Skip to main content
NCBI home page
Biophysical Journal logoLink to Biophysical Journal
. 2013 Mar 5;104(5):1065–1072. doi: 10.1016/j.bpj.2013.01.037

Myosin Heads Are Displaced from Actin Filaments in the In Situ Beating Rat Heart in Early Diabetes

Mathew J Jenkins †,‡,▵, James T Pearson ‡,§,¶,▵,∗, Daryl O Schwenke ‖, Amanda J Edgley †,‡, Takashi Sonobe ††, Yutaka Fujii ††, Hatsue Ishibashi-Ueda ††, Darren J Kelly †, Naoto Yagi ‡‡, Mikiyasu Shirai ††
PMCID: PMC3870813  PMID: 23473489

Abstract

Diabetes is independently associated with a specific cardiomyopathy, characterized by impaired cardiac muscle relaxation and force development. Using synchrotron radiation small-angle x-ray scattering, this study investigated in the in situ heart and in real-time whether changes in cross-bridge disposition and myosin interfilament spacing underlie the early development of diabetic cardiomyopathy. Experiments were conducted using anesthetized Sprague-Dawley rats 3 weeks after treatment with either vehicle (control) or streptozotocin (diabetic). Diffraction patterns were recorded during baseline and dobutamine infusions simultaneous with ventricular pressure-volumetry. From these diffraction patterns myosin mass transfer to actin filaments was assessed as the change in intensity ratio (I1,0/I1,1). In diabetic hearts cross-bridge disposition was most notably abnormal in the diastolic phase (p < 0.05) and to a lesser extent the systolic phase (p < 0.05). In diabetic rats only, there was a transmural gradient of contractile depression. Elevated diabetic end-diastolic intensity ratios were correlated with the suppression of diastolic function (p < 0.05). Furthermore, the expected increase in myosin head transfer by dobutamine was significantly blunted in diabetic animals (p < 0.05). Interfilament spacing did not differ between groups. We reveal that impaired cross-bridge disposition and radial transfer may thus underlie the early decline in ventricular function observed in diabetic cardiomyopathy.

Introduction

Diabetes is associated with impaired myocardial function in the absence of underlying coronary vascular disease or hypertension (1). This has been termed diabetic cardiomyopathy (DCM) and is characterized by impaired heart muscle relaxation with a progressive increase in left ventricular (LV) muscle stiffness and fibrosis eventually leading to congestive heart failure (2). Several studies suggest that diabetes and DCM are associated with cardiomyocyte changes at the subcellular level resulting from abnormalities in myocardial contractile and regulatory proteins (3–5). One study using electron microscopy demonstrated disarray within cardiac sarcomeric order and disorganization of the mitochondrial matrix within advanced DCM hearts (6). Accumulating evidence from animal models of DCM also shows that the reduction in cardiac function is associated with altered regulatory protein phosphorylation, a reduction in myofibrillar ATPase activity, shifts in myosin enzyme from the faster V1 to the slower V3 isoform, and a reduced calcium uptake by the sarcoplasmic reticulum (7–10). Furthermore, a slower time course of Ca2+ transients, reduced decay time, slower contraction, and diminished developed tension have been reported in the isolated DCM heart (11). Combined with an increase in the activation of oxidative stress cascades in DCM (12) leading to reactive oxygen species-induced protein carbonylation of myosin (13), and potential redox modifications of thin filament proteins is likely to lead to functional defects in actin-myosin cross-bridge formation (14). Joseph et al. (15) have linked these functional changes in isolated LV papillary muscle to altered cross-bridge dynamics by demonstrating that DCM is associated with a reduction in the number of active cross-bridges and subsequent alterations in cross-bridge dynamics. Taken together, these studies strongly suggest that functional depression of the cardiac contractile apparatus contributes to DCM development. However, previous work has primarily been performed with cardiomyocytes, isolated cardiac muscle, or Langendorff preparations and, although informative, uncertainty remains as to the role changes in cardiac cross-bridge regulation play in DCM in vivo.

Synchrotron radiation as a source for small-angle x-ray diffraction has allowed the investigation of cardiac cross-bridge dynamics in real time, in isolated muscle (16), perfused hearts (17), and in situ in beating hearts (18–20). Notably, the combined use of LV pressure-volumetry with small-angle x-ray scattering (SAXS) has allowed us to correlate regional actin-myosin cross-bridge dynamics with changes in global cardiac function on a beat-to-beat basis (19), which is essential to understanding DCM. The aim of this investigation was to determine whether altered regulation of cross-bridge disposition and myosin interfilament spacing underlie the in vivo impairment in cardiac function in a streptozotocin (STZ) rat model of DCM using synchrotron SAXS.

Materials and Methods

Animals and experiments at the synchrotron

Experiments were conducted at Beamline 40XU at the Japan Synchrotron Radiation Research Institute (SPring-8), Hyogo, Japan. Male Sprague Dawley rats (Japan SLC, Kyoto, Japan, 7 weeks old) received either a vehicle injection of sodium citrate (0.1M, pH 4) (control, n = 7) or STZ (65 mg/kg i.p.) (Sigma-Aldrich, Saint Louis, MI) to induce type 1 diabetes (n = 9). Three weeks after vehicle or STZ injection all rats underwent terminal experiments at Beamline 40XU.

