Intraventricular and interventricular cellular heterogeneity of inotropic responses to α₁-adrenergic stimulation

Abstract

α₁-Adrenergic receptors (α₁-ARs) elicit a negative inotropic effect (NIE) in the mouse right ventricular (RV) myocardium but a positive inotropic effect (PIE) in the left ventricular (LV) myocardium. Effects on myofilament Ca²⁺ sensitivity play a role, but effects on Ca²⁺ handling could also contribute. We monitored the effects of α₁-AR stimulation on contraction and Ca²⁺ transients using single myocytes isolated from the RV or LV. Interestingly, for both the RV and LV, we found heterogeneous myocyte inotropic responses. α₁-ARs mediated either a PIE or NIE, although RV myocytes had a greater proportion of cells manifesting a NIE (68%) compared with LV myocytes (36%). Stimulation of a single α₁-AR subtype (α_1A-ARs) with a subtype-selective agonist also elicited heterogeneous inotropic responses, suggesting that the heterogeneity arose from events downstream of the α_1A-AR subtype. For RV and LV myocytes, an α₁-AR-mediated PIE was associated with an increased Ca²⁺ transient and a NIE was associated with a decreased Ca²⁺ transient, suggesting a key role for Ca²⁺ handling. For RV and LV myocytes, α₁-AR-mediated decreases in the Ca²⁺ transient were associated with increased Ca²⁺ export from the cell and decreased Ca²⁺ content of the sarcoplasmic reticulum. In contrast, for myocytes with α₁-AR-induced increased Ca²⁺ transients, sarcoplasmic reticulum Ca²⁺ content was not increased, suggesting that other mechanisms contributed to the increased Ca²⁺ transients. This study demonstrates the marked heterogeneity of LV and RV cellular inotropic responses to stimulation of α₁-ARs and reveals a new aspect of biological heterogeneity among myocytes in the regulation of contraction.

previously, we found interventricular differences in inotropic responses to stimulation of α₁-adrenergic receptors (α₁-ARs), with an overall positive inotropic effect (PIE) in left ventricular (LV) myocardium in contrast to an overall negative inotropic effect (NIE) in right ventricular (RV) myocardium (27, 28). Alterations in myofilament Ca²⁺ sensitivity played a role in the contrasting α₁-AR inotropic responses of the LV versus RV myocardium (27, 28). However, the role of Ca²⁺ handling in α₁-AR inotropic responses remains unclear. Previously, an interventricular difference in Ca²⁺ handling was reported, with smaller Ca²⁺ transients noted in RV myocytes versus LV myocytes (15). Here, we investigated the role of Ca²⁺ handling in the differing α₁-AR inotropic responses of the RV versus LV. We used adult mouse ventricular myocytes from the RV or LV free wall and monitored Ca²⁺ transients and contractile responses in response to stimulation of α₁-ARs.

Interestingly, we found that for both RV and LV myocytes, there was considerable cellular heterogeneity in the effects of α₁-AR stimulation on inotropic responses. Moreover, compared with the LV, the RV had a significantly greater proportion of cells in which α₁-AR stimulation elicited a NIE (68% vs. 36%). Differences in α₁-AR inotropic responses appeared to be driven by effects on Ca²⁺ transients; moreover, α₁-AR effects on Ca²⁺ loading of the sarcoplasmic reticulum (SR) played a role.

This study demonstrates cellular heterogeneity of α₁-AR inotropic responses, which is a new aspect of biological heterogeneity among myocytes that potentially might extend to other types of receptors.

METHODS

The Animal Studies Subcommittee of the San Francisco Veterans Affairs Medical Center approved all procedures.

Previously, we measured α₁-AR inotropic responses using multicellular myocardial samples (trabeculae) dissected from the LV or RV (27, 28). The present study used single myocytes isolated from the LV or RV and measured α₁-AR inotropic responses and fura-2-assessed Ca²⁺ transients.

