Early development of intracellular calcium cycling defects in intact hearts of spontaneously hypertensive rats

Abstract

Defects in excitation-contraction coupling have been reported in failing hearts, but little is known about the relationship between these defects and the development of heart failure (HF). We compared the early changes in intracellular Ca²⁺ cycling to those that underlie overt pump dysfunction and arrhythmogenesis found later in HF. Laser-scanning confocal microscopy was used to measure Ca²⁺ transients in myocytes of intact hearts in Wistar-Kyoto (WKY) rats and spontaneously hypertensive rats (SHRs) at different ages. Early compensatory mechanisms include a positive inotropic effect in SHRs at 7.5–9 mo compared with 6 mo. Ca²⁺ transient duration increased at 9 mo in SHRs, indicating changes in Ca²⁺ reuptake during decompensation. Cell-to-cell variability in Ca²⁺ transient duration increased at 7.5 mo, decreased at 9 mo, and increased again at 22 mo (overt HF), indicating extensive intercellular variability in Ca²⁺ transient kinetics during disease progression. Vulnerability to intercellular concordant Ca²⁺ alternans increased at 9–22 mo in SHRs and was mirrored by a slowing in Ca²⁺ transient restitution, suggesting that repolarization alternans and the resulting repolarization gradients might promote reentrant arrhythmias early in disease development. Intercellular discordant and subcellular Ca²⁺ alternans increased as early as 7.5 mo in SHRs and may also promote arrhythmias during the compensated phase. The incidence of spontaneous and triggered Ca²⁺ waves was increased in SHRs at all ages, suggesting a higher likelihood of triggered arrhythmias in SHRs compared with WKY rats well before HF develops. Thus serious and progressive defects in Ca²⁺ cycling develop in SHRs long before symptoms of HF occur. Defective Ca²⁺ cycling develops early and affects a small number of myocytes, and this number grows with age and causes the transition from asymptomatic to overt HF. These defects may also underlie the progressive susceptibility to Ca²⁺ alternans and Ca²⁺ wave activity, thus increasing the propensity for arrhythmogenesis in HF.

heart failure (HF) is a chronic maladaptive state in which perturbations of Ca²⁺ cycling occur concurrent with deteriorating cardiac function and increasing arrhythmogenesis (22). Cellular defects in cardiac excitation-contraction (E-C) coupling have been identified in failing hearts from human and animal models and contribute to mechanical and electrophysiological dysfunction (49). One particularly serious defect is a disruption in the T-tubule network (5) so that junctional sarcoplasmic reticulum (SR) and ryanodine receptors (RYRs) that regulate Ca²⁺ release from the SR are isolated from the normal trigger supplied by L-type Ca²⁺ channels (LCCs). Activation occurs instead by diffusion from neighboring release units, a defect that was first identified in spontaneously hypertensive rats (SHRs) (45) but also found in mouse (35), human (34) and dog (5) hearts. The ensuing defective activation of “orphaned” RyRs (45) causes a poorly coordinated release of Ca²⁺ throughout the cell. In combination with a slowing of reuptake because of reduced activity of the sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA) (21), the result is a serious dysfunction in Ca²⁺ cycling.

In contrast to overt HF, there is little information on the progression of defects in Ca²⁺ cycling and E-C coupling and their electrophysiological implications during HF development. In the SHR model, early compensation in response to hypertension includes a period during which Ca²⁺ sparks are increased, suggesting an early positive inotropic response (44). However, the subsequent changes in Ca²⁺ cycling involved in the progression from compensated HF to the clinically apparent decompensated HF, both in animals and humans, are very poorly understood. Several studies have focused on active myocardial properties, fibrosis, Ca²⁺ sparks, and Ca²⁺ transients during the transition from hypertension to HF (46, 56). However, a comprehensive study of changes in Ca²⁺ cycling during the development of HF is lacking.

In addition, although previous aging studies have demonstrated negative inotropy in rodents and positive inotropic effects in sheep (15, 23, 25, 39, 58), there are few data regarding the Ca²⁺ handling properties at different stages throughout the aging process. Even fewer studies have compared Ca²⁺ cycling changes during normal aging to aging in HF animals. It is also important to note that all of these previous observations were made in isolated myocytes, with little information available about Ca²⁺ cycling at the cellular level in the intact heart.