Animals and surgical preparation

Surgical preparations were similar to those described by us previously (19). Under sodium pentobarbital anesthesia (50 mg/kg, i.p) (Nembutal, Dainippon Sumitomo Pharma, Osaka, Japan), rats were intubated and artificially ventilated. Supplemental anesthesia was maintained via further doses of pentobarbital (30–40 mg/kg/h i.p.). Pancuronium bromide (Mioblock; 2 mg/kg i.m., Sankyo, Tokyo, Japan) was also administered to eliminate spontaneous breathing when the ventilator was briefly switched off at end-expiration during diffraction recordings, thereby limiting large fluctuations in venous return. Rats were then thoractomized, providing access to the heart, and a continuous flow of lactate Ringers solution was maintained to prevent drying of the exposed heart and lungs. The right jugular vein was catheterized to facilitate drug delivery and fluid replacement, thereby maintaining blood volume and a stable LV volume (lactate Ringers solution, 2.0 ml/h, Otsuka Pharmaceuticals, Osaka, Japan after an initial infusion of 50 units of heparin). The apex of the heart was then raised by a micromanipulator and partly restrained by a plastic receptacle inserted beneath the posterior apex, to limit vertical movements during recordings and ensure maintenance of myocardial depth.

Cardiac catheterization and volume calibration

Cardiac catheterization was performed to allow continuous LV pressure-volume recordings simultaneous with all SAXS and arterial pressure recordings. Calibration of the LV volume signal from relative volume units to absolute volume was performed at the end of the experiment (19). Pressure-volumetry was used to establish the timing of the cardiac cycle in all treatment periods and to permit assessment of actin-myosin influence on global LV function. Heart rate (HR) was determined from the interval between end-diastolic (ED) events in the pressure-volume loops. Hemodynamic data were recorded using CHART (v5.5.6, ADInstruments, NSW) at a sampling rate of 1000/s.

X-ray source, camera, and diffraction recordings

As described previously (19), a collimated quasimonochromatic beam with 0.08 nm wavelength (15 keV), dimensions 0.2 × 0.1 mm (horizontal × vertical) and beam flux ∼1012 photons/s (ring current 90–100 mA) was focused on the surface myocardium at an oblique tangent (rat ∼3m from the detector). SAXS sequences (12 bit, 144 × 150 pixels) each lasting <2.1 s were collected at a sampling interval of 15 ms with the aid of an image intensifier (V5445P, Hamamatsu Photonics, Hamamatsu, Japan) and a fast charge-coupled device camera (C4880-80-24A, Hamamatsu Photonics). Patterns were then digitally recorded using HiPic32 software (v5.1.0 Hamamatsu Photonics); see the Supporting Material.

Experimental protocol

SAXS recordings were sequentially obtained during baseline intravenous infusion of lactate Ringers (2.0 ml/h), 4.0 μg/kg/min and 8.0 μg/kg/min dobutamine (Dobutrex, Eli Lilly Japan, Kobe Japan,) to assess myocardial responses to increasing cardiac contractility and HR. At the termination of experiments animals were euthanized with an overdose of potassium chloride (KCl) solution (100 mM) and diffraction recordings were acquired during complete muscle relaxation.

Diffraction pattern analysis

SAXS patterns were analyzed using customized in-house software (HDA version 3.0) over a minimum of five cardiac cycles (19). Lattice spacing calibration was made using the 63.5-nm meridional reflection of collagen present in a dried chicken tendon sample at the start of the experiments. Using HDA custom software the average radial line profile around the center of the spectrum was calculated using a three point background curve fitting process with manual definition of peak spectra limits. Background subtraction was then performed between user-defined inner and outer limits on either side of the 1,0 and 1,1 reflections. The integrated intensity of the 1,0 and 1,1 reflection intensities was then determined from the areas under the reflection peaks, defined as I1,0 and I1,1, respectively (21); see Fig. 1. The 1,1 intensity, I1,1 was corrected with a 3 multiplication after integration as the averaging along arcs underestimated the 1,1 intensity by 3 relative to the 1,0 reflection (17,22). The integrated intensity was then further multiplied by 3 for the Lorentz factor; see theoretical consideration in the Supporting Material. Because the number of fibers in the beam path changes during contraction the equatorial intensity ratio (I1,0/I1,1) was used to determine the shift of myosin mass (assumed to be predominately cross-bridges) from the region of the thick filament to the thin filament. The 1,0 reflection lattice spacing (d1,0) and 1,1 (d1,1) were obtained from the center of gravity of the integrated 1,0 and 1,1 reflection at ED, end-systolic (ES) and the maximal spacing.

Figure 1.

Figure 1

Open in a new tab

Representative changes in epicardial equatorial reaction intensities recorded during 15-ms diffraction patterns in control and diabetic hearts in vivo. Left panels (A, D, G, J) show x-ray diffraction patterns at ED (A, G) and ES (D, J) in a single control and diabetic heart. The intensity distributions are shown in pseudocolor. The inner ring is the 1,0 reflection, whereas the outer equatorial reflection is 1,1. The middle panels (B, E, H, K) indicate average radial intensity profiles around the center of the spectrum. Dashed lines indicate background intensity. In the rightmost panels (C, F, I, L) integrated reflection intensities for 1,0 and 1,1 were fitted with Gaussian functions, where xc is reflection peak position and w is the peak-width (2 SD of peak).

The relative myosin mass transfer index, Δintensity ratio, was defined as the change in intensity ratio between ED and the minimum intensity ratio, which usually corresponded with ES. In addition, we calculated the radial transfer of myosin heads to the thin filament as an absolute measure of cross-bridge formation using the intensity ratios obtained in the minimum and maximum cross-bridge attachment states. Peak and ED myosin mass transfer to actin in control and diabetic hearts was therefore interpolated using linear regression based on the intensity ratio in arrested hearts following KCl (minimum cross-bridge attachments) and the minimum attainable intensity ratio of 0.3 in the rigor state (maximum actin binding site occupancy), based on previous findings (16,21,23,24). The number of cross-bridges at diastole and during peak contraction was determined from absolute myosin mass transfer, but no attempt was made to distinguish weak and strongly binding states in the absence of meridional reflections.

Statistical analysis

Values are expressed as mean ± SE unless otherwise stated. Analyses to assess the effect of drug infusions and myocardial layer were performed using two-way and one-way ANOVA with Bonferroni post hoc testing to account for repeated measures. An independent 2-tailed Student’s t-test was performed to confirm differences between control and diabetic groups. Linear regression was also used to test for significant correlations. The Statistical Package Software System (SPSS v15, SPSS, Chicago, IL) was used for all analysis with values of p < 0.05 deemed significant.

Results

Animal characteristics and basal LV function

Diabetic animals had a markedly elevated blood glucose concentration (18.3 ± 1.2 vs. 5.8 ± 0.6 mmol, p < 0.001) while maintaining a lower body weight (260 ± 13 vs. 355 ± 36 g, p < 0.05) compared to controls. Mean arterial pressure before LV catheterization was not significantly different between diabetic and control animals (55.5 ± 5.9 vs. 66.0 ± 5.4 mm Hg, NS), although diabetes did cause a depression in HR of ∼80 bpm (p < 0.05). An impairment in diabetic systolic function was evident as a lower ES pressure (p < 0.05) and a mild suppression of LV contractility evident as a diminished dP/dtmax (p < 0.05) relative to the controls. Diastolic relaxation was also impaired in diabetic animals, with a ∼40% lower dP/dtmin during baseline (p < 0.05) compared to controls. There was also a nonsignificant trend toward a prolonged diastolic active relaxation time with an increase in tau (time constant of isovolumetric LV relaxation); see summary in Table S1.

Myosin mass transfer during basal contractions

Equatorial reflections from all animals showed clear 1,0 and 1,1 reflections due to the electron density around actin-myosin filaments (see Fig. 1). Representative intensity ratio sequences from a control and a diabetic rat illustrate the well-defined cardiac cycles in intensity ratio; see Fig. S1. Notably, the diabetic rat group was found to have a markedly elevated ED intensity ratio in all layers compared to controls (p < 0.05, Fig. 2 A). Minimum intensity ratio, indicative of peak systolic cross-bridge attachments, was also unaffected by myocardial depth in control rats. However, minimum intensity ratio in diabetic animals was significantly greater in all myocardial layers (p < 0.05, Fig. 2 B) compared to controls and tended to increase with depth. Here, Δintensity ratio (difference between ED and minimum ratio) in control animals decreased slightly but not significantly as myocardial depth increased (Fig. 2 C). Thus, nondiabetic rats appeared to have a similar extent of cross-bridge formation during baseline contractions throughout the examined LV wall depth. This contrasted with diabetic rats where Δintensity ratio increased with depth, and was markedly higher in the subepicardial (p < 0.05) and subendocardial (p < 0.05) layers (84% and 164%, respectively) compared to control rats (Fig. 2 C). The larger Δintensity ratio in diabetic rats suggests that myosin mass transfer is enhanced compared to control rats, but absolute myosin transfer did not agree with this finding.

Figure 2.

Figure 2

Open in a new tab

ED (A), systolic minimum (B), and systolic intensity ratio change (C) in LV of control and diabetic rats during baseline (lactate 2 ml/h), low-dose dobutamine (4.0 μg/kg/min), and high-dose dobutamine (8.0 μg/kg/min) sequential infusions in relation to myocardial layer. Control, n = 7 and diabetic n = 9. †p < 0.05, ‡p < 0.01, ††p < 0.001 vs. controls of same layer. ∗p < 0.05, #p < 0.01, ∧p < 0.001 vs. baseline of same layer.

SAXS patterns recorded following diastolic arrest with KCl infusion (see Fig. S2) were used to determine myosin mass transfer with no cross-bridge attachments and subsequently absolute myosin mass transfer during contractions. The quiescent intensity ratio found in diabetic rats (4.0–4.5) was higher than control rats and all previous reports at ∼3.0 (16,23). A trend toward a ∼15–20% lower basal peak systolic myosin mass transfer was found in all layers of diabetic hearts compared to controls (Fig. 3 A). Of importance, the number of myosin heads in the proximity of actin at ED in control rats was ∼20–40% of the total potential myosin mass transfer, thus a significant amount of myosin projections remain close to actin between contractions without developing force. A trend toward a lower basal diabetic ED myosin mass transfer relative to controls was found in all myocardial layers, and significantly lower in the subepicardium (p < 0.05, Fig. 3 B).

Figure 3.

