Mechanisms of Electrical Remodeling in Lipotoxic Guinea Pig Heart.

Abstract

Objectives:

To develop an adult guinea pig model of lipotoxicity and explore the underlying mechanisms associated with changes in the expression of the delayed rectifier potassium current (I_K).

Background:

Lipotoxicity may represent a common link among metabolic disorders and a higher vulnerability to arrhythmias.

Methods:

Whole-cell patch clamp, and palmitic acid (PA, a potent inducer of lipotoxicity), were used to assess mechanisms of short-term (~50 days) high-fat diet (HFD) feeding on atrial electrophysiology in guinea pig hearts and myocytes.

Results:

HFD fed guinea pigs were significantly heavier, displayed hypertriglyceridemia and hypercholesterolemia; but no signs of hyperglycemia or inflammation compared to low-fat diet fed controls. Increasing cardiac PA levels, resulted in shortened atrial action potential duration, and increased I_K density. Inhibition of phosphoinositide 3-kinase (PI3K) prevented increases in I_K due to PA. Acute (≥ 1hr) exposure of atrial myocytes to exogenous PA (1 mM) increased the density of the rapid delayed potassium current I_Kr, while it was decreased with the unsaturated oleic acid (OA, 1 mM). Serine-threonine protein phosphatase-2 (PP2A) inhibition with cantharidin reversed the effect of OA on I_Kr.

Conclusion and Implications:

Our data provide evidence of a novel lipotoxic guinea pig model with signs of vulnerability to arrhythmias. Inhibition of PA/PI3K/I_K and/or activation of the OA/PP2A/I_Kr pathways may be therapeutically beneficial for lipotoxic arrhythmias.

Introduction

Obesity is a major contributor to the increasing prevalence of cardiovascular diseases worldwide [1], with significant implications for supraventricular arrhythmias [2]. Obesity mechanisms may have a direct impact on the electrical activity of the atria (Figure 1A). Lipotoxicity or abnormal accumulation of serum free fatty acids (FFAs) [3], looms as a key mechanism underlying these interlinked pathologies [4].

Figure 1. Effects of LFD and HFD feeding in guinea pig.

A, Cartoon illustration of HFD induced pathological changes and effects on heart with implications for arrhythmogenesis. B, Cartoon representation of guinea pigs fed a LFD or HFD. C, Progressive body weight changes relative to time of guinea pigs fed with either LFD (◯, n=3), or HFD (Δ, n=5). Arrow (Thick) indicates the switch to a specific special diet.

Lipotoxicity is associated with changes in protein kinases (phosphatidylinositide 3-kinase (PI3K)) [5], and protein phosphatases (serine-threonine protein phosphatase-2, or PP2A) [6], PI3K upregulates I_Kr function in HEK293 cells, while dominant negative mutants reduced the current [7]. Reduced PP2A activity is associated with arrythmias including atrial fibrillation (AF) [8], and there is growing interest in PP2A activators in treating arrhythmias [8]. Thus, PP2A and PI3K are strong candidates for modulation of I_K function in lipotoxic heart.

The study presents a novel guinea pig model of lipotoxicity with early signs of atrial electrical remodeling (shortened action potential duration (APD) and increased I_K density), which may be useful for investigating mechanisms that initiate arrhythmias. Furthermore, PI3K inhibition prevented increases of I_K due to palmitic acid (PA), while PP2A inhibition prevented oleic acid (OA)-mediated depression of I_Kr. The data suggest that individual or multiple combinations of dietary (OA) and therapeutic interventions (potassium channel blockers, PI3K inhibitors, and PP2A activators) may be beneficial as early treatment options in obese patients with atrial arrhythmias.

Methods

This study was performed in accordance with the guidelines of the Columbia University and the Veterans Healthcare System Animal Care and Use Committees.

Electrophysiology

Whole-cell current recordings in adult guinea pig atria myocytes were performed as previously described [9, 10].

High-fat diet feeding in guinea pig model

Male and female guinea pigs (200-250 g or 2-3 weeks of age) were purchased from Charles River Laboratories (Wilmington, MA). Guinea pigs were fed ad libitum a low-fat diet (LFD), and high-fat diet (HFD) as previously described [11].

Intramyocardial injection of palmitic acid in guinea pigs

Bovine serum albumin-conjugated palmitic acid (PA-BSA, 1 mM, or 50 mg/kg) or vehicle (albumin alone, 50 mg/kg) was introduced into slightly anaesthetized [12] guinea pig hearts through the cranial vena cava, over 10 days. 12.5 mg/kg of PA-albumin or albumin alone were injected at day 0, 2, 8, and 10 (Figure 2A), to achieve a cumulative concentration of 50 mg/kg. All animals recovered within 2-5 minutes of the injection of PA-BSA or control albumin.

Figure 2. Modulation of atrial I_K by intramyocardial elevations of palmitic acid.

A, BSA conjugated palmitic acid (50 mg/kg) was injected in guinea pig cranial vena cava. B, Exemplar superimposed traces of atrial action potentials recorded from myocytes isolated from BSA-alone and PA+BSA injected guinea pigs. C. Comparison of average data of RMP, APD₉₀, and APD₃₀, measured in atrial myocytes from BSA-alone (Basal, ◻, n=3), and PA+BSA (lipotoxic, ∎, n=4) hearts. D, Exemplar traces of atrial I_K recorded using the voltage protocol shown on top of the traces. E, Pooled I_peak-V curves for I_K measured in atrial myocytes from BSA-alone (⚫, n=6 separate cells), and PA+BSA (▲, n=6 separate cells). Arrows indicate I_peak-V for the specific experimental condition. *P<0.05.

Isolation of guinea pig atrial myocytes

Primary atrial myocytes were isolated by enzymatic dissociation as previously described [9].

Preparation of bovine serum albumin conjugated FFA solutions

Palmitic acid (PA) stock solution was prepared as previously described [9].

Data and statistical analyses

Electrophysiological data were analyzed off-line using built in functions in Fitmaster (HEKA), Clampfit and Origin software. Current densities are expressed as current amplitudes (in pA) relative to cell size (in pF) or pA/pF. Data are reported as means±S.E.M. Statistical differences were determined using one-way ANOVA with Bonferroni post-hoc analysis or two-tailed unpaired t test for comparisons between groups and considered significant at P < 0.05.

Results

Effects of HFD feeding in guinea pig

The effects of short-term (~50 days), LFD and HFD feeding were investigated in male and female guinea pigs. As illustrated in Figure 1 switching from normal chow to the respective LFD or HFD led to progressive weight gain relative to time (Figure 1C). After ~50 days, HFD guinea pigs (Open triangles) were heavier (733±48 g, n=5, or a 19% increase) when compared to LFD-fed controls (589±10g, n=3, Open circles), in line with our previous report [9].

Short-term HFD feeding separates hyperlipidemia from hyperglycemia and inflammation in guinea pig

The Table shows the blood cell counts and glucose values before and after LFD or HFD feeding. We found that both complete blood cell (white blood cells, red blood cells, platelets, hemoglobin) count analyses and serum glucose measurements revealed that no significant changes were associated with HFD feeding, compared to LFD-fed guinea pigs.

Table.

Effect of LFD- and HFD-feeding on blood cell counts in guinea pigs 500 after 50 days.

	LFD (n=3)		HFD (n=3)
	PRE-DIET	POST-DIET	PRE-DIET	POST-DIET
White Blood Cells (x103)	4.07 ± 0.27	3.86 ± 0.44	4.61 ± 0.32	4.46 ± 0.41
Red Blood Cells (x 106)	4.92 ± 0.13	5.55 ± 0.25	4.57 ± 0.27	5.43 ± 0.15
Platelets (x 103/μl)	341.2 ± 59	519 ± 55	245.8 ± 63	427 ± 53
Hemoglobin (g/dL)	11.9 ± 0.44	13.2 ± 0.58	12.5 ± 0.71	13.28 ± 0.28
Glucose (mM)	9.39 ± 0.26	12.7 ± 1.19	9.07 ± 0.50	10.09 ± 0.21

Data are means ± S.E.M.

P > 0.05 compared to LFD non-obese controls, one-way ANOVA and Bonferroni test.

Determination of intramyocardial palmitic acid effects on atrial myocyte electrophysiology

Next the effect of lipotoxicity directly on cardiac function was evaluated using intracardiac injection of saturated PA (a potent inducer of lipotoxicity) [13]. After 40 days, electrophysiological assays were conducted (Figure 2A). Figure 2B shows representative superimposed AP traces recorded in myocytes isolated from guinea pigs injected with PA-BSA and vehicle control (BSA alone). On average atrial myocytes from PA-BSA guinea pigs displayed significantly shorter APD₉₀ (107.9±13.3 ms, n=3 or 33%, *P<0.05, Figure 2C) compared to BSA alone control (151.3±3.42 ms, n=4, Figure 2C). Average APD₃₀ was reduced to 41.3±2.67ms, (or by 68%, n=3, *P<0.05, Figure. 2C) from a control value of 92.5±10ms (n=4). Resting membrane potential (RMP) was more depolarized in guinea pigs injected with PA-BSA (−47±2 mV, n=3, *P<0.05, Figure. 2C), compared to control (−70 ±1.44 mV, n=4), suggesting that a reduced function of other major atrial ion channels (TWIK-related acid sensitive potassium channel or TASK-2) could depolarize RMP [14] in our lipotoxic guinea pig model and contribute to the vulnerability to atrial arrhythmogenic events.

Effects of intramyocardial palmitic acid on atