Targeted Disruption of K_ATP Channels Aggravates Cardiac Toxicity in Cocaine Abuse

Abstract

Cocaine is the most frequently used illicit drug among individuals seeking emergency-room care, with fatal outcome most often attributable to the cardiovascular manifestations of drug abuse. While the symptomatic presentations of cocaine toxicity are increasingly understood, the molecular determinants that define outcome remain largely unknown. Here, we report that the susceptibility to cocaine-induced cardiotoxicity is genetically regulated. Targeted deletion of the KCNJ11-encoded Kir6.2 pore-forming subunit of sarcolemmal K_ATP channels resulted in amplified vulnerability to the toxic effects of chronic cocaine abuse. Under the hyperadrenergic stress, imposed by daily 3-week-long intraperitoneal administration of 30 mg/kg cocaine in Kir6.2-knockout mice, failure to maintain cardiac homeostasis translated into decreased exercise tolerance revealed by poor treadmill stress performance, and dilated hypokinetic left hearts with aggravated cellular hypertrophy and pathognomonic characteristics of chronic cocaine-induced cardiac toxicity. This study therefore reveals a previously unrecognized role of Kir6.2-encoded K_ATP channels in determining cardiovascular outcome in chronic cocaine abuse, identifying a novel molecular determinant of cocaine cardiotoxicity.

Introduction

Cocaine is a powerful stimulant that accounts for over 2 million of illicit-drug abusers and recurring dependents in the United States.¹ The alkaloid's addictive properties are associated with near-pandemic abuse. Intoxication with cocaine is the most frequent cause of drug-related visits to the emergency department worldwide, mainly related to adverse cardiovascular manifestations of the drug. By inhibiting reuptake of catecholamines, cocaine induces an overdriven sympathomimetic effect resulting in mismatch of energy demand and supply responsible for drug-induced toxicity.² Although introduced in medical practice as a local anesthetic, cocaine can produce irreversible structural damage to the heart, accelerate cardiovascular disease progression, and precipitate sudden cardiac death.³ Indeed, cocaine cardiopathology is the most common substrate leading to hospitalization and mortality among chronic drug abusers.⁴ Of note, individual variations in the severity of presentation have been documented in cases of cardiac cocaine toxicity.³ However, the underlying molecular determinants of outcome in cocaine abuse remain largely unknown.

Recently, KCNJ11 has been identified as a safeguard molecule conferring survival advantage in response to the hyperadrenergic state imposed by acute cocaine overdose.⁵KCNJ11 encodes the pore-forming Kir6.2 subunit of ATP-sensitive potassium (K_ATP) channels, recognized biosensors that adjust membrane potential-dependent processes to match metabolic demand.^6–8 In the heart, Kir6.2 is integral to the make-up of sarcolemmal K_ATP channels.^9–11 Integrated with cellular metabolic pathways,^12–16 cardiac K_ATP channels process energetic signals of distress.^6,17,18 Kir6.2 is required in cardiac adaptation to physiological and pathophysiological stress,^19–22 with K_ATP channel malfunction implicated in heart disease.^23,24 A deficit in K_ATP channels impairs tolerance to systemic stressors, including sympathetic surge.^25,26 Genetic disruption of K_ATP channels compromises cardioprotection,²⁷ while variants in channel proteins have been linked to cardiac pathology in patient populations.^28–30 The increased understanding of the K_ATP channel involvement in regulation of heart homeostasis^8,31,32 raises the hypothesis that these unique channels contribute toward protection of the myocardium from the hyperadrenergic state of cocaine cardiotoxicity.

Here, we tested the relevance of K_ATP channels in securing cardiac adaptation to repetitive exposures to cocaine. A chronic, sublethal cocaine administration regimen resulted in reduced exercise capacity accompanied by pathological left ventricular enlargement and impaired blood ejection velocity, as well as increased myocyte cross-sectional area in the Kir6.2-knockout compared to wild-type counterparts. Th is study identifies K_ATP channel functionality as a novel determinant of outcome in cocaine abuse.

Methods

Murine model of chronic cocaine abuse

Mice deficient in K_ATP channels (Kir6.2-KO) were generated by targeted disruption of the KCNJ11 gene encoding the pore-forming Kir6.2 subunit of sarcolemmal K_ATP channels, and backcrossed for five generations into a C57BL/6 background.^33,34 In accordance with NIH guidelines, and following approval by the Mayo Clinic Institutional Animal Care and Use Committee, 12-week-old Kir6.2-KO and sex/age-matched C57BL/6 wild-type (WT) mice were administered a sublethal dose of cocaine (30 mg/kg i.p. every 12 hours) for 20 days (Figure 1). All mice were given standard rodent chow, were housed individually with a 12-hour day/night cycle, and were monitored daily. Body weight and food consumption were measured prior to initiation of cocaine administration (drug-free state), and following the 20 day-long cocaine administration regimen.