The goal of this study was to investigate the early changes in intracellular Ca²⁺ cycling that occur in response to hypertension and before the development of overt HF in an intact rat heart model. These results from young animals were also compared with those obtained in old SHRs with overt HF similar to those reported by others (45). We tested the hypothesis that defects in intracellular Ca²⁺ cycling in HF are distinct, occur early (long before HF develops), and differ from those seen in normal aging. We chose to study the SHR model because hypertension occurs early (within 2 to 3 mo of age) and is maintained constant throughout life (12) and therefore mimics the clinical course of many patients who develop HF secondary to hypertension, making the SHR a particularly good model of the slow onset of hypertrophy and HF in response to hypertension. To investigate the cellular function in the intact heart in response to hypertension, we used laser-scanning confocal microscopy to measure intracellular Ca²⁺ cycling with sarcomere-level resolution in myocytes of whole hearts (1).

METHODS

All animal use protocols were approved by the Institutional Animal Care and Use Committee in accordance with National Institutes of Health guidelines. Male SHRs and Wistar-Kyoto (WKY) rats were anesthetized with ketamine-xylazine (80:8 mg/kg ip), and the heart was removed and placed on a Langendorff apparatus for retrograde perfusion with a modified Tyrode solution containing (in mM) 140 NaCl, 5.4 KCl, 1.8 CaCl₂, 17 NaHCO₂, 0.5 MgCl₂, 0.4 NaH₂PO₄, and 10 glucose and 2% bovine serum albumin (by weight), bubbled with 95% O₂-5% CO₂ (pH 7.35). Four distinct age groups were studied: 6 mo (6–6.5 mo), 7.5 mo (7–8 mo), 9 mo (8.5–9.5 mo), and 22 mo (20–25 mo). The numbers were as follows (hearts/sites/myocytes): 6-mo WKY, 3/6/85; 6-mo SHR, 2/5/81; 7.5-mo WKY, 3/5/76; 7.5-mo SHR, 2/4/66; 9-mo WKY, 2/7/81; 9-mo SHR, 2/4/73; 22-mo WKY, 5/11/193; and 22-mo SHR, 6/13/212. Advanced stages of HF were demonstrated in the 22-mo population by high heart weight-to-body weight ratios (3.99 ± 0.19 mg/g in WKY rats compared with 8.88 ± 0.29 in SHRs, P < 0.001), fluid in the lungs, dyspnea, and greatly diminished activity levels in the SHRs. The heart was then placed in an experimental chamber on the stage of a Zeiss LSM510 laser-scanning confocal microscope. Temperature was maintained at 26 ± 1°C. Three successive additions of fluo-4 AM (10–15 μM diluted from an ethanol stock; Invitrogen) were made to the solution at 20-min intervals during recirculation, after which the heart was washed with normal solution for 10 min. Cytochalasin-D (50 μM dissolved in ethanol; Sigma) and blebbistatin (12 μM dissolved in ethanol; Sigma) were then added to the solution to prevent contraction, and recirculation was again initiated. The right atrium was crushed to slow the intrinsic heart rate. Hook platinum electrodes were inserted into the left ventricular apex so that basal pacing at a basic cycle length (BCL) of 700 ms could be interspersed with 10-s epochs of rapid pacing at different test cycle lengths (CLs).

Once dye loading was complete, a section of the left ventricular midepicardial surface was scanned to identify the sites with well-loaded myocytes. Fluo-4 was excited with 488-nm laser light from a 25-mW argon laser at ≤10% transmission, and fluorescence > 505 nm was collected via a long-pass filter. The scan line was then placed across the short axis of 10–25 myocytes, and pacing protocols were initiated. The characteristics of Ca²⁺ transients were recorded during both basal pacing (BCL = 700 ms) and rapid pacing. The rate sensitivity of Ca²⁺ transient alternans magnitude was measured as alternans ratio (53) (AR = 1 − small/large) at steady state (10 s) for each test BCL. Each epoch was followed by a 3-s pause before a return to basal pacing for 1 min before the next test train was initiated at a 10-ms shorter BCL. The effects of rapid pacing were tested in the range of BCL = 500–140 ms or until 2:1 block or tachycardia occurred. Ventricular tachycardia was defined as a train of at least five spontaneous beats following the test train. Details of these methods have been published elsewhere (1, 27, 48).

Confocal fluorescence images were measured using Zeiss and ImageJ software. Fluorescence intensities for each cell were cut from the original images, and intensity profiles were analyzed using MatLab and pCLAMP8 software. The SD of the mean for each parameter (35), also termed heterogeneity indexes (HIs) (48), was calculated to permit comparisons of the overall intercellular heterogeneity between the different sites. Gradient indexes (GIs) were calculated to compare cell-to-cell heterogeneity (or gradients between adjoining cells) and are defined as the SD of the difference between adjacent cells for each parameter. All results are expressed as means ± SE. Differences were analyzed by one-way and two-way ANOVA, followed by Student-Newman-Keuls post hoc analysis. Results were considered statistically significant at P < 0.05.

RESULTS

Characteristics of Ca²⁺ transients during basal pacing.

Figure 1 shows recordings of Ca²⁺ transients from each of the eight conditions studied during basal pacing: WKY (A, left) and SHR (A, right) hearts at 6, 7.5, 9, and 22 mo. All cells (black bars at left of each image) show simultaneous Ca²⁺ release during stimulation and subsequent decay as Ca²⁺ is removed from the cytoplasm (Fig. 1A). Figure 1B shows the average intensity profile for all WKY and SHR myocytes within the site, comparing WKY and SHRs at each time point. Figure 1C, i and ii, shows the intensity profiles of WKY and SHRs at all time points and reflects changes in Ca²⁺ cycling with age.

Fig. 1.

Line scan images comparing intracellular Ca²⁺ transients of intact Wistar-Kyoto (WKY) and spontaneously hypertensive rat (SHR) hearts. A: transverse recordings from multicellular sites of WKY (left) and SHR (right) hearts at 6, 7.5, 9, and 22 mo of age. Black bars at left indicate each cell. Two beats are shown at basal basic cycle length (BCL) = 700 ms. Intensity profiles of a single beat from a selected cell from each image are shown to the right of the line scan images and compare WKY and SHRs at the same age (B). C: superimposed intensity profiles at different ages.

Figure 1A shows changes in basal Ca²⁺ cycling that occur both in normal aging and in HF. Ca²⁺ transients in WKY and SHRs are similar in duration and amplitude at 6 mo and show little variability between myocytes. At 7.5 mo, the SHR heart demonstrated an overall positive inotropic effect but has a highly mixed population of Ca²⁺ transients in which some cells have increased amplitude and duration, whereas others are similar to the 6-mo animals, resulting in increased cell-to-cell variability. At 9 mo, most SHR cells have increased amplitude and duration but again exhibit less cell-to-cell variability. The Ca²⁺ transient duration is longer in the 22-mo SHR despite the decreased amplitude. Note also that the 22-mo WKY rat has longer Ca²⁺ transients than the younger WKY rats and lower intercellular variability in Ca²⁺ transient magnitude and duration.

The time-dependent changes in Ca²⁺ transient characteristics during early development in WKY and SHR myocytes are summarized in Fig. 2. There were no differences between Ca²⁺ transients in SHRs and WKY rats at 6 mo (Fig. 1B,i). At 7.5 mo, the Ca²⁺ transients are similar in all characteristics except for the increased amplitude of the SHR myocytes. At 9 mo (Fig. 1B,iii), the increase in amplitude persists in the SHR myocytes, but there is now an increase in Ca²⁺ transient duration at 80% of recovery (TD₈₀, Fig. 2B). Increased duration occurs as a result of both slower rising and falling phases (Fig. 2, D and E). Increased amplitude and duration result in an increased integral of Ca²⁺ release (Fig. 2C). Decreases in the maximal rate of Ca²⁺ release (+dF/dt, Fig. 2F) and maximal rate of decline (−dF/dt, Fig. 2G) are consistent with slower rise and decay times. In comparison, HF myocytes in 22-mo SHRs (Fig. 1B,iv) show smaller Ca²⁺ transient amplitudes than in WKY rats despite the increased duration (Fig. 2, B, D, and E), as is confirmed by the decreased Ca²⁺ transient integral (Fig. 2C). Both +dF/dt and −dF/dt were depressed in HF (Fig. 2, F and G).

Fig. 2.

Summary of characteristics of Ca²⁺ transients at basal pacing (BCL = 700 ms) for all groups. ±dF/dt, maximal rate of Ca²⁺ release and decline, respectively; TD₈₀, Ca²⁺ transient duration at 80% of recovery. *P < 0.05 compared with WKY at the same age. †P < 0.05 compared with 6-mo SHR. ‡P < 0.05 compared with 6-mo WKY.

The normal aging process in WKY rats causes little change in Ca²⁺ transient amplitude and duration from 6 to 9 mo (Figs. 1C,i and 2). Only at 22 mo do WKY Ca²⁺ transients show decreased amplitude and increased duration (Fig. 2, A and B) compared with younger animals. This increased duration is due to the increased decay time at 22 mo in the absence of changes in the rise time and +dF/dt (Fig. 2, D and E). In contrast, SHRs show changes in Ca²⁺ cycling starting at 7.5 mo as an increase in amplitude (Figs. 1C,ii and 2A), whereas duration changes (TD₈₀, rise and decay) occur starting at 9 mo (Fig. 2, B, D, and E).