Figure 3

Open in a new tab

Systolic (A) and ED (B) myosin mass transfer in control and diabetic rat hearts during baseline (lactate 2 ml/h), low-dose dobutamine (4.0 μg/kg/min), and high-dose dobutamine (8.0 μg/kg/min) infusion. Control, n = 7 and diabetic n = 9. †p < 0.05, ‡p < 0.01 vs. controls of same layer.

Among the rats in the diabetic group basal ED intensity ratio was positively correlated with the minimum rate of LV pressure decrease during diastole (dP/dtmin) and therefore global LV relaxation (Fig. 4). This relation in diabetic animals was significantly different in slope from zero in the subepicardial layer (p < 0.05); similar trends were also evident in other layers; see Fig. S3. As the ED intensity ratio increased in diabetic rat hearts diastolic function worsened. No such correlation was found in control animals.

Figure 4.

Figure 4

Open in a new tab

Diastolic relaxation (dP/dtmin) in relation to ED intensity ratio in individual control and diabetic rats during baseline (lactate 2 ml/h) conditions. In diabetic animals there was a significant difference in slope from zero in the subepicardium. Control, n = 7 and diabetic n = 9.

Global LV response to dobutamine

Despite the lower absolute HR in diabetic animals, the change in HR and MAP from baseline during dobutamine infusion was not significantly different between groups. Although the ejection fraction was slightly enhanced during low-dose dobutamine infusion in diabetic animals (p < 0.05), contractility was depressed during high-dose dobutamine as the LV dP/dtmax increase from baseline was greatly reduced (∼30% of control). Notably, ES LV pressure remained ∼70 mmHg lower in diabetic rats compared to controls during both low-dose dobutamine (p < 0.01) and high-dose dobutamine (p < 0.05); as detailed in Table S1. LV relaxation (dP/dtmin) in diabetic rats was impaired and remained ∼50% lower during both low-dose (p < 0.01) and high-dose dobutamine (p < 0.05) infusions when compared to control animals.

Effect of dobutamine stimulation on local myosin mass transfer and cross-bridge formation

In general, differences in ED, minimum intensity ratio, and Δintensity ratio between control and diabetic animals observed at baseline were unchanged following low- or high-dose dobutamine infusion. In response to low-dose dobutamine, systolic mass transfer increased comparably in both control (∼9%) and diabetic groups (∼8%) relative to baseline treatment across all myocardial layers (Fig. 3 A). However, systolic cross-bridge formation was ∼18% lower in diabetic animals compared to controls, notably in the subendocardium (p < 0.05). The change in ED myosin mass transfer during low-dose dobutamine infusion from baseline was similar in both groups but remained ∼30 to ∼40% lower in diabetic hearts compared to controls across all myocardial layers, especially in the subepicardium (p < 0.05, Fig. 3 B).

Systolic and ED mass transfer were further increased by high-dose dobutamine infusion in controls in contrast to the response in diabetic rodents (Fig. 3). Nevertheless, the extent of systolic myosin mass transfer was consistent across the myocardial wall within both groups, but was significantly lower in diabetics compared to controls (p < 0.05). Systolic myosin mass transfer and cross-bridge attachments decreased in some diabetic rat hearts relative to the basal state during high-dose dobutamine; see Fig. S4.

Effect of diabetes and dobutamine on interfilament spacing

Under basal conditions there was no significant transmural difference in interfilament (d1,0) spacing in either group. We did however observe a smaller ED d1,0 spacing (∼1.0 nm smaller) in diabetic animals compared to control rats in the subepicardial (p < 0.05) and subendocardial (p < 0.05) layers (Fig. 5). Neither low-dose nor high-dose dobutamine infusion had a significant effect on d1,0 spacing, see Fig. S5. Taken together, the diminished changes in myosin transfer and LV pressure development changes observed in diabetic rat hearts paralleled the reduced change in myofilament lattice expansion.

Figure 5.

Figure 5

Open in a new tab

ED d1,0 spacing in LV of control and diabetic rats during baseline (lactate 2 ml/h), low-dose dobutamine (4.0 μg/kg/min), and high-dose dobutamine (8.0 μg/kg/min) sequential infusions. Values expressed as mean ± SE. †p < 0.05 vs. controls for the same layer.

Discussion

To our knowledge, this is the first in situ study to investigate cardiac cross-bridge regulation and interfilament spacing in the early diabetic rodent heart using a synchrotron radiation SAXS protocol. Our findings show that early diabetic cardiac dysfunction is associated with an increased intensity ratio, a measure of the relative electron density shift between myosin thick filaments and actin thin filaments, and a corresponding reduction in cardiac cross-bridge formation that was in part due to displacement of myosin heads from actin filaments particularly during diastole. Furthermore, diabetic subendocardial rather than epicardial layers appear to be more susceptible to early contractile impairment as greater intensity ratios were observed with increased depth in the myocardium. This early impairment was further confirmed as cross-bridge formation remained markedly depressed following cardiac-stress testing with the β-adrenergic agonist dobutamine. These early changes occur in the absence of marked alterations in interfilament spacing and therefore, we also assume to occur in the absence of significant sarcomere length changes because d1,0 is directly related to sarcomere length in intact muscle (25). Hence, these findings provide evidence that early type 1 diabetic cardiomyopathy also results from impairment in myosin head extension and/or flexibility.

