Heterogeneous abnormalities of in-vivo left ventricular calcium influx and function in mouse models of muscular dystrophy cardiomyopathy

Mice were anaesthetized using 5% isofluorane and anaesthesia maintained at 1.5% in oxygen with a flow rate of 0.5 L/min. The tail-vein was cannulated and mice placed on a MR compatible sled with surface ECG electrodes on the chest wall, respiration pillow and cutaneous temperature probe (Dazai Research Instruments, MICe, Toronto, Canada). This was connected to MR compatible monitoring equipment (SA Instruments Inc., Stony Brook, NY 11790). Mice on the sled were then placed on a bed holding the nose cone for anaesthesia delivery and then all of this slid into a 39 mm diameter quadrature birdcage volume coil (Rapid Biomedical GmbH). Images were acquired on a 7 Tesla horizontal bore microimaging system equipped with a 12-cm microimaging gradient insert (maximum gradient 40 gauss/cm) (Varian Inc., Palo Alto, CA, USA). Following power calibration and global shimming a series of four pilot transverse images were acquired over the heart. Single slice coronal and sagittal images were then obtained in order to view the apex and mitral valve planes. These images were used to plan for the true short axis plane [14]. To measure left ventricular function, contiguous short axis slices were acquired to cover the entire left ventricle using a spoiled gradient-echo cine sequence (TR=5 ms, TE=1.42 ms, flip angle 15°, FOV 30×30 mm, data matrix 128×128, 1 mm slice thickness, max. of 10 slices depending on heart size, 4 averages). Images were ECG triggered to the R wave with a cine delay of 15 ms and typically 30 phases were acquired distributed through the cardiac cycle (depending on HR). Images were zero-filled to a matrix size of 256×256. Scans were converted to matfiles (using a matlab script kindly provided by Dr Johannes Riegler, University College London, UK) and analysed using the freely available analysis software Segment v1.8 (http://​segment.​heiberg.​se) [15] to measure left ventricular mass and the LV functional parameters – end-systolic and end-diastolic volumes, stroke volume, ejection fraction and cardiac output.

For the manganese contrast enhancement, 60 mM manganese chloride (Sigma-Aldrich 244589) was given by intravenous infusion through the tail vein cannula at a flow-rate of 0.6 ml/hour, flow time adjusted according to weight to give a total dose of 190 nmol/g body weight. For a 30 g mouse, this would result in a 9.5 minute infusion. Gradient echo short axis images at the level of the papillary muscles (T₁ weighted parameters: TR=35 ms, TE=3.5 ms, flip angle 60°, FOV 30 mm × 30 mm, data matrix 128 × 128, 1 mm slice thickness, 6 averages) were taken. Four baseline images were acquired in order to average any variations due to changes in TR as a result of fluctuations in heart rate and then at 5 minute intervals for 50 minutes during and after manganese infusion. A relative increase in T₁ weighted contrast indicates increased manganese uptake (Figure 1). Images acquired from each timepoint were extracted using a Python script (kindly written by Dr Conor Lawless, Newcastle University), opened in Image J (http://​rsb.​info.​nih.​gov/​ij/​) and converted to a stack using the stack builder plugin. On the first image an area of interest was drawn to fit inside the myocardium and average signal intensity measured. Minor adjustments of this drawn region were made for subsequent images in the stack and the increase in myocardial contrast enhancement expressed as a percentage increase from the average of the four baseline images (which showed little or no variation).

The L-type calcium channel antagonist diltiazem (5–10 mg/kg body weight i.p., 10 mg/kg being maximal tolerated dose, Sigma Chemical Co., St Louis, MO) was given separately to subgroups of the younger mice. The hyperpolarization-activated cyclic nucleotide-gated (HCN) channel inhibitor ZD 7288 (0.3 mg/kg body weight i.p., Sigma Chemical Co., St Louis, MO) was also given to separate groups to study the effects of selective heart rate reduction on manganese contrast enhancement [16, 17]. For the ZD 7288 experiments we also included a 2.5 minute scan as we were particularly interested in the early kinetics of manganese uptake related to heart rate. These medications were given prior to placing the animal in the scanner which was approximately 30–40 minutes before the manganese infusion.

To test effects of time and age on manganese-contrast enhancement, repeated measures of analysis of variance was used combining data from young and old groups. Within each age group, analysis of variance was used to test differences amongst groups comparing an average value of steady state manganese contrast enhancement, and MR functional data. Post-hoc testing was performed with the Scheffe′ test. Single comparisons were made with Student’s t-tests. Data are expressed as mean ± standard error of the mean.