Cell-to-cell variability of Ca²⁺ transients.

One of the most striking observations was increased cell-to-cell variability in Ca²⁺ transient characteristics within each recording site (Fig. 1A), particularly in Ca²⁺ transient duration. Figure 3, A, C, E, and G, summarizes the HI (48) for Ca²⁺ transient characteristics in WKY and SHR myocytes in each age group. HI compares intercellular heterogeneity for each parameter within a site (35). We found increased heterogeneity only in the characteristics related to duration (Fig. 3, C, E, and G) of SHR myocytes at 7.5 mo compared with both 6 and 9 mo with no changes in heterogeneity at any time during aging in WKY rats. A high variability in duration was also present in failing hearts (22-mo SHRs). These changes reflect the prolongation of Ca²⁺ transients that develops between 6 and 9 mo so that intercellular variability is greatest at 7.5 mo and the greatest uniformity occurs at 6 and 9 mo. Different processes affect E-C coupling in overt HF and will be discussed later.

Fig. 3.

WKY vs. SHRs heterogeneity indexes (HIs) and gradient indexes (GIs). Summary of HIs and GIs for basal Ca²⁺ transient characteristics as a function of age in WKY and SHRs. *P < 0.05 compared with WKY at the same age. †P < 0.05 compared with 6-mo SHR.

Further investigation of the source for intercellular variability is shown in measurements of the GIs shown in Fig. 3, B, D, F, and H, for amplitude and durational components of WKY and SHRs. GI is a measure of variability between adjoining cell pairs and is calculated as the SD of the differences in each parameter between each cell pair in a recording site. Once again, GI is greater for all three durational components of SHR myocytes at 7.5 mo and also in HF compared with WKY myocytes (Fig. 3, D, F, and H). Thus much of the intercellular variability within the sites for duration comes from the large heterogeneities that exist between adjoining cells.

Increased sensitivity to Ca²⁺ transient alternans development occurs early in SHRs.

Figure 4 shows twelve consecutive beats at steady state during rapid pacing (BCL = 300 ms) in line scan images of sites from WKY and SHRs at 6, 7.5, 9, and 22 mo. The average intensity profiles of the site are shown above each image, and the profiles of two selected cells (black bars) are shown for the last two beats at the right of each image. Ca²⁺ transient amplitude, morphology, and duration are constant at 6 mo for both WKY and SHRs. At 7.5 mo, there is a small Ca²⁺ transient alternans in the WKY rat (AR = 0.13) and a large alternans in the SHR (AR = 0.54) at the same CL, although alternans phase is identical in all cells in each site (intercellular concordant alternans). At 9 mo, the WKY site shows increased Ca²⁺ transient alternans (AR = 0.46), but the SHR site, in addition to an increased overall cellular AR, now demonstrates intercellular discordant alternans where myocytes are phase mismatched and large Ca²⁺ transients in some cells occur while neighboring myocytes respond with small Ca²⁺ transients. Note that the overall intensity profile shows less Ca²⁺ transient alternans, as the average of simultaneous large-small and small-large cycles results in an overall site average Ca²⁺ transient alternans of intermediate magnitude. At 22 mo, both WKY and SHR sites display intercellular discordant alternans, with the WKY rat showing only one discordant cell (fifth from bottom), whereas the SHR shows ∼50% discordance.

Fig. 4.

Line scan recordings of intracellular Ca²⁺ transients in cells from intact WKY and SHR hearts during rapid pacing. Images from WKY (left) and SHRs (right) at each time point are shown. Average fluorescence intensity profiles for each cell are placed above the image. Images show 12 consecutive beats at 300-ms cycle length. Intensity profiles of 2 myocytes for the final 2 beats are shown to right of line scans.

Figure 5A summarizes the rate sensitivity of AR for all conditions. The curves show fitted averages for all cells in each category (including original data points for all 22-mo SHR cells for comparison) and demonstrate that AR increases with heart rate. To quantify the rate sensitivity for Ca²⁺ transient alternans development, the estimated CL at 50% maximal alternans (ECL₅₀) was calculated (1). The HF (22-mo SHR) myocytes are the most vulnerable of all groups to alternans and are therefore shifted farthest to the right in Fig. 5A and give the largest ECL₅₀ values. Figure 5B shows increased vulnerability to alternans at both 9 and 22 mo in SHRs with no change between 6 and 7.5 mo. There were no changes in vulnerability to alternans in young WKY rats, although normal aging also increased alternans vulnerability, as shown by an increased ECL₅₀ in the oldest WKY rats compared with younger WKY rats.

