2−, it abrogated diabetes-induced changes in RhoA mRNA, active RhoA protein, NO, and O2− production. However, mRNA levels of the Rho-associated kinase, ROCKI/II were not affected by diabetes but were rather reduced by the IGF-I transgene itself. Our data further revealed that STZ-induced elevation in O2− was blocked by the NOS inhibitor l-NAME (100 μmol/l), indicating a NOS-dependent mechanism for O2− accumulation or “uncoupled eNOS” (Fig. 3).
Expression of Akt, eNOS, their phosphorylation, and DHFR.
Fig. 4 indicates that diabetes significantly downregulated Akt and its phosphorylation (Ser473) (pAkt or pAkt/Akt ratio), which was reversed by IGF-I. Consistent with reduced NO release in diabetes (Fig. 3), phosphorylation of the Akt downstream target eNOS (Ser1177) was also reduced in diabetic mice, although total eNOS levels were unchanged. Consistent with its effect on mechanical function, RhoA, and NO release, IGF-I reversed diabetes-elicited reduction in Akt, pAkt, and phospho-eNOS. IGF-I itself did not affect Akt, eNOS, or phospho-eNOS, although it significantly enhanced basal Akt phosphorylation as reported earlier in these IGF-I transgenic mice. (41) Our further study revealed that diabetes downregulated the BH4 salvage enzyme DHFR, which was reconciled by IGF-I.
Expression of p38 MAPK, ERK1/2, Kv1.2, and MHC.
IGF-I has been shown to neutralize diabetes-induced oxidative stress (17). We evaluated p38 MAPK and ERK1/2, two key signaling molecules in the MAPK superfamily. Data in Fig. 5 showed that diabetes significantly enhanced p38 MAPK phosphorylation, which was abolished by IGF-I. Neither diabetes nor IGF-I affected total p38 and ERK1/2, as well as phosphorylation of ERK1/2. Voltage-dependent K+ channels (Kv1.2) are essential to cardiac excitation-contraction coupling (23) and may be negatively regulated by RhoA (3). Consistently, our data revealed that Kv1.2 expression was significantly reduced by diabetes, which was restored by IGF-I. Since relative abundance of β-MHC isozyme is associated with cardiac function phenotype (20), we examined the MHC isozyme distribution in normal and diabetic hearts. Our results revealed an increased proportion of β-MHC isozyme in FVB-STZ mouse hearts. Although IGF-I reduced the percentage of β-MHC in both normal and diabetic hearts, it failed to reduce STZ-induced percent increase (∼150%) in β-MHC isozyme.
Effect of y27632, folate, and methotrexate on glucose toxicity-induced cell shortening, eNOS coupling, and Kv1.2 expression.
To further examine the role of RhoA and DHFR in diabetic cardiac defects, we used our established culture system, simulating diabetic condition. Consistent with the previous data (29), myocytes maintained in HG medium for 24 h exhibited significantly decreased PS and prolonged TR90 compared with NG cells. Interestingly, both Y27632 and the DHFR recoupler folate significantly attenuated HG-induced mechanical defects without eliciting any effects in NG cells themselves. Consistently, Y27632 and folate alleviated HG-induced downregulation of DHFR and Kv1.2, as well as elevated O2− and reduced NO production. Neither Y27632 nor folate itself elicited any effect on DHFR and Kv1.2 expression or levels of O2− and NO. These data indicated a role of eNOS coupling and K+ channels in glucose toxicity-induced cardiac dysfunction and Y27632/folate-exerted cardioprotection. Our further study revealed that direct inhibition of DHFR with methotrexate significantly compromised cardiomyocyte contractile function (reduced PS and prolonged TR90), increased O2− production, suppressed NO release, and downregulated Kv1.2 channel without affecting DHFR expression in normal cardiomyocytes. More interestingly, the beneficial effects of Rho inhibition against HG, including myocyte mechanics, levels of O2−/NO and expression of Kv1.2 were antagonized by the DHFR inhibitor (Fig. 6).
Our current data indicated that selective IGF-I overexpression protects against diabetic cardiac defects, possibly through rescuing RhoA-mediated inhibition of eNOS/NO, elevation of O2− and downregulation of Kv1.2 channel. The reduced DHFR expression in conjunction with the l-NAME-inhabitable O2− accumulation reveals the presence of “eNOS uncoupling” in diabetic hearts, which is consistent with reduced phosphorylation of eNOS (Ser1177) and Akt (Ser473). The involvement of Rho and eNOS uncoupling in diabetes-associated suppression of Kv1.2 and cardiac dysfunction was further substantiated by our in vitro study when Y27632 and folate protected against HG, whereas methotrexate ameliorated the beneficial effect of Y27632. Our results do not support any involvement of ERK1/2 and MHC in diabetes-induced cardiomyocyte dysfunction or selective IGF-I overexpression-elicited beneficial effect, although p38 MAPK seems to contribute to the IGF-I-elicited favorable effects.
Reduced contractility and prolonged diastole are hallmarks of diabetic cardiomyopathy (8, 17, 24). Our study depicted reduced fractional shortening, PS, ±dL/dt, and prolonged TR90 in diabetic heart, similar to our previous findings (8, 24, 25). Several mechanisms have been postulated for diabetes-related defects such as MHC isozyme switch (from the fast α to the slow β isoform), impaired intracellular Ca2+ homeostasis, and reduced myofilament Ca2+ sensitivity (4, 6, 14, 25). Dysfunction of K+ channels may contribute to prolonged action potential repolarization and diastolic duration in diabetic hearts (35). Our observation of downregulated Kv1.2 in diabetic hearts is supported by prolonged relaxation duration, reduced ΔFFI, compromised intracellular Ca2+ decay, and β-MHC isozyme switch in STZ-induced diabetic hearts. Data from our laboratory (8), as well as others (21), suggest upregulation of RhoA may be essential to diabetic cardiac defects. Although the precise mechanism is not fully clear behind upregulated RhoA in diabetes, the cyclic AMP response element, which may be stimulated by hyperglycemia and p38 MAPK-dependent oxidative stress (16, 36), may facilitate RhoA promoter activity (34). This is consistent with elevated O2− levels, enhanced p38 phosphorylation, and Rho activation in our diabetic FVB mice. Paradoxically, Rho activation has been demonstrated to regulate cardiac contractility and gene expression through the MAPK superfamily (19, 33). Thus, p38 MAPK may participate in cardiac regulation both upstream and downstream of RhoA (Fig. 7). Nevertheless, our present study failed to reveal an increased ROCK I/II mRNA expression under diabetic state despite elevated RhoA activity in STZ-treated FVB mouse hearts. Such discrepancy between ROCK I/II mRNA and RhoA activity levels may indicate possible involvement of a posttranslational rather than a transcriptional modification of the small G protein in STZ-induced diabetes. Cardiac-specific overexpression of RhoA has been shown to result in action potential prolongation and diminished ventricular contractility (33), through interrupted Kv1.2 conductance (3), consistent with our current observation of upregulated RhoA mRNA, Rho activation, and diminished Kv1.2 in diabetic hearts. Our findings favor the hypothesis that eNOS-NO may serve as the downstream messengers for Rho-mediated diabetic cardiac defects. Rho-induced inhibition of eNOS was suggested in diabetic erectile dysfunction (2). Our current study provided evidence for the first time that diabetes and hyperglycemia reduce phosphorylation of Akt and eNOS, as well as NO production, and promote O2− accumulation in a Rho-dependent manner in the hearts. The l-NAME-inhabitable O2− production associated with reduced DHFR expression has further consolidated the case of “eNOS uncoupling” in diabetic hearts. Our in vitro glucose toxicity study seems to provide evidence that the “eNOS uncoupling” is Rho-dependent since Y27632 effectively corrected glucose-elicited alteration in DHFR, NO, and O2−. Given the essential role of eNOS, NO, and K+ channels in the maintenance of cardiac contractile function (11, 23), our data suggested that Rho-mediated eNOS uncoupling and downregulation of K+ channels may be a permissive step in the development of cardiac contractile defects as outlined in Fig. 7. This notion received further support in that the eNOS recoupler folate protects against HG-induced cardiac dysfunction, reduced NO, and elevated O2− levels. On the other hand, the DHFR inhibitor methotrexate antagonized Rho inhibition-elicited cardiac protection, while simulating cardiomyocyte dysfunction, NO, and O2− production in a manner reminiscent of diabetes or hyperglycemia.
Perhaps our most significant finding is that cardiac-selective IGF-I overexpression corrected diabetes-induced cardiac functional and geometric change, cardiomyocyte dysfunction, as well as alterations in RhoA, p38 MAPK, phosphorylation of Akt and eNOS, DHFR, NO, O2−, and Kv1.2 channel. IGF-I facilitates cardiac growth (increased heart and LV mass/size), increases insulin sensitivity, and improves lipid profile (32), suggesting its therapeutic potential. Our data showed that IGF-I enhanced cell cross-sectional area, consistent with enhanced heart-to-body weight ratio and normalized LV mass in IGF-I mice, as seen in our current and earlier studies (24, 25). STZ failed to elicit any hypertrophic effect on cardiomyocyte cross-sectional area, absolute heart weight, and LV mass, although it enhanced the heart-to-body weight ratio and normalized LV mass likely due to STZ-elicited body weight loss. The IGF-I-induced increase in heart weight was not affected by the diabetic state, possibly because of antagonism of IGF-I against diabetes-induced cardiac apoptosis (17). The rationale for a therapeutic role of IGF-I in patients with diabetes is also consistent with the fact that both circulating and tissue IGF levels are reduced in clinical (39) and experimental (25) diabetes, as seen in our study.