At 3 weeks post STZ induction all diabetic rats had impaired diastolic relaxation and systolic contraction as a result of persistent hyperglycemia. This is consistent with previously reported changes in cardiac function in early diabetes (26). Of importance, in early stage diabetes this cardiac dysfunction likely occurs in the absence of cardiac structural changes such as collagen accumulation (27), and suggests that other mechanisms such as subcellular alterations in the contractile apparatus must play a role.

Baseline cross-bridge disposition in diabetes

We confirmed that ED and ES intensity ratios in nondiabetic hearts were within the previously described range using synchrotron SAXS of around 0.8–2.5 (17,19). In contrast, we found that diabetic rats had a marked increase in the minimum, ED, and ES intensity ratios between 1.5 and 4.5, with higher ratios consistently found in the subendocardium. This is notable as there are no reports of diastolic or resting intensity ratios >3.5 for perfused hearts or ex vivo cardiac muscle (21,24,25). Because the 1,1 reflection is observed continuously over the cardiac cycle in our SAXS recordings of the diabetic hearts, the observation of reduced mass associated with actin is unlikely to be an artifact.

Between animal comparisons revealed that the elevated baseline intensity ratio observed in diabetic hearts at ED was directly negatively correlated with HR (see Fig. S6), and the rate of LV pressure change during contraction positively correlated with dP/dtmin, the rate of relaxation. The negative correlation between HR and intensity ratio is expected as HR is intimately related to calcium homeostasis, contractile force, and thus cardiac function (28). However, bradycardia only explains the abnormally elevated ED intensity ratio in diabetic rodents in part because the ED ratio exceeded the intensity ratio of KCl relaxed nondiabetic muscle, and in the same diabetic hearts the ratio increased with myocardial depth from the epicardium. These findings suggest that beyond a prolonged interbeat interval an underlying impairment in actin-myosin filament regulation is the primary cause of cardiac dysfunction in early DCM.

Mass transfer during basal contractions in diabetic hearts

We found that absolute basal myosin mass transfer at ED was appreciably reduced in early diabetes, in contrast to inferences drawn from the Δintensity ratio as the quiescent intensity ratio was elevated. In one of the first studies to apply SAXS to cross-circulated heart preparations, ∼50% of myosin projections remained in the presence of the thin actin filament during diastole in papillary muscle under near physiological cyclic contractions (29); even though no force is developed by these myosin heads between systolic contractions. This is similar to our control values but is markedly greater than that which we observed in early diabetes where only ∼5% of myosin projections remained in the presence of the thin filament. Displacement of myosin mass to the thin filaments during the early phase of contraction has been shown to alter the rate of force development (22). Indeed, Matsubara and associates have shown that the number of myosin heads near actin during diastole is directly related to the inotropic state in cardiac muscle (30). In diabetes myosin isoform changes are known to play a large role in the reduced cross-bridge kinetics (3,31), which is consistent with the reduced ESP in diabetic hearts. However, in the early stage of diabetes it is likely that the greatly diminished proportion of myosin heads in the proximity of actin, which also contributes to the depression in the rate of force development and cardiac dysfunction. This agrees with the previous finding in STZ-induced diabetic rat heart muscle where the impairment in contractile function was due to a reduction in total active cross-bridge number and altered kinetics rather than a change in the force of a single cross-bridge (15). As such, the early decline in diabetic cardiac function likely results from a deficit in strongly bound cross-bridge attachments.

Nonuniformity of transmural cross-bridge regulation in diabetes

The subendocardium has long been thought to be the most susceptible myocardial region to ischemic injury (32), although more recently susceptibility has been shown to be more closely related to the extent of collateral vessels (33). Nevertheless, our findings in diabetic rodents show that impairment in cross-bridge dynamics initially occurs to a greater extent in the subendocardial region. Preliminary evaluation of cross-bridge disposition in progressively deeper myocardial layers of the diabetic heart uncovered a strong trend toward more exaggerated dysfunction due to reduced myosin mass transfer, but 1,1 reflection intensity is no longer consistently present for detailed analysis. Greater subendocardial pathology appears to be consistent with reports in human type 2 diabetes patients who have no history of clinical heart disease but present initially with subendocardial LV dysfunction at rest and during peak stress (34). One difficulty associated with the in situ method is that diffraction recordings from the deeper subendocardial and subepicardial layers contain information about the epicardial regions, making it challenging to delineate specific contributions due to the possibility of differential activation. Nevertheless, these findings show that subendocardial rather than epicardial impairment in cross-bridge regulation occurs preferentially in early diabetes.

No clear role for interfilament spacing in early diabetic dysfunction

Interfilament spacing at ED is directly correlated with the number of cross-bridges that can form attachments in the next cardiac cycle (19). Potentially, the increased diabetic intensity ratio and reduction in cross-bridge formation could result from a change in lattice spacing as demonstrated in relaxed skinned cardiac muscle (35) and/or reduced myofilament overlap at short sarcomere length (36). However, in diabetics a slightly smaller ED d1,0 was balanced by normal d1,0 spacing during isovolumetric contraction compared to controls.

Diminished cross-bridge recruitment during β-adrenergic stimulation

A major finding of this study is that dobutamine stimulation uncovered a blunted β-adrenoceptor response in diabetic systolic cross-bridge formation and diastolic mass transfer compared to both our control rats and another study in nondiabetic mice (20). The data presented here clearly show that diastolic myosin mass transfer is an important determinant of cross-bridge recruitment in positive inotropy. In control rat hearts diastolic myosin mass transfer was 60–80% at the higher dobutamine dose. This pronounced reduction in the intensity ratio is assumed to be due to radial mass transfer of myosin projections and not azimuthal movement as first demonstrated by Haselgrove and Huxley (37).Therefore, it is suggested that an increase in diastolic mass transfer results in an increase in weak cross-bridge formation as these projections do not develop force. In the early stage of diabetic heart, we found little evidence of a significant increase in the proximity of myosin heads to actin filaments or an appreciable increase in global cardiac contractility during dobutamine stimulation. A trend for a small increase in diastolic myosin mass transfer in the deeper myocardial layers did not result in a concomitant increase in systolic mass transfer. β-adrenergic stimulation did not affect diabetic lattice spacing (20) and therefore we assume did not alter the sarcomere length operating range. It should be noted however that neither low-dose nor high-dose dobutamine was sufficient to increase HR from baseline in either group of open-chest rats. This contrasts with closed-chest mice where β-adrenoceptor activation evoked a modest increase in HR (20). The lack of a positive chronotropic response to dobutamine in our study appears to be due to the open-chest preparation. Nevertheless, the increase in diastolic myosin mass associated with actin during the increase in positive inotropy is not directly due to a shorter diastolic interval between contractions and cannot be explained by a reduced interval for relaxation of fibers to allow myosin heads to return to the myosin backbone.

Possible mechanisms underlying the reduction in the number of myosin heads in close proximity to actin filaments are unclear, however prior work suggests that myosin binding protein-C (MyBP-C) or regulatory light chain (RLC) might be important. MyBP-C plays a role in altering myosin head order, myosin head extension (38), and plays a role in sustaining cardiac force and stiffness to maintain cardiac function (39). Phosphorylation of MyBP-C leads to greater cross-bridge extension from the thick filament backbone and an increase in myosin filament order and orientation (38). Notably, in Drosophila it has been shown that RLC is a more important determinant of the orientation and proximity of myosin heads to actin and twitch force development than interfilament spacing (40). Therefore, the roles MyBP-C and RLC play in reduced myosin head interaction and cross-bridge formation in early stage diabetes requires further investigation. Although alterations in troponin-I and myosin heavy chain isoforms have been demonstrated to be a major determinant of reduced cross-bridge kinetics in diabetic cardiac muscle (3,31) neither of these changes influence myosin head extension.

In summary, we have shown with real-time synchrotron x-ray diffraction, that early in diabetes, impairment in the regulation of myosin head extension occurs in the heart, contributing to the reduction in its ability to increase cardiac force. Our data show that in diabetes, myosin heads are displaced from actin filaments in beating heart muscle fibers. A lack of diastolic myosin mass transfer to actin was directly related to impaired LV function and contributed to the reduction in systolic strong cross-bridge formation during increased positive inotropic stimulation and a progressively greater impairment in deeper myocardial layers.

Acknowledgments

The authors acknowledge support to M.J.J. and J.T.P from the Access to Major Research Facilities Programme (Australian Nuclear Science and Technology Organisation, AMRFP proposal AS-IA101), to M.S. from the Ministry of Health, Labour and Welfare, and a Grant-in-Aid (A) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (No. 20590242, 23650213, 23249038), and to D.J.K. from a National Health and Medical Research Council (NH&MRC) Program Grant (No. 546272). The authors declare no conflict of interest.

Supporting Material

Document S1. Materials, methods, six figures, and reference (41)
mmc1.pdf (1.1MB, pdf)