Fig. 5.

Heart failure and aging increase susceptibility to Ca²⁺ transient alternans. A: fitted sigmoidal curves of Ca²⁺ alternans ratio versus cycle length for all groups. Summary data points are shown on the 22-mo curve of SHRs, demonstrating a typical quality of fit. Curves are drawn for the mean of all myocytes in each group. B: estimated cycle length at 50% maximal concordant Ca²⁺ alternans (ECL₅₀) for all data. C: cycle length at 50% maximal intercellular discordant Ca²⁺ alternans (DCL₅₀) for all data. D: threshold cycle length at which subcellular Ca²⁺ alternans (Subcell Alt) threshold occurs for all groups. *P < 0.05 compared with WKY at the same age. †P < 0.05 compared with 6-mo SHR. ‡P < 0.05 compared with 6-mo WKY.

Intercellular discordant Ca²⁺ alternans was quantified by comparing the values of the CL at which 50% of maximal discordance (DCL₅₀) is achieved. Note that DCL₅₀ is the estimated CL at which 25% of cells are phase mismatched with the other 75%, since maximal discordance occurs when 50% of cells are out of phase. Figure 5C shows a plot of DCL₅₀ for WKY and SHRs for each age group and shows more discordance in 7.5–22-mo SHRs compared with WKY rats. Intercellular discordant alternans also increases in 22-mo WKY rats compared with 6–9-mo WKY rats.

A third form of Ca²⁺ alternans, subcellular alternans, refers to Ca²⁺ alternans in which alternans in one cellular region is phase mismatched with other regions in the same cell (28). Figure 5D shows that the threshold for subcellular alternans is increased for SHRs at 7.5–22 mo over WKY rats. There is also an increased susceptibility to subcellular alternans with aging in the WKY group.

Another form of aberrant Ca²⁺ cycling behavior occurred rarely in WKY rats (Fig. 6A,i) but more commonly in SHRs. Both spontaneous (Fig. 6A,ii) and triggered Ca²⁺ waves (Fig. 6A,iii) occurred at a higher incidence at all ages in SHRs compared with WKY rats (Fig. 6B). Triggered Ca²⁺ waves are defined as Ca²⁺ waves that occur during pacing and are absent during quiescence, in contrast with spontaneous Ca²⁺ waves that occur only following pacing and during quiescence (49). These forms of Ca²⁺ cycling occur at all stages of disease development in SHRs compared with age-matched control WKY rats and are especially prevalent in HF, as we have reported previously (49).

Fig. 6.

Spontaneous and triggered Ca²⁺ waves in WKY and SHRs. A,i: Ca²⁺ transients activated by the last 3 stimulated beats (♦) (BCL = 280 ms) in a 22-mo WKY heart, followed by a 3-s pause when a spontaneous beat (◊) occurred. A,ii: a spontaneous wave (*) occurs in the second cell from top during the pause (22-mo SHR). A,iii: triggered waves during both basal and rapid pacing (arrows). B: summary of the incidence of spontaneous (Spont; B,i) and triggered (Trig; B,ii) waves in each experimental group. *P < 0.05 compared with WKY at the same age. †P < 0.05 compared with 6-mo SHR.

Altered restitution of SR Ca²⁺ release in early SHRs and aging.

A major determinant of Ca²⁺ alternans is the rate of recovery of Ca²⁺ release from the SR during rapid stimulation (48). Figure 7A shows the averaged restitution of all WKY and SHR myocytes from 6–22 mo. Data from young WKY rats show a rapid recovery of SR release with an increasing interval, with nearly full recovery at ∼450 ms. Nearly identical results occurred in 6- and 7.5-mo SHRs, but the rate of restitution in SHRs is slowed at 9 mo and even more in HF where, in many instances, recovery was so slow that it was not possible to test short enough intervals without inducing ventricular tachycardia.

Fig. 7.

Rates of restitution of sarcoplasmic reticulum Ca²⁺ release in WKY and SHRs. A: average Ca²⁺ transient restitution curves for all groups. Curves are drawn for average of all myocytes. B: cycle length at 50% fractional recovery (R₅₀) for all groups. *P < 0.05 compared with WKY at the same age. †P < 0.05 compared with 6-mo SHR. ‡P < 0.05 compared with 6-mo WKY.

Figure 7B summarizes this change in restitution by plotting the time of recovery of restitution for all the groups. R₅₀ is the calculated CL at which fractional recovery of the sigmoid fit achieved 0.5. SHR myocytes showed slower restitution compared with WKY rats at 9 and 22 mo. Aging itself also slows restitution, with R₅₀ in 22-mo WKY rats greater than that found in young animals, although this slowing is significantly less than that in HF.