It was suggested that IGF-I offers its protection against diabetes via regulation of membrane ionic channels, oxidative stress, and Akt activation (17, 24, 30). It has been suggested that the cardiac beneficial effects of IGF-I may be mediated through an insulin-like activation of the phosphatidylinositol-3 (PI-3)-kinase-Akt cascade (5, 30). Our results of dampened Akt and pAkt in diabetic hearts were consistent with earlier reports (8, 15, 30). Given that Akt is essential for cardiac function and cell survival (15) and pAkt inactivates p38 MAPK and phosphorylates eNOS (13), it is plausible that the IGF-I-elicited beneficial effect is associated with Akt activation-mediated inhibition of p38 MAPK phosphorylation, eNOS phosphorylation, and eNOS recoupling (levels of DHFR, NO, and O2−). Our cardiac-specific IGF-I transgenic mice display much higher basal levels of pAkt and pAkt-to-Akt ratio in the absence of diabetes, suggesting an intrinsic phosphorylation state of Akt, which may be cardioprotective. This beneficial effect of IGF-I is consistent with our observation that IGF-I inhibits the STZ-induced increase in p38 MAP kinase phosphorylation. In addition, activation of p38 MAP kinase has also been speculated to be downstream of the Rho signaling pathway (18, 40). IGF-I transgene reduced ROCK I/II mRNA, as well as STZ-induced increase in RhoA expression and activity, consistent with the notion that IGF-I inactivates RhoA signaling cascade (38). The lack of effect on ERK1/2 phosphorylation in response to either diabetes or IGF-I is somewhat consistent with our earlier report using the high glucose culture system (30).
There were several experimental limitations in our study. First, although the cellular study with the RhoA inhibitor may provide a good linkage between Rho signaling and specific cardiomyocyte mechanical defects, our nonphysiological in vitro high glucose system may not represent the actual diabetic milieu from the in vivo setting. Diabetes mellitus is a rather complex metabolic disease, and its cardiac complications are likely due to multiple factors in addition to hyperglycemia (such as dyslipidemia and insulin resistance). It is not realistic for us to replicate all of these factors in the same cell culture model. Second, the caveats of our techniques used for NO and superoxide quantification have been well documented (44). Better assay methods such as fluorescent dyes should greatly enhance the reliability of our measurement. Third, we failed to obtain action potential profiles in myocytes from normal and diabetic mice, thus making our Western blot analysis on Kv1.2 expression somewhat descriptive and speculative. Last but not least, the IGF-I overexpression-elicited beneficial effect may be simply due to the higher basal levels of IGF-I considering the higher than normal IGF-I levels in STZ-treated IGF-I mice. The higher basal IGF-I levels may make the transgenic mice more resistant to the diabetes-induced insult, given that IGF-I is an insulin analog.
Perspectives and Significance
Our study revealed that diabetes impairs cardiac function associated with upregulation of Rho, dampened Akt phosphorylation, eNOS uncoupling and decreased K+ channel expression. While IGF-I alleviated diabetes-induced echocardiographic and cardiomyocyte defects through inhibition of Rho, it rescues dampened Akt phosphorylation, eNOS uncoupling, and suppressed K+ channel in diabetes. The involvement of Rho and uncoupled eNOS in diabetes and IGF-I-mediated cardiac responses was consolidated by the protective effect from RhoA inhibition and eNOS recoupling against glucose toxicity-induced cardiomyocyte dysfunction, whereas DHFR inhibition cancelled out the beneficial effect of RhoA inhibition and mimicked the hyperglycemic effect itself. These data suggest that RhoA plays a key role in uncoupled eNOS, reduced NO bioavailability, enhanced O2−, and decrease in K+ channel expression in diabetic condition. Our current findings should shed some light on a better understanding of the role of RhoA in the pathogenesis of diabetic cardiomyopathy. More importantly, our observations suggest the therapeutic potential of maintaining eNOS coupling through selective IGF-I overexpression in the management of diabetic complications.
This work was supported by American Diabetes Association (Grant 7-00-RA-21), American Heart Association Pacific Mountain Affiliate (Grant 0355521Z), National Institute on Aging 1R03 AG21324-01, and National Natural Science Foundation of China (30728023) (to J. Ren).
The authors greatly appreciate the helpful discussion from Dr. Ranganath Muniyappa at the National Center for Complementary and Alternative Medicine and the technical support of Drs. Hai-Ying Zhang and Joseph N. Benoit from University of North Dakota School of Medicine.
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