References

  • 1.Rubler S., Dlugash J., Grishman A. New type of cardiomyopathy associated with diabetic glomerulosclerosis. Am. J. Cardiol. 1972;30:595–602. doi: 10.1016/0002-9149(72)90595-4. [DOI] [PubMed] [Google Scholar]
  • 2.Kannel W.B., Hjortland M., Castelli W.P. Role of diabetes in congestive heart failure: the Framingham study. Am. J. Cardiol. 1974;34:29–34. doi: 10.1016/0002-9149(74)90089-7. [DOI] [PubMed] [Google Scholar]
  • 3.Kita Y., Shimizu M., Mabuchi H. Correlation between myocardial dysfunction and changes in myosin isoenzymes in diabetic rat hearts. J. Diabetes Complications. 1996;10:38–42. doi: 10.1016/1056-8727(94)00064-6. [DOI] [PubMed] [Google Scholar]
  • 4.Machackova J., Barta J., Dhalla N.S. Molecular defects in cardiac myofibrillar proteins due to thyroid hormone imbalance and diabetes. Can. J. Physiol. Pharmacol. 2005;83:1071–1091. doi: 10.1139/y05-121. [DOI] [PubMed] [Google Scholar]
  • 5.Liu X., Takeda N., Dhalla N.S. Troponin I phosphorylation in heart homogenate from diabetic rat. Biochim. Biophys. Acta (BBA) 1996;1316:78–84. doi: 10.1016/0925-4439(96)00007-5. [DOI] [PubMed] [Google Scholar]
  • 6.Shimabukuro M., Higa S., Takasu N. Cardioprotective effects of troglitazone in streptozotocin-induced diabetic rats. Metabolism. 1996;45:1168–1173. doi: 10.1016/s0026-0495(96)90018-9. [DOI] [PubMed] [Google Scholar]
  • 7.Malhotra A., Sanghi V. Regulation of contractile proteins in diabetic heart. Cardiovasc. Res. 1997;34:34–40. doi: 10.1016/s0008-6363(97)00059-x. [DOI] [PubMed] [Google Scholar]
  • 8.Dhalla N.S., Liu X., Takeda N. Subcellular remodeling and heart dysfunction in chronic diabetes. Cardiovasc. Res. 1998;40:239–247. doi: 10.1016/s0008-6363(98)00186-2. [DOI] [PubMed] [Google Scholar]
  • 9.Flagg T.P., Cazorla O., Nerbonne J.M. Ca2+-independent alterations in diastolic sarcomere length and relaxation kinetics in a mouse model of lipotoxic diabetic cardiomyopathy. Circ. Res. 2009;104:95–103. doi: 10.1161/CIRCRESAHA.108.186809. [DOI] [PubMed] [Google Scholar]
  • 10.Zhao X.Y., Hu S.J., Xia Q. Decreased cardiac sarcoplasmic reticulum Ca2+-ATPase activity contributes to cardiac dysfunction in streptozotocin-induced diabetic rats. J. Physiol. Biochem. 2006;62:1–8. doi: 10.1007/BF03165800. [DOI] [PubMed] [Google Scholar]
  • 11.Ishikawa T., Kajiwara H., Kurihara S. Alterations in contractile properties and Ca2+ handling in streptozotocin-induced diabetic rat myocardium. Am. J. Physiol. 1999;277:H2185–H2194. doi: 10.1152/ajpheart.1999.277.6.H2185. [DOI] [PubMed] [Google Scholar]
  • 12.Shen X., Zheng S., Epstein P.N. Protection of cardiac mitochondria by overexpression of MnSOD reduces diabetic cardiomyopathy. Diabetes. 2006;55:798–805. doi: 10.2337/diabetes.55.03.06.db05-1039. [DOI] [PubMed] [Google Scholar]
  • 13.Oh-Ishi M., Ueno T., Maeda T. Proteomic method detects oxidatively induced protein carbonyls in muscles of a diabetes model Otsuka Long-Evans Tokushima Fatty (OLETF) rat. Free Radic. Biol. Med. 2003;34:11–22. doi: 10.1016/s0891-5849(02)01239-x. [DOI] [PubMed] [Google Scholar]
  • 14.Sumandea M.P., Steinberg S.F. Redox signaling and cardiac sarcomeres. J. Biol. Chem. 2011;286:9921–9927. doi: 10.1074/jbc.R110.175489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Joseph T., Coirault C., Lecarpentier Y. Changes in cross-bridge mechanical properties in diabetic rat cardiomyopathy. Basic Res. Cardiol. 2005;100:231–239. doi: 10.1007/s00395-005-0512-5. [DOI] [PubMed] [Google Scholar]
  • 16.Matsubara I., Kamiyama A., Suga H. X-ray diffraction study of contracting heart muscle. J. Mol. Biol. 1977;111:121–128. doi: 10.1016/s0022-2836(77)80118-6. [DOI] [PubMed] [Google Scholar]
  • 17.Yagi N., Shimizu J., Kajiya F. X-ray diffraction from a left ventricular wall of rat heart. Biophys. J. 2004;86:2286–2294. doi: 10.1016/S0006-3495(04)74286-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Pearson J.T., Shirai M., Yagi N. In situ measurements of cross-bridge dynamics and lattice spacing in rat hearts by x-ray diffraction: sensitivity to regional ischemia. Circulation. 2004;109:2976–2979. doi: 10.1161/01.CIR.0000133322.19340.EF. [DOI] [PubMed] [Google Scholar]
  • 19.Pearson J.T., Shirai M., Yagi N. Effects of sustained length-dependent activation on in situ cross-bridge dynamics in rat hearts. Biophys. J. 2007;93:4319–4329. doi: 10.1529/biophysj.107.111740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Toh R., Shinohara M., Yagi N. An X-Ray diffraction study on mouse cardiac cross-bridge function in vivo: effects of adrenergic β-stimulation. Biophys. J. 2006;90:1723–1728. doi: 10.1529/biophysj.105.074062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yagi N., Saeki Y., Kurihara S. Cross-bridge and calcium behavior in ferret papillary muscle in different thyroid states. Jpn. J. Physiol. 2001;51:319–326. doi: 10.2170/jjphysiol.51.319. [DOI] [PubMed] [Google Scholar]