Figure 1. Experimental design.

At a drug-free state, prior to initiation of cocaine administration, a treadmill stress test and baseline food consumption measurements were performed. Day 0 represents the beginning of the twice daily, 30 mg/kg intraperitoneal cocaine administration that lasted 20 days. On completion of this regimen, a second exercise test, food consumption measurements, and echocardiography were performed, followed by postmortem histopathological evaluation of heart tissue samples.

Treadmill exercise stress test

To assess the impact of chronic cocaine abuse on physical capacity in wild-type and Kir6.2-KO mice, a comparison of performance was made based on exercise stress tests at baseline and at day 21 (Figure 1). Stepwise increments of incline and velocity at 3-minute intervals on a graded treadmill (Columbus Instruments, Columbus, OH, USA) allowed the determination of total tolerated workload (W, in Joules), a composite parameter calculated as the sum of kinetic energy E_k = m · v²/2 and potential energy E_p = m · g · v · t sin(θ), where m represents animal mass, v treadmill velocity, g acceleration due to gravity, t elapsed time at a protocol level, and θ the angle of incline.^20,26

Cardiac structure and function

Trans-thoracic cardiac ultrasound imaging (15L8 transducer, Sequoia c256, Acuson, Mountain View, CA, USA) was performed in lightly sedated (1.25% isoflurane) mice before and after the chronic cocaine administration protocol (Figure 1). Images were digitally acquired and stored for offline blinded analysis. Echocardiographic measurements of left ventricular dimensions were recorded at end-diastole (LVDd) and end-systole (LVDs) from three consecutive cardiac cycles, using the leading edge method. Similarly, interventricular septum (IVST) and posterior wall thickness (PWT) were measured from M-mode images. The ratio of relative wall thickness to the left ventricular end-diastolic dimension, a parameter of heart geometry, was expressed as [(IVST + PET)/LVDd]. Left ventricular fractional shortening was calculated as [(LVDd − LVSd)/LVDd]×100.^20,35 Peak ejection velocity was determined from the actual pulsed-wave Doppler tracings on the parasternal long-axis view of trans-aortic flow by measuring the absolute magnitude of three consecutive tracings.^36,37

Histochemistry

At 21 days, one day after the last cocaine administration, left ventricles (LV) including septum were removed, rinsed, blotted dry, and stored in 10% buffered formalin. Micrographs of 5 μm thick, paraffin-embedded, LV sections stained with hematoxylin-eosin were taken with a KS-4000 Axiocam digital camera (Carl Zeiss MicroImaging, Thornwood, NY, USA) mounted on an Axioplan 2 light microscope (Carl Zeiss MicroImaging, Thornwood, NY, USA) and analyzed using the MetaMorph software (Visitron Universal Imaging, Union City, CA, USA).^38,39 All quantification was performed blinded to the sample origin.

Statistical analysis

Results are expressed as mean ± SEM. Student's t-test was used to compare variables between experimental groups and define p-values (JMP 7, SAS, Cary, NC). P-values less than 0.05 were predetermined to indicate statistical significance.

Results

Chronic cocaine administration results in exercise intolerance in K_ATP channel knockout

A potent sympathoadrenal stimulant, exercise amplifies the catecholamine response induced by cocaine administration increasing stress load, particularly to the heart.⁴⁰ Here, under a treadmill stress-test, chronic cocaine administration rendered Kir6.2-KO mice vulnerable to physical exertion as demonstrated by reduced tolerated workload and exercise duration. These performance parameters were 25% and 11% lower, respectively (p < 0.05), than those observed at a drug-free state, an effect not seen in the wild type (WT) in which cocaine consumption enhanced tolerated workload (+27%; p < 0.05) when compared to the drug-free baseline (Figure 2A). The decreased exercise capacity of cocaine-abusing Kir6.2-KO animals was not a consequence of reduced caloric intake, as revealed by equivalent or even slightly increased food consumption compared to WT (198.7 ± 9.4 mg/day/g vs. 224.5 ± 9.0 mg/day/g before, and 204.4 ± 12.4 mg/day/g vs. 210.1 ± 6.8 mg/day/g after cocaine administration in WT vs. Kir6.2-KO, respectively; Figure 2B), nor of an accentuated body weight loss resulting from cocaine administration (30 ± 0.8 g in WT vs. 28.1 ± 0.8 g in Kir6.2-KO before, and 27.1 ± 0.9 g in WT vs. 25.5 ± 0.9 g in Kir6.2-KO after cocaine administration; Figure 2C).

Figure 2. Impaired physical capacity with cocaine challenge in the absence of K_ATP channels.

(A) Kir6.2 knock out compromises physical endurance and impairs cardiac performance in chronic cocaine abuse. Under exercise stress testing, Kir6.2-KO exhibited impaired endurance compared to matched cocaine-challenged WT with decreased workload tolerance (25% less than at drug-free state for Kir6.2-KO vs. 27% above baseline for WT) and exercise duration (11% decrease from baseline for Kir6.2-KO vs. no change for WT) after 20 days of cocaine administration. (B) Food consumption normalized to body weight was constant before (198.7 ± 9.4 mg/day/g vs. 224.5 ± 9.0 mg/day/g) and after (204.4 ± 12.4 mg/day/g vs. 210.1 ± 6.8 mg/day/g) cocaine administration in WT vs. Kir6.2-KO. (C) Both WT and Kir6.2-KO animals experienced a reduction in body weight. However, body weight was indistinguishable between groups before (30 ± 0.8 g in WT vs. 28.1 ± 0.8 g in Kir6.2-KO) and after (27.1 ± 0.9 g in WT vs. 25.5 ± 0.9 g in Kir6.2-KO) the cocaine administration regimen. In each group, n = 5; ^†p < 0.05 when comparing each group to baseline drug-free state. *p < 0.05 when comparing WT versus Kir6.2-KO.

Ventricular dilatation and impaired contractility in K_ATP channel knockout after cocaine abuse

Cocaine abuse can lead to clinically relevant left ventricular dysfunction.^41,42 Here, after 20 days of imposed cocaine consumption, left ventricular diastolic (LVDd) and systolic (LVSd) dimensions reached 3.3 ± 0.2 mm and 2.4 ± 0.2 mm in Kir6.2-KO, respectively, significantly exceeding measurements obtained in WT (2.1 ± 0.2 mm LVDd, p < 0.01, Figure 3A; 1.3 ± 0.2 mm LVSd, p < 0.05, Figure 3B). Exaggerated left ventricular chamber dilatation was accompanied by prominent wall thinning, as the ratio between relative left ventricular wall thickness and LVDd ((IVST + PWT)/LVDd, a parameter reflecting heart geometry, was lower at 0.5 ± 0.1 in Kir6.2-KO versus 1 ± 0.1 in WT (p < 0.01; Figure 3C), as revealed by M-mode echocardiography (Figure 3D).

Figure 3. Dilated hypokinetic hearts in cocaine-challenged K_ATP channel deficient hearts.

Examination by ultrasound revealed dilated hearts in Kir6.2-KO compared to WT after chronic cocaine administration. (A) Left ventricular end-diastolic dimension (LVDd; 3.3 ± 0.2 mm vs. 2.1 ± 0.2 mm; n = 5) and (B) left ventricular end-systolic dimension (LVSd; 2.4 ± 0.2 mm vs. 1.3 ± 0.2 mm; n = 4) were augmented in Kir6.2-KO compared to WT after 20 days of cocaine. (C) The ratio between relative ventricular wall thickness (the sum of interventricular septum thickness, IVST, and posterior wall thickness, PWT) and LVDd was lower in Kir6.2-KO (0.5 ± 0.1; n = 5) versus WT (1 ± 0.1; n = 4) after the cocaine regimen, as shown in representative echocardiographic M-mode recordings (D). (E) Deteriorated cardiac performance following chronic cocaine injections was observed as a significantly reduced fractional shortening from the drug-free baseline in Kir6.2-KO (48.2 ± 4.2% pre- vs. 28.3 ± 4.2% post-cocaine; n = 5 in each group) in contrast to sustained function in WT (47.5 ± 4.2% pre- vs. 35.4 ± 4.7%; n = 4 in each group). (F) Peak ejection velocity appeared lower in Kir6.2-KO (0.49 ± 0.05 m/s; n = 5) versus WT (0.67 ± 0.03 m/s; n = 5), as documented by representative Doppler ultrasound (C). ^†denotes p < 0.05 when comparing each group to baseline drug-free state. *denotes p < 0.01 when comparing WT versus Kir6.2-KO.

Beyond structure, Kir6.2-KO also demonstrated significantly reduced cardiac performance manifested as reduction in fractional shortening after chronic cocaine administration compared to the baseline drug-free state (48.2 ± 4.2% pre- vs. 28.3 ± 4.2% post-cocaine; p < 0.05), a functional decline not developed by WT (47.5 ± 4.2% pre- vs. 35.4 ± 4.7% post-cocaine; Figure 3E). Moreover, peak ejection velocity after chronic cocaine abuse was significantly reduced in Kir6.2-KO compared to WT (0.49 ± 0.05 m/s vs. 0.67 ± 0.03 m/s, respectively; p < 0.01, Figure 3F), as demonstrated by Doppler ultrasound imaging (Figure 3G).