Myocyte isolation.

Twelve-week-old male C57BL/6 mice were deeply anesthetized with pentobarbital (100 mg/kg ip) mixed with heparin (100 units). A midline thoracotomy was performed, and the heart was rapidly removed, immersed in ice-cold arrest solution [containing (in mM) 120 NaCl, 30 KCl, and 0.1 CaCl₂], and mounted to a cannula by the aorta. Myocytes were isolated by enzymatic digestion of the heart by retrograde perfusion of collagenase solution through the coronary vasculature (20). After collagenase treatment, the heart was placed in a “stop buffer” to halt enzymatic digestion (20). The free wall of the RV or LV was dissected, placed in 10 ml of stop buffer, and gently teased with fine forceps followed by repeated pipetting to release the cells. Cells from only one ventricle were studied per animal. The cell isolation buffers contained 10 mM 2,3-butanedione monoxime to prevent cell contraction. Cells were used within 4 h of isolation.

The RV yielded ∼350,000 cells. The LV free wall (representing ∼50% of the LV) yielded ∼720,000 cells. These yields suggest a cell yield for the entire heart of ∼1.8 million cells, which is similar to the cell yield obtained in our previous study (17) of mouse hearts. Moreover, the ratio between RV versus LV cell yield is consistent with the ratio of RV versus LV weight for the mouse heart in our previous study (27).

Overall, we studied 163 cells from 64 animals. Few cells (2–3 cells) were studied per heart due to the difficulty of maintaining stable cell contractions for the 20–30 min required for an equilibration period followed by recording the response to an agonist. Technical problems, such as cell movement and arrhythmias, resulted in a success rate ∼50%.

Fura-2 loading.

Myocytes were loaded with fura-2 by exposure to 1 μM fura-2 AM for 20 min. After being washed, myocytes were equilibrated for 20 min to allow for deesterification of the indicator and used within 1.5 h after completion of the loading protocol.

Contractility and fluorescence measures.

Myocytes were superfused in a small glass-floored chamber with Krebs-Henseleit solution containing (in mM) 112 NaCl, 5 KC1, 1.2 MgCl₂, 10 glucose, 24 NaHCO₃, 1.2 Na₂SO₄, 2.0 NaH₂PO₄, and 1 CaCl₂. The perfusate was oxygenated with 95% O₂-5% CO₂ to give a pH of 7.4 at 22°C. The chamber was mounted on an inverted Nikon Diaphot microscope, and cells were visualized at ×40 magnification. Cells were electrically stimulated with 4-ms square-wave pulses at 25 V and at a frequency of 0.5 Hz. Cells were selected based on a rod-shaped appearance with clear striations and contractions (≥2% basal length) in response to electrical stimulation. Cell contraction and fura-2 fluorescence were monitored using an IonOptix Hyperswitch system (Milton, MA). Cell contraction was computed by monitoring changes in muscle sarcomere length measured from the myocyte striation spacing. Fura-2 fluorescence at an emission wavelength of 510 nm was measured, with the excitation alternated between 340 and 380 nm at a frequency of 240 Hz. On each experimental day, for both excitation wavelengths, the background autofluorescence of the instrument plus unloaded cells was assessed (n = 5 cells), and the background autofluorescence at each wavelength was subtracted before the fura-2 fluorescence ratio was computed. Calibration factors were determined to calibrate the fura-2 ratio to units of Ca²⁺ concentration. Briefly, fura-2 loaded cells (n = 10) were metabolically inhibited, permeablilized, and exposed to calibration solutions of various Ca²⁺ concentrations, as previously described (18). Fura-2 ratio was converted to Ca²⁺ concentrations using the formula of Grynkiewicz (12).

Inotropic responses.