DISCUSSION

The pathophysiology of HF has been studied in many animal models and has revealed numerous defects in Ca²⁺ cycling in both animal and human HF. The SHR model is particularly attractive because it mimics clinical developments common to many hypertensive patients. The early development of hypertension (11, 26) leads to a compensatory positive inotropic effect after 6 mo in SHRs (10, 44), but there is no information about the changes in Ca²⁺ cycling as this form of compensation transitions to decompensation. Contractile performance then declines slowly, ending in HF after 18 mo of age (7, 14, 40). Our investigation focused on the early changes in E-C coupling in the 6–9-mo window before decompensation occurs. We found distinct changes in E-C coupling that could explain both the late decline in pump function in HF and the increased vulnerability to ventricular arrhythmias that occurs beginning as early as 4 mo in SHRs (17, 57).

Defects in Ca²⁺ cycling develop before HF.

We found that fundamental alterations in basal Ca²⁺ transients occur long before HF develops. The early positive inotropic effect has been reported previously (44, 47), and the negative inotropic state in overt HF has also been reported in numerous studies (20, 46). The positive inotropic phase may arise because of altered interactions between LCC and RyRs in junctional SR that enhance release without changing the LCC trigger (44), although no precise mechanisms have been proposed and no changes in LCC or RyR number have been reported. In contrast, the negative inotropic state in HF may develop because of the disruption in the T-tubule network, leaving orphaned RyRs without activation by LCC, which has been found in SHRs (45) and other HF models as well (5, 34, 35).

Our results are the first to document cellular changes in Ca²⁺ cycling during the transition from positive to negative inotropy. The first change was a prolongation of Ca²⁺ transient duration that occurred in a highly heterogeneous manner between individual myocytes even during the overall positive inotropic phase. Thus the greatest variability occurred at 7.5 mo, presumably because there was a nearly equal split between cell populations with short and long Ca²⁺ transients. Decreased SERCA (or Na⁺/Ca²⁺ exchange) expression or activity could be responsible for a slowing in Ca²⁺ removal from the cytoplasm. These results are consistent with the time course and heterogeneity of decay time and −dF/dt. Furthermore, the GIs for these parameters indicate that cell-to-cell differences occur randomly between myocytes and are not the result of systematic or regional changes in Ca²⁺ reuptake. This important early change in Ca²⁺ transient kinetics probably contributes to the development of diastolic dysfunction as the result of a slowing in relaxation which occurs despite the overall positive inotropic effect. Nearly identical time-dependent changes in Ca²⁺ release also occur. Not only is there a slowing in rise time and release rate (+dF/dt) between 6 and 9 mo, but the mixture of normal and defective cells peaks midway between these two time points and then declines with age. Presumably, the slowing in Ca²⁺ release rate could be the result of an early disruption in T-tubule organization, which would increase the number of orphaned RyRs (45), as has recently been reported in another model of hypertension-induced HF in rat heart (51).

We interpret these findings to mean that the defects in Ca²⁺ cycling present in HF actually are present at a much earlier stage of the disease process than would be predicted from the development of symptoms of HF. Even during the positive inotropic phase, the slowing in both Ca²⁺ release and reuptake begins to develop, possibly in response to a hypertensive stimulus and as part of the compensatory hypertrophy. Throughout this compensated phase, defects in E-C coupling continue to develop in individual myocytes so that there is a shift in the number of cells with normal to abnormal Ca²⁺ cycling, thus explaining the biphasic HI for both the rising and declining phases of the Ca²⁺ transient. At the end of this process (≥9 mo), more myocytes demonstrate defective E-C coupling, thus initiating a long decline from mild and asymptomatic dysfunction to abnormal pump function and symptomatic HF, consistent with the known literature (8).

The question then remains about why the heart is performing relatively normally while these myocytes are showing clear signs of altered Ca²⁺ cycling. One possible answer comes from our additional observations about spontaneous Ca²⁺ waves during the development of HF. Waves occur at a high incidence during HF, suggesting that additional defects in E-C coupling have developed during the transition from abnormal E-C coupling with normal function to abnormal E-C coupling and HF. Spontaneous Ca²⁺ waves occur when SR Ca²⁺ content exceeds the threshold for release, causing Ca²⁺ to be released spontaneously into the cytoplasm. Moreover, we found that triggered waves occur in myocytes whose normal E-C coupling mechanisms are overtly impaired, so that LCC activation is unable to activate normal Ca²⁺ transients even though SR Ca²⁺ load is high [current study and Wasserstrom et al. (49)]. One consequence of Ca²⁺ waves is an increase in resting cytoplasmic Ca²⁺ levels as a result of the slower release/reuptake cycling that both contributes to and is a product of the overall Ca²⁺ overload driving the spontaneous waves. This Ca²⁺ overload has also been shown to be critically important in causing apoptosis in these cells (13). Thus it is likely that the decline in cardiac function beginning after 9 mo in SHRs is the result of myocyte loss and replacement with fibrous tissue that occurs in HF. Not only is there a steady decline in myocytes with relatively normal E-C coupling throughout this period (reducing the efficiency of myocyte contraction), but there is also a decline in myocytes with even marginally functioning Ca²⁺ cycling. As cells develop defects that lead to increased spontaneous Ca²⁺ release, they are removed from providing any contribution to mechanical performance. HF development may also be influenced by the fact that these cells are likely to be replaced by fibrosis, which increases myocardial stiffness in addition to the loss of viable myocyte population. This possibility is supported by results from a completely different rat model of severe ventricular dysfunction as a result of aging where end-diastolic pressure was increased, first derivative of pressure was decreased, and fibrosis increased from 8 to 16% (3). Moreover, it has been suggested by others that collagen accumulation with age reduces myocardial compliance and may thus be responsible for both diastolic and systolic HF (4, 40, 50). Taken together, these data suggest that the development of HF occurs as a result of a slow loss of abnormally functioning myocytes that fall victim to defective Ca²⁺ cycling and Ca²⁺ overload.

Increased susceptibility to three forms of Ca²⁺ transient alternans: rate-dependent changes in Ca²⁺ cycling during the transition to HF.

T-wave alternans (TWA) is an important prognostic marker of arrhythmogenesis in HF (42) and is closely related to an increased susceptibility to Ca²⁺ transient alternans in HF (9, 16, 37, 49, 52). We found a progressive increase in susceptibility to several forms of cellular Ca²⁺ alternans, including intercellular concordant alternans, intercellular discordant alternans, and subcellular alternans. An increased vulnerability to intercellular concordant alternans, which is most closely associated with TWA, was greater in SHRs than in WKY animals at 9 and 22 mo. Ca²⁺ transient duration was also prolonged beginning at 9 mo as was the rate of restitution of Ca²⁺ release, both of which might contribute to an increased alternans susceptibility (1, 48) because less time is allowed for recovery of SR release between heart beats (48). This is an important finding since the current thinking is that intercellular concordant Ca²⁺ alternans may be responsible for the development of repolarization gradients, thus establishing the substrate for reentry. A greater susceptibility to intercellular concordant alternans at lower heart rates could explain why HF patients, whose resting heart rate is higher than normals, have a high incidence of TWA and are more vulnerable to arrhythmias. Furthermore, our data show a greater vulnerability to Ca²⁺ alternans when Ca²⁺ transient duration is prolonged (9 mo) and restitution is slowed, suggesting that these characteristics of HF might contribute to the development of Ca²⁺ alternans, pulses alternans, TWA, and reentry long before overt HF is detected (41).

We also found increased susceptibility to the highly heterogeneous state of Ca²⁺ cycling where intercellular discordant alternans ensues, also called dyssynchronous Ca²⁺ transient alternans (1), in SHRs as early as 7.5 mo. The fact that there is greater susceptibility to intercellular discordant alternans during the early peak of HI for Ca²⁺ transient duration is expected. Although the physiological significance of intercellular discordant alternans is not yet clear, it has been implicated as a precursor to lethal arrhythmias, especially in the setting of HF (6). This is not unexpected if intercellular discordant alternans emerged in a regionally heterogeneous manner, as the highly heterogeneous Ca²⁺ cycling in intercellular discordant alternans would (via Ca²⁺→voltage coupling) promote electrical heterogeneities (43). If, however, intercellular discordant alternans emerged in a more regionally homogeneous manner, the intercellular phase-mismatched Ca²⁺ alternans may be effectively averaged out across the myocardium, similar to the recordings of mean fluorescence in the 9- and 22-mo SHR recordings in Fig. 4. This may actually reduce the likelihood of reentrant arrhythmias and serve to maintain a more constant beat-to-beat cardiac output even at high rates, thus diminishing the impact of Ca²⁺ alternans on cardiac output in HF.