K_ATP channel deficiency precipitates cardiac myocyte enlargement after repeated cocaine administration

Activation of the sympathetic system by cocaine targets the cardiovascular system resulting in augmented cardiac workload and in turn cardiac hypertrophy.⁴⁰ Here, as a result of repetitive cocaine exposure, the cross-sectional area of cardiac myocytes was enlarged in Kir6.2-KO compared to WT (460 ± 15 μm² vs. 374 ± 23 μm², respectively; p = 0.01) as revealed by histological evaluation of postmortem heart preparations (Figure 4A), indicating maladaptive cardiac remodeling.

Figure 4.

(A) Aggravated hypertrophy in cocaine-challenged K_ATP channel deficient cardiomyocytes. Representative images of hematoxylin-eosin-stained cardiac tissue. (B) Analysis of histological heart sections demonstrated increased myocyte cross-sectional area in Kir6.2-KO (460 ± 15 μm², n = 4) in contrast to WT (374 ± 23 μm², n = 4) after 20 days of cocaine i.p. injections. Bars are mean ± SEM. *denotes p ≤ 0.01.

Discussion

Cocaine is one of the most commonly used illicit drugs, and a major public health problem.² Indeed, recent reports suggest that 3% of Americans between the ages of 15 and 64 currently abuse cocaine.⁴³ Although the effects of cocaine intoxication are systemic, cardiovascular distress is a major substrate of adverse outcome. Cocaine exerts its cardiac effect through a sympathomimetic action by inhibiting catecholamine re-uptake and stimulating adrenergic receptors. While activation of proximal and distal components of the cardiac sympathetic axis has been associated with outcome,^3,4 the present study provides evidence linking K_ATP channel genes as molecular determinants of cocaine cardiotoxicity.

Sarcolemmal K_ATP channels have the unique property to operate as high-fidelity homeostatic molecular rheostats capable of adjusting membrane excitability to match energy demand.^24,30 In this study, the diminished exercise capacity of K_ATP channel-deficient mice was accompanied by cardiac structural and functional abnormalities on echocardiography under chronic cocaine intoxication. Kir6.2-KO developed left ventricular dilatation and relative wall thinning compared to wild type as a result of cocaine abuse, indicating a progression toward cardiomyopathy in the absence of protective K_ATP channel function. This maladaptation underscores the inability of K_ATP channel-deficient organs to withstand continuous sympathomimetic action.²⁶ Indeed, the contractile function of Kir6.2-KO hearts deteriorates under isoproterenol challenge²⁶ or, as shown here, with chronic administration of intraperitoneal cocaine. Fractional shortening was compromised in Kir6.2-KO following 20 days of cocaine, with a reduction in peak ejection velocity compared to WT counterparts, implicating an essential role played by these biosensors in supporting cardiac performance under chronic cocaine use. In addition, the enlarged cardiomyocyte cross-sectional area revealed by histopathological evaluation of cardiac tissue in Kir6.2-KO compared to WT, as a result of repetitive cocaine exposure, indicates a pivotal role for the K_ATP channel in protecting myocardial cytostructure in response to cocaine abuse.

By processing signals of cellular distress, K_ATP channels control cardiac action potential duration, thus limiting Ca²⁺ influx during cardiac contraction and protecting the heart against calcium overload and downstream calcium-dependent signaling events leading to pathological remodeling under stress.^22,24,26 As one of the major mechanisms of cocaine's action is enhancement of sympathetic tone, leading to increased currents through L-type Ca²⁺ channels⁴⁴ and aggravated myocardial oxygen demand⁴⁵ with deterioration of ventricular performance,^41,46 K_ATP channels may provide a significant safeguard. Indeed, K_ATP channel deficiency has been related to intracellular Ca²⁺-overload predisposing to malignant structural remodeling under physiological or pathophysiological stress.⁷ In fact, Ca²⁺ channel blocking drugs, successfully used in protection against the toxic effects of cocaine in the heart, have been shown to rescue the impaired cardioprotection imposed by knockout of K_ATP channels.¹⁹

While the pharmacodynamics of cardiac cocaine toxicity are increasingly understood,³ the notion that susceptibility to cocaine-induced mortality may be genetically controlled has been only recently documented.⁵ In this regard, variance in K_ATP channel genes could contribute in determining cardiac outcome in chronic cocaine abuse. Indeed, there is growing evidence that implicates K_ATP mutations or polymorphisms in the susceptibility to human cardiac disease.^28–30 Studies in well-defined human cohorts will be necessary to establish the clinical implications of K_ATP channels in protection against cocaine cardiotoxicity.

Conclusion

In summary, this proof-of-principle study uncovers the importance of sarcolemmal K_ATP channels in determining the cardiovascular outcome in an experimental setting of chronic cocaine abuse. The presented data implicate the K_ATP channel as a candidate regulator of cardiovascular performance under the sympathomimetic effects of cocaine. This report thus identifies a previously unrecognized molecular determinant of cocaine cardiotoxicity, providing the foundation for translational studies in patient populations.