Age had significant effects on haemodynamics (ANOVA, all 3 groups combined), with significant increases in end-diastolic volume (P<0.01), end-systolic volume, and stroke volume (both P<0.05) (data not shown). As there were significant differences in body weights, left ventricular mass, volume measurements and cardiac output are indexed to body weight (i.e. cardiac index), and with that changes in volumes were no longer significant (Table 1). Sgcd−/− mice had lower body weights, higher heart rates, increased left ventricular mass index at the older age, normal volumes, and increased cardiac index at both ages. The increased cardiac index was a result of the higher heart rate, as stroke volume index was not significantly different to WT mice. The mdx mice had significant left ventricular hypertrophy in the older age group (though left ventricular mass index did not actually change in these mice with age, rather it was decreased in older WT mice), and reduced stroke volume index when considering both age groups combined compared to WT. The mdx mice also had significantly lower ejection fractions compared to the Sgcd−/− mice. Thus, there were markedly different haemodynamic characteristics in the 2 muscular dystrophy mice models. Whereas both had left ventricular hypertrophy at older ages when compared to older WT, there was a hyperdynamic circulation in the Sgcd−/− mice with increased cardiac index, and left ventricular systolic dysfunction with reduced stroke volume in the mdx mice. Average surface temperatures measured during the scans were not significantly different between the groups (WT: 34.3±0.3°C, mdx 33.5±0.4°C and Sgcd−/− 34.3±0.3°C (ANOVA p=NS).

Contrast-enhancement was significantly slower in all older mouse groups compared to younger mice (Figures 2A and B, P<0.05, time*age interaction, repeated measures ANOVA combining both age groups and all 3 groups). For instance, in older WT mice contrast-enhancement at 5 minutes was 29±5% compared with 53±5% in younger WT mice. This analysis also showed that when considering both ages together contrast enhancement was significantly elevated in the Sgcd−/− mice relative to controls (P<0.05 post hoc Scheffe test), and this was of borderline significance for the mdx mice (P=0.08). To study contrast-enhancement within the 2 age groups, a mid point value of contrast enhancement was defined as the mean contrast enhancement of the steady state period of 10 to 25 minutes after start of the manganese infusion, with 5 minute interval scans. In the older mice, mid point contrast enhancement was borderline increased in the Sgcd−/− compared to WT mice (P=0.07, post hoc Scheffe test), and not significantly different in the mdx mice (Figure 2C). In the younger mice, mid point contrast enhancement was significantly increased in both the Sgcd−/− and mdx compared to WT mice (both P<0.05, post hoc Scheffe test). As the Sgcd−/− mice were on a different strain background to the C57Bl/10 controls we performed initial experiments to determine if there was a strain difference between C57Bl/10 and C57Bl/6 mice. In these experiments the mid point contrast enhancement was 50±6% in C57Bl/10 (N=5) and 50±10.0% in C57Bl/6 mice (N=3, p=NS) indicating that there was no strain effect.

We concentrated on the younger mice for the pharmaceutical studies as these were the ages where the most significant increases in contrast enhancement were seen. The L-type calcium channel blocker diltiazem (5 mg/kg i.p.) significantly reduced contrast enhancement (Figure 3A) in WT (21%, P<0.05) and to a greater extent in mdx mice (38%, P<0.001), though had no significant effects in the Sgcd−/− mice (6%, P=NS). A higher dose of diltiazem (10 mg/kg) did produce significant reductions in contrast enhancement in Sgcd−/− mice (21%, P<0.05), indicating that the L-type calcium channels were partially resistant in the Sgcd−/− mice. Consistent with the variable sensitivity to L-type calcium channel blockade, diltiazem 5 mg/kg significantly reduced heart rates in the WT and mdx mice, but not in the Sgcd−/− mice, though 10 mg/kg did produce significant reductions in the Sgcd−/− mice (Figure 3B).

There was increased contrast-enhancement in the Sgcd−/− mice which accompanied a significantly higher heart rate compared to WT mice (Table 1), and thus the mechanism of this increased contrast-enhancement could be due to more heart beats and action potentials transmitting Mn²⁺ into the cell, though could also be related to a cellular channel defect with increased Mn²⁺ entry into the myocyte per action potential. To clarify this issue we used the HCN channel inhibitor ZD 7288 to selectively reduce heart rate in all 3 groups of younger mice. ZD 7288 significantly reduced contrast enhancement in all 3 groups of mice, and this was associated with significant reductions in heart rate (Figure 4A and B). This does suggest in general that the manganese contrast enhancement signal is sensitive to heart rate, and care must therefore be taken when comparing mouse models with different heart rates with this technique. However, when comparing the time plots of contrast enhancement and heart rates in the WT and Sgcd−/− mice (Figure 4C and D) we see that at a very early stage in the increase in contrast enhancement (2.5 and 5 minutes) when heart rates of both groups of mice are very similar that there was still a marked increase in contrast-enhancement in the Sgcd−/− mice compared to wild type. Thus, while contrast-enhancement is sensitive to heart rate, by reducing the heart rate in the Sgcd−/− mice to WT levels we can see that there is clearly a component of increased cellular uptake of Mn²⁺ per action potential. The pattern of early increased contrast enhancement was not however seen in the mdx mice. For instance, at 2.5 minutes with similar heart rates to WT mice contrast enhancement was 22±5% in the mdx mice compared to 17±6% in the WT mice (p=NS), so this illustrates a different pattern to the Sgcd−/− mice, and potentially different mechanism of abnormal contrast enhancement.