  • 22.Matsubara I. X-ray diffraction studies of the heart. Annu. Rev. Biophys. Bioeng. 1980;9:81–105. doi: 10.1146/annurev.bb.09.060180.000501. [DOI] [PubMed] [Google Scholar]
  • 23.Matsubara I., Umazume Y., Yagi N. Lateral filamentary spacing in chemically skinned murine muscles during contraction. J. Physiol. 1985;360:135–148. doi: 10.1113/jphysiol.1985.sp015608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Sowerby A.J., Harries J., Stier A. X-ray diffraction studies of whole rat heart during anoxic perfusion. Biochem. Biophys. Res. Commun. 1994;202:1244–1251. doi: 10.1006/bbrc.1994.2064. [DOI] [PubMed] [Google Scholar]
  • 25.Yagi N., Okuyama H., Kajiya F. Sarcomere-length dependence of lattice volume and radial mass transfer of myosin cross-bridges in rat papillary muscle. Pflugers Arch. 2004;448:153–160. doi: 10.1007/s00424-004-1243-z. [DOI] [PubMed] [Google Scholar]
  • 26.Cheng X., Cheng X.S., Pang C.C. Inhibition of iNOS augments cardiovascular action of noradrenaline in streptozotocin-induced diabetes. Cardiovasc. Res. 2004;64:298–307. doi: 10.1016/j.cardiores.2004.06.023. [DOI] [PubMed] [Google Scholar]
  • 27.Stilli D., Lagrasta C., Quaini F. Preservation of ventricular performance at early stages of diabetic cardiomyopathy involves changes in myocyte size, number and intercellular coupling. Basic Res. Cardiol. 2007;102:488–499. doi: 10.1007/s00395-007-0665-0. [DOI] [PubMed] [Google Scholar]
  • 28.Katz A.M., Lorell B.H. Regulation of cardiac contraction and relaxation. Circulation. 2000;102(Suppl 4):IV69–IV74. doi: 10.1161/01.cir.102.suppl_4.iv-69. [DOI] [PubMed] [Google Scholar]
  • 29.Matsubara I., Suga H., Yagi N. An X-ray diffraction study of the cross-circulated canine heart. J. Physiol. 1977;270:311–320. doi: 10.1113/jphysiol.1977.sp011954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Matsubara I., Yagi N., Endoh M. The state of cardiac contractile proteins during the diastolic phase. Jpn. Circ. J. 1982;46:44–48. doi: 10.1253/jcj.46.44. [DOI] [PubMed] [Google Scholar]
  • 31.Rundell V.L.M., Geenen D.L., de Tombe P.P. Depressed cardiac tension cost in experimental diabetes is due to altered myosin heavy chain isoform expression. Am. J. Physiol. Heart Circ. Physiol. 2004;287:H408–H413. doi: 10.1152/ajpheart.00049.2004. [DOI] [PubMed] [Google Scholar]
  • 32.Reimer K.A., Lowe J.E., Jennings R.B. The wavefront phenomenon of ischemic cell death. 1. Myocardial infarct size vs. duration of coronary occlusion in dogs. Circulation. 1977;56:786–794. doi: 10.1161/01.cir.56.5.786. [DOI] [PubMed] [Google Scholar]
  • 33.Leshnower B.G., Sakamoto H., Gorman R.C. Progression of myocardial injury during coronary occlusion in the collateral-deficient heart: a non-wavefront phenomenon. Am. J. Physiol. Heart Circ. Physiol. 2007;293:H1799–H1804. doi: 10.1152/ajpheart.00590.2007. [DOI] [PubMed] [Google Scholar]
  • 34.Vinereanu D., Nicolaides E., Fraser A.G. Subclinical left ventricular dysfunction in asymptomatic patients with Type II diabetes mellitus, related to serum lipids and glycated haemoglobin. Clin. Sci. 2003;105:591–599. doi: 10.1042/CS20030168. [DOI] [PubMed] [Google Scholar]
  • 35.Martyn D.A., Adhikari B.B., Yu L.C. Response of equatorial x-ray reflections and stiffness to altered sarcomere length and myofilament lattice spacing in relaxed skinned cardiac muscle. Biophys. J. 2004;86:1002–1011. doi: 10.1016/S0006-3495(04)74175-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Millman B.M. The filament lattice of striated muscle. Physiol. Rev. 1998;78:359–391. doi: 10.1152/physrev.1998.78.2.359. [DOI] [PubMed] [Google Scholar]
  • 37.Haselgrove J.C., Huxley H.E. X-ray evidence for radial cross-bridge movement and for the sliding filament model in actively contracting skeletal muscle. J. Mol. Biol. 1973;77:549–568. doi: 10.1016/0022-2836(73)90222-2. [DOI] [PubMed] [Google Scholar]
  • 38.Weisberg A., Winegrad S. Alteration of myosin cross-bridges by phosphorylation of myosin-binding protein C in cardiac muscle. Proc. Natl. Acad. Sci. USA. 1996;93:8999–9003. doi: 10.1073/pnas.93.17.8999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Nyland L.R., Palmer B.M., Vigoreaux J.O. Cardiac myosin binding protein-C is essential for thick-filament stability and flexural rigidity. Biophys. J. 2009;96:3273–3280. doi: 10.1016/j.bpj.2008.12.3946. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Farman G.P., Gore D., de Tombe P.P. Myosin head orientation: a structural determinant for the Frank-Starling relationship. Am. J. Physiol. Heart Circ. Physiol. 2011;300:H2155–H2160. doi: 10.1152/ajpheart.01221.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Streeter D.D., Jr., Spotnitz H.M., Sonnenblick E.H. Fiber orientation in the canine left ventricle during diastole and systole. Circ. Res. 1969;24:339–347. doi: 10.1161/01.res.24.3.339. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Materials, methods, six figures, and reference (41)
mmc1.pdf (1.1MB, pdf)

Articles from Biophysical Journal are provided here courtesy of The Biophysical Society

RESOURCES