A third form of Ca²⁺ transient alternans, subcellular alternans (28), was also more prevalent in SHRs starting at 7.5 mo. It is not yet known how subcellular alternans affects cardiac function because of the small scale of any effects on mechanical and electrical activity. However, the fact that subcellular Ca²⁺ transient alternans is also increased at 7.5 mo suggests that the heterogeneity in Ca²⁺ cycling occurs not only on a cellular level but may also occur at the intracellular level, which together with rapid pacing-induced instabilities in Ca²⁺ cycling dynamics is sufficient to induce subcellular Ca²⁺ alternans (2). We have demonstrated a possible means by which subcellular alternans can lead to intercellular discordant alternans (2) and are currently examining the subcellular heterogeneities in Ca²⁺ cycling during HF development to further investigate the role of subcellular Ca²⁺ alternans in arrhythmogenesis susceptibility in normal versus diseased heart.

Ca²⁺ waves, reduced cardiac output, and arrhythmias.

The high incidence of Ca²⁺ waves, both spontaneous and triggered, is likely to have profound effects on both mechanical and electrophysiological function long before HF has developed. The presence of spontaneous Ca²⁺ waves is associated with spontaneous beats and triggered arrhythmias, so it is interesting that there is a higher incidence of both types of waves at all ages, suggesting the potential for triggered arrhythmias at all stages of disease progression but particularly in advanced HF. Ca²⁺ waves are thought to arise only under conditions of SR Ca²⁺ overload, although it is possible that the threshold for spontaneous release may be altered in HF because of an increased phosphorylation state (32). The same is probably true of triggered waves, but we found that these cells show extremely poor Ca²⁺ release during cardiac activation despite high SR load. Their failure to release during normal pacing leads to progressive Ca²⁺ overload in the form of a wave that is activated in response to normal E-C coupling mechanisms but which cannot induce uniform Ca²⁺ transients along the entire cell length (49). Thus the efficiency of E-C coupling is impaired in these cells, making them unable to contribute to mechanical function. This is probably not a major issue early when most cells can function fairly normally (through 9 mo) but may contribute later to a reduced mechanical performance and to an acceleration of HF as the proportion of defective myocytes increases.

Changes in E-C coupling with aging.

There are many studies that report changes in E-C coupling during aging which have generally shown a decrease in Ca²⁺ transient magnitude (and accompanying negative inotropic effect) in rodents but a positive inotropic effect in sheep (15, 23, 25, 58). Human studies show that cardiac contractile function is relatively well preserved at rest regardless of age, although the ability to increase contractile force in response to increased demand is compromised, and myocardial relaxation is slowed in older adults (29, 30). Similar results have been seen in rodents and related specifically to intracellular Ca²⁺ dynamics (23) where the decline in cardiac contractile function with aging occurs concurrently with alterations in Ca²⁺ handling (18, 25, 31, 54). In young myocytes, contractions and Ca²⁺ transients rise more rapidly at higher stimulation frequencies. However, aged myocytes produce much smaller increases in peak Ca²⁺ transients than younger cells when myocytes are paced at rapid rates. In addition, the rates of decay are prolonged in aged cells compared with younger cells under these experimental conditions. A reduced expression of SERCA in aged myocytes may be responsible for the observed effects (33) although this issue is controversial (55). In addition, a recent study showed that Na⁺/Ca²⁺ exchange activity actually increases with age in intact ventricular myocytes (36) which could compensate, at least in part, for the age-related decline in SERCA activity and help remove Ca²⁺ from the aging cardiac myocyte during relaxation. Proteins involved in SR Ca²⁺ release have also been shown to change with age, including a reduction in RyR2 expression in the aging heart (38). Aging also alters RyR2 function, with data suggesting that the frequency of spontaneous Ca²⁺ sparks increases with age in mouse ventricular myocytes, although the duration of individual Ca²⁺ sparks declines (24). These studies demonstrate both molecular and physiological changes in Ca²⁺ handling in normal ventricular myocytes with age, an observation that is strongly supported by our results in WKY, which provide more precision about when some of these changes occur during senescence. Interestingly, we found that Ca²⁺ transient amplitude and duration are unchanged until WKY rats are nearly 2 yr old. The prolongation of the Ca²⁺ transient occurs consequent to a slowed reuptake rate in these older rats just as described above in other normal rat models and which is consistent with reports that SERCA is less active (21). These older rats also show slowed Ca²⁺ transient restitution, possibly related to the slower SR reuptake, which is also likely to be responsible for an increased vulnerability to intercellular concordant alternans found in the present study and in studies of aging animals (19). These effects could explain why there is an increase in TWA and reentrant arrhythmias in elderly patients in the absence of fibrosis (52).

GRANTS

This work was supported in part by National Heart, Lung, and Blood Institute Grants HL-68733 and HL-071865 (to C. W. Balke) and HL-101196 (to R. Arora).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).