We were interested in L-type calcium channel blockers for 2 reasons – 1) they can reduce manganese contrast enhancement [11], and 2) have differential effects in the mdx and Sgcd−/− mice [8]. It is known that intracellular calcium is increased in muscular dystrophy cardiomyopathy [7, 18]. Cohn et al. [8] have shown that the L-type calcium channel blocker verapamil had significant beneficial effects in the Sgcd−/− mice on troponin I levels and cardiac histology, though not in the mdx mice. To reconcile the findings in this study and our findings it is necessary to examine doses used. The doses used in Cohn et al. compared to the present one are different when corresponding human doses calculated by weight are examined, though a consistent theme is that high doses are required. The doses of diltiazem in the current study of 5–10 mg/kg would equal 350–700 mg in a 70 kg human, the upper limit of a normal human dose being 240 mg. The dose of verapamil used by Cohn et al. was 3.5 mg/day, which would be approximately 35 fold higher than an appropriate human dose. One possible consistent explanation is that only higher doses of L-type calcium channel antagonists are effective in the Sgcd−/− mice, as we only show responses to higher doses of diltiazem. There are strain differences in the backgrounds of the wild type and mdx (both C57Bl/10) versus the Sgcd−/− mice (C57Bl/6). Though we could find no differences between Bl/10 and Bl/6 mice in terms of manganese uptake, we cannot rule out some strain effect in the response to L-type calcium channel blockade.

When comparing models with different heart rates, it may be difficult to determine whether a difference in manganese contrast enhancement is due to the increased heart rate resulting in more action potentials over the period of the manganese infusion or whether there is an underlying cellular defect with an altered calcium influx per action potential [19]. To determine whether the increased heart rate in the Sgcd−/− mice accounts for the increased manganese contrast enhancement we have used the HCN channel inhibitor ZD 7288 which inhibits the I _f channel producing a selective reduction in heart rate [16, 17]. These studies clearly show that at matched heart rates, the Sgcd−/− mice have increased manganese contrast enhancement compared to wild type, indicating a primary cellular defect in calcium influx. These experiments also show clearly the important role that heart rate plays in manganese contrast enhancement, so that for instance, in our experiments with diltiazem it is reasonable to assume that some of the reduction in contrast enhancement is due to a reduction in heart rate. Of note, a small amount of calcium may enter cardiac myocytes through the I _f channel (0.5% of the current from the HCN2 channel) [20], though this small amount is unlikely to have significantly effected our results. Furthermore, as the mdx mice do not have significantly different heart rates compared to wild type mice the increase in manganese contrast enhancement will be due to a primary cellular defect in calcium influx. Surface temperatures of the mice were below physiological levels during the scans, and so heart rates were also low. Surface temperatures are approximately 1°C lower than core body temperature [21]. Higher temperatures would have likely resulted in higher heart rates and this may alter manganese uptake rates.

Isenberg et al. [22] have shown that in aged mice calcium transients peak late and decay slowly. Salameh et al. [23] have shown that the L-type calcium (I_Ca.L) current is abnormal in aging rabbit hearts with the I_Ca.L density normalised to cell volume significantly reduced, maximum conductance also significantly decreased and steady state inactivation shifted to more positive potentials in aged hearts. However, expression of the α_1c subunit of the L-type channel, the SERCA2a-ATPase, and the Na⁺/Ca²⁺ exchanger did not differ significantly between the two age groups. Direct comparison of cellular calcium handling and MECMR would be very informative in studying age-related changes in calcium influx. Our young and old mice were separate groups and in future, serial MECMR studies within the same mouse will also be important.

We have previously shown that there are differences in function between the mdx and Sgcd−/− mouse models using the conductance catheter [10], and these results are very consistent despite the very different techniques, though adding information about left ventricular mass and calcium influx. Townsend et al. [24] have compared these mouse models and also found distinct differences (at 6 – 8 months of age). Isolated myocytes from the mdx mice show reduced compliance and are susceptible to terminal contracture, whereas the Sgcd−/− mice had normal compliance. There was increased fibrosis in the Sgcd−/− hearts, though not in the mdx. We have also shown that mdx mice exhibit improved hemodynamics with 8 weeks of β-blocker treatment though there is significant deterioration in the Sgcd−/− mice with reduced heart rate, cardiac output and impaired active relaxation [10]. Despite this information a specific molecular pathway by which these differences occur is still unclear.

We have shown at an early stage in the development of muscular dystrophy cardiomyopathy that in-vivo calcium influx is increased. Increased calcium influx and left ventricular hypertrophy are seen in both models, though other important haemodynamic and pharmacological differences are apparent, so that all these factors need to be considered together when assessing mechanisms of the cardiomyopathy and response to therapies. A limitation of these in-vivo techniques is that it is difficult to say definitively if there are specific abnormalities in channels such as the L-type calcium channel in these models. Thus, the future utility of these non-invasive in-vivo techniques will likely be in serially assessing responses to novel preclinical treatments.