Contracting myocytes were equilibrated for 10 min in the presence of the β-AR antagonist timolol (10 μM). Myocytes were then exposed to the α₁-AR agonist phenylephrine (PE) at a dose (10 μM) that mediates a near-maximal inotropic response in the mouse myocardium (13, 19, 21, 26). In a subset of experiments, myocytes were exposed to the α_1A-AR subtype-selective agonist A-61603 (100 nM). Previously, we noted that myocardium from mice lacking the α_1A-AR subtype did not respond to A-61603, indicating that A-61603 did not stimulate other α₁-AR subtypes (18). Contraction was monitored for ∼15 min, and fura-2 transients were monitored intermittently during this time (to limit photobleaching).

Caffeine-induced contractures.

In some experiments, rapid application of caffeine was used to assess SR Ca²⁺ loading. Before or 10 min after exposure to PE, electrical stimulation was stopped, and, 5 s later, cells were rapidly exposed to a stream of superfusate containing 20 mM caffeine passed through a micropipette placed close to the cell (1).

Statistical analysis.

Data are presented as means ± SE. Groups were compared using Students t-test or a χ²-test.

RESULTS

RV cells have smaller Ca²⁺ transients and slower relaxation than LV cells.

Figure 1 shows the contraction and Ca²⁺-handling baseline characteristics of LV versus RV cells. Consistent with a previous study (15), RV cells had smaller Ca²⁺ transients and slower relaxation than LV cells. New findings were that compared with LV cells, the decline phase of the Ca²⁺ transient was slower in RV cells and the diastolic sarcomere length was longer in RV cells.

Fig. 1.

Baseline recording of Ca²⁺ transients and contractions of right ventricular (RV) myocytes (n = 31) and left ventricular (LV) myocytes (n = 28). A: diastolic and systolic Ca²⁺ levels. B: sarcomere length. C: time constant (τ) of Ca²⁺ decline. D: τ of relaxation of contraction. **P < 0.01; ***P < 0.001.

Consistent with a previous report (15), for electrically stimulated cells, RV cells had significantly lower systolic Ca²⁺ levels compared with LV cells (1.11 ± 0.06 μM, n = 30, vs. 1.35 ± 0.1 μM, n = 26, P < 0.05; Fig. 1A).

Figure 1B shows that for contracting cells, the diastolic sarcomere length of RV cells was 24 nm longer than for LV cells (1.823 ± 0.005 μm, n = 31, vs. 1.799 ± 0.007 μm, n = 28, P < 0.01). This was not due to differences in diastolic Ca²⁺ levels, which did not differ between LV and RV cells (Fig. 1A). Furthermore, this difference in sarcomere length was evident in quiescent cells maintained in the cell isolation buffer containing 2,3-butanedione monoxime and without Ca²⁺ (not shown). Thus, effects of contraction did not play a role in the longer diastolic sarcomere length of RV versus LV cells.

The systolic sarcomere length of RV cells was 37 nm longer than LV cells (1.763 ± 0.008 μm, n = 31, vs. 1.726 ± 0.007 μm, n = 28, P < 0.01). Consistent with smaller Ca²⁺ transients in RV cells, there was a trend for the mean contraction amplitude (difference between diastolic and systolic sarcomere lengths) for RV cells to be smaller (80%) compared with LV cells, but the difference did not reach statistical significance (P = 0.09).

Quantitation of the relaxation phase used the time constant estimated for the exponential decline of signals for Ca²⁺ concentration and sarcomere length. Compared with LV cells, RV cells had slower Ca²⁺ transient decline (greater time constant value; Fig. 1C). Furthermore, RV cells had a slower time course of relaxation than LV cells, consistent with a previous report (15). The slower Ca²⁺ transient decline might have contributed to the slower relaxation in RV cells compared with LV cells.

In summary, compared with LV myocytes, RV myocytes have smaller Ca²⁺ transients with slower Ca²⁺ decline and somewhat smaller contractions with slower relaxation.

Cellular heterogeneity of α₁-AR inotropic responses.

Previously, we found that α₁-AR stimulation resulted in a NIE in the RV myocardium but a PIE in the LV myocardium (27, 28). Interestingly, in this study, when examined at the cellular level, considerable heterogeneity of α₁-AR inotropic responses was noted. Figure 2 shows that α₁-AR stimulation manifested either a NIE or PIE in cells from the RV free wall (A and B) or cells from the LV free wall (C and D). For all cells studied, there was no relation between baseline contractility and the direction of the inotropic response. For example, the cells shown in Fig. 2, C and D, had similar baseline contraction amplitudes but directionally opposite inotropic responses.

Fig. 2.

Slow time-based recordings of electrically stimulated contractions before and after stimulation of α₁-adrenergic receptors (α₁-ARs) with 10 μM phenylephrine (PE) plus the β-blocker timolol (10 μM). A: addition of PE (arrow) elicited a negative inotropic effect (NIE) in some RV cells. B: PE elicited a positive inotropic effect (PIE) in other RV cells. C and D: LV cells also manifest a NIE or PIE after α₁-AR stimulation. E: there was in no inotropic response to treatment with the vehicle (Veh) control.

For RV cells, α₁-AR stimulation with PE elicited a NIE in the majority (68%) of cells (21 of 31 cells, P < 0.05; Fig. 3A). This is consistent with the overall NIE of α₁-AR stimulation in the intact RV myocardium. In contrast, for LV cells, α₁-AR stimulation elicited more of a mixed response, with a NIE in a minority (36%) of cells (10 of 28 cells) and a PIE in a majority (64%) of cells (18 of 28 cells), perhaps reflecting the overall PIE elicited by α₁-AR stimulation in the intact LV myocardium.

Fig. 3.

Summary of α₁-AR inotropic responses of RV and LV myocytes (expressed as the percent change from the baseline contraction level for each cell). Heterogeneity of inotropic responses was observed using the nonsubtype-selective α₁-AR agonist PE (A) and also using the α_1A-AR subtype-selective agonist A-61603 (B).

Few cells from either the RV or LV were studied per heart (range: 1–5 cells, average 2.4 cells/heart). Nevertheless, for the majority (66%) of the 21 hearts from which at least 2 cells were studied, α₁-AR stimulation with PE elicited both PIE and NIE types of the cellular response. This indicates that the observed cellular heterogeneity of α₁-AR responses existed within hearts rather than arising from variability between animals.

Cellular heterogeneity of α_1A-AR subtype inotropic responses.

There are two predominant α₁-AR subtypes on cardiac myocytes (α_1A and α_1B), which may have different roles in cardiac inotropy (6, 9, 11, 16, 22). Therefore, in a subgroup of cells, we tested the role of a single α₁-AR subtype (α_1A-ARs) in cellular inotropic responses. For both RV and LV myocytes, we found that the highly selective α_1A-AR agonist A-61603 mediated either a PIE or a NIE (Fig. 3B). Thus, the single α_1A-AR subtype could mediate fundamentally different inotropic responses, suggesting that the type of inotropic response (PIE or NIE) for a particular cell depends on events downstream from the receptor. Similar to PE, A-61606 elicited a PIE in the majority (82%) of LV cells (14 of 17 cells).

Cellular heterogeneity of α₁-AR effects on Ca²⁺ handling.

Contraction and relaxation are triggered by the rise and fall of activator Ca²⁺. Therefore, we determined the role of activator Ca²⁺ in the heterogeneous α₁-AR inotropic responses of LV and RV myocytes.

Figure 4 shows that fura-2-induced Ca²⁺ transients and sarcomere length transients measured in LV and RV myocytes before and after α₁-AR stimulation with PE elicited either a NIE (A and C) or a PIE (B and D). Evidently, the α₁-AR-mediated NIE was associated with a marked decrease of the Ca²⁺ transient, whereas the PIE was associated with an increase of the Ca²⁺ transient.

Fig. 4.

Examples of a NIE and PIE elicited by α₁-ARs in LV and RV myocytes. Records of fura-2-induced Ca²⁺ transients and cell contractions (assessed from sarcomere length measures) are shown. Compared with before the addition of PE, α₁-AR stimulation caused decreased Ca²⁺ transients and decreased contractions in some cells (A and C) or increased Ca²⁺ transients and contraction in other cells (B and D).

Although changes in Ca²⁺ and sarcomere length are not in a dynamic equilibrium during myocyte contraction, nevertheless there is a positive correlation between the size of the Ca²⁺ transient and the corresponding magnitude of contraction (10). The pooled data shown in Fig. 5 demonstrate the relationship between cell contraction amplitude and the corresponding Ca²⁺ transient amplitude. For LV myocytes, there was a linear relationship between contraction amplitude and Ca²⁺ transient amplitude for cells before and after α₁-AR stimulation with PE. This relationship suggests that the NIE or PIE of α₁-AR stimulation was appreciably driven by corresponding decreases or increases of Ca²⁺ transient amplitude. Figure 5 shows that RV myocytes also manifested a relationship between cell contraction and Ca²⁺ transient amplitude. This relationship appeared steeper for RV myocytes than for LV myocytes, suggesting that for a PIE elicited by PE, the myofilament Ca²⁺ responsiveness of RV myocytes was greater than for LV myocytes.

Fig. 5.

Relationship between the amplitude of myocyte contraction and Ca²⁺ transient amplitude for RV cells (open symbols) and LV cells (solid symbols). Data are shown for cells before the addition of PE (●) and after 10 min in the presence of PE for cells that manifested a PIE (▲) or a NIE (▼).

Numerous factors contribute to the amplitude of the Ca²⁺ transient. We investigated if α₁-AR effects on Ca²⁺ transients were linked to changes in Ca²⁺ loading of the SR. SR Ca²⁺ loading was assessed by a rapid application of 20 mM caffeine, which causes rapid unloading of SR Ca²⁺. Figure 6A shows that caffeine caused a rapid rise of intracellular Ca²⁺ concentration followed by a slow decline of Ca²⁺ concentration. Figure 6A shows that for cells manifesting an α₁-AR-induced NIE, the peak of the caffeine-induced Ca²⁺ transient was reduced compared with cells without α₁-AR stimulation by PE. In contrast, cells manifesting an α₁-AR-induced PIE did not have an increase in the peak of the caffeine-induced Ca²⁺ transient compared with cells without α₁-AR stimulation.

Fig. 6.

Effects on α₁-AR stimulation on Ca²⁺ handling as assessed with caffeine-induced contractures. A: records of caffeine-induced Ca²⁺ transients in LV cells before the addition of PE and 10 min after PE induced either a NIE or a PIE. B: summary of caffeine-induced Ca²⁺ transients before (Con) and after α₁-AR stimulation induced a NIE or PIE in RV or LV cells. C: summary of the time course of decline of the caffeine-induced Ca²⁺ transient as assessed from the relaxation half-time (RT₅₀; same groups as in B). RV and LV myocyte numbers were 17 and 14 for Con, 8 and 8 for NIE, and 9 and 7 for PIE, respectively. *P < 0.05; ***P < 0.001.

The pooled data (Fig. 6B) demonstrate that for both RV and LV cells, an α₁-AR-induced NIE was associated with a reduced caffeine-induced Ca²⁺ transient. In contrast, an α₁-AR-induced PIE was not associated with a change in the caffeine-induced Ca²⁺ transient compared with cells without α₁-AR stimulation. This suggests that for LV and RV cells, the α₁-AR-induced NIE may have been contributed to by a reduction in SR Ca²⁺ loading. In contrast, for LV and RV cells, the α₁-AR-induced PIE did not involve increased SR Ca²⁺ loading.

The decline of Ca²⁺ concentration in the presence of caffeine reflects Ca²⁺ transport from the cytosol by other Ca²⁺-handling systems besides sarco(endo)plasmice reticulum Ca²⁺-ATPase, principally, Ca²⁺ export from the cell by the Na⁺/Ca²⁺ exchange (1). Figure 6C shows that for both LV and RV cells, α₁-AR stimulation tended to increase the rate of Ca²⁺ decline by these other Ca²⁺ transport systems. After α₁-AR stimulation, faster Ca²⁺ export from the cell via Na⁺/Ca²⁺ exchange could contribute to decreased SR Ca²⁺ load and a NIE. In contrast, faster Ca²⁺ export from the cell would not contribute to an α₁-AR-induced PIE.

DISCUSSION

The major finding of this study was that the free walls of both the RV and LV are composed of myocytes that manifest considerable cell to cell heterogeneity in their Ca²⁺ handling and contractile responses to stimulation of α₁-ARs. This finding reveals a new aspect of biological heterogeneity among cardiac myocytes that might encompass other regulatory systems. Moreover, this study illuminates a complexity to the effects of α₁-ARs on Ca²⁺ signaling and contraction that was not suspected from previous studies of the intact myocardium.

Cellular heterogeneity in the heart.

The heart is known to manifest regional heterogeneity, with reports of transmural differences between the endocardium and epicardium in voltage-gated K⁺ currents (2), action potential duration (14), excitation-contraction coupling (7), myosin light chain kinase abundance (5), and myofilament properties (3, 4, 25).

We have previously reported that α₁-ARs mediated fundamentally different inotropic responses in the RV versus LV myocardium (27, 28). This difference was associated with different effects of α₁-AR stimulation on myofilament Ca²⁺ sensitivity of the LV versus RV myocardium (28). However, α₁-ARs also influence Ca²⁺ handling. Moreover, differences in Ca²⁺ handling between the RV and LV have been reported (15). Therefore, we investigated the effects of α₁-AR stimulation on Ca²⁺ handling in RV and LV myocytes.

Cellular heterogeneity of Ca²⁺ signaling and inotropic responses to α₁-adrenergic stimulation.

At the cellular level, the effects of α₁-AR stimulation were complex. We found considerable cell-to-cell variability in α₁-AR effects on Ca²⁺ signaling and contraction. For cells from the RV or LV free wall, α₁-ARs elicited a NIE in some cells but a PIE in other cells. The α₁-AR-induced NIE was associated with a decreased Ca²⁺ transient, and the α₁-AR-induced PIE was associated with an increased Ca²⁺ transient. These findings suggest that α₁-AR-induced changes in the Ca²⁺ transient played a key role in the inotropic response.

We found that cells with an α₁-AR-induced NIE and reduced Ca²⁺ transient manifested a reduction in SR Ca²⁺ loading (as assessed from the caffeine-induced Ca²⁺ transient). Reduced SR Ca²⁺ load induced by α₁-ARs might contribute to the observed reduction of Ca²⁺ transients and the NIE. Moreover, with transport of Ca²⁺ to the SR prevented in the presence of caffeine, the decay of the caffeine-induced Ca²⁺ transient is thought to reflect other Ca²⁺ transport systems, predominantly the Na⁺/Ca²⁺ exchanger, which exports Ca²⁺ out of the cell (1). We found that α₁-AR stimulation accelerated the decay of the caffeine-induced Ca²⁺ transient. This is consistent with α₁-AR-induced stimulation of Na⁺/Ca²⁺ exchange (19). Faster export of Ca²⁺ from the cell could contribute to a decreased SR Ca²⁺ load and the NIE elicited by α₁-AR stimulation.

In contrast, cells with an α₁-AR-induced PIE and increased Ca²⁺ transients did not manifest an increased SR Ca²⁺ load, suggesting that other mechanisms contributed to the increased Ca²⁺ transient [e.g., prolongation of the action potential (8), increased Ca²⁺ current, and increased gain of Ca²⁺-induced Ca²⁺ release from the SR].

Cellular heterogeneity of α₁-AR inotropic responses in the RV versus LV.

The inotropic response to α₁-AR stimulation has remained uncertain because it has varied considerably among studies and may be influenced by experimental conditions (23, 28). The results of the present study demonstrate that cellular heterogeneity within both the RV and LV is another complexity to α₁-AR inotropic responses. Moreover, the fraction of cells manifesting an α₁-AR-mediated NIE was significantly greater in the RV (68%) than in the LV (36%). This trend appears consistent with a previous report (28) of α₁-ARs mediating a NIE in the RV myocardium. Thus, both intraventricular and interventricular factors may play a role in α₁-AR inotropy.

Previous studies (6, 9, 11, 16, 22) have suggested that α₁-AR subtypes may have different roles in cardiac inotropy, with the α_1A-AR subtype mediating positive inotropy and the α_1B-AR playing a negative modulatory role. However, in myocytes, we found that the α_1A-AR subtype could mediate either a positive or negative inotropic response. This finding is consistent with our previous report (28), which showed that the α_1A-AR agonist A-61603 mediated negative inotropy in the RV myocardium but positive inotropy in the LV myocardium, again indicating that the α_1A-AR subtype can mediate a PIE or a NIE. Together, these findings suggest that events downstream from the α_1A-AR subtype determine whether a PIE or NIE is elicited. The present study identifies that differential effects of α₁-ARs on Ca²⁺ handling play a critical role.

Limitations.

The clinical disease relevance of the presence of heterogeneous α₁-AR responses is unclear. Potentially, the balance of cells manifesting a PIE versus a NIE elicited by α₁-ARs may shift in disease and thereby impact the myocardial α₁-AR inotropic response. For example, we (27) have previously reported a shift in the myocardial α₁-AR inotropic response from a NIE in the nonfailing RV to a PIE in the failing RV.

This study indicates that effects of α₁-ARs on Ca²⁺ handling contribute to the heterogeneous inotropic responses of myocytes. Further studies will be required to identify the specific mechanisms downstream from α₁-ARs that determine the inotropic response.

Like most studies of myocytes, only a small number of myocytes per heart were studied. However, the myocytes that survive the isolation process may not be truly representative of the total myocyte populations of the RV or LV. Cells were studied under artificial in vitro conditions, and thus it is not clear how well the in vitro cell behavior reflects that of cells in vivo. We studied mouse cells, which are similar to humans in cardiac α₁-AR abundance (24). However, it will be important to study cells from other species. As we studied only a single class of receptors, it will be interesting to determine if there is cellular heterogeneity in responses to other receptor classes.

Conclusions.

This study demonstrates the intraventricular and interventricular heterogeneity of cellular inotropic responses to stimulation of α₁-ARs. A single α₁-AR subtype (α_1A) can manifest either a PIE or a NIE. Multiple mechanisms, including effects on Ca²⁺ transients, SR Ca²⁺ loading, and export of Ca²⁺ from the cell, are involved. This study reveals a new aspect of biological heterogeneity among cardiac myocytes in the regulation of contraction.

GRANTS

This work was supported by Veterans Affairs Merit Awards (to A. J. Baker and to D. H. Lovett), an American Heart Association-Western States grant-in-aid (to A. J. Baker), and National Heart, Lung, and Blood Institute Grant HL-31113 (to P. C. Simpson).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

Author contributions: C.C., K.V.T., K.W.P., P.W., O.M., and D.H.L. performed experiments; C.C., K.V.T., K.W.P., P.W., and A.J.B. analyzed data; D.H.L., P.C.S., and A.J.B. edited and revised manuscript; D.H.L., P.C.S., and A.J.B. approved final version of manuscript; A.J.B. conception and design of research; A.J.B. interpreted results of experiments; A.J.B. prepared figures; A.J.B. drafted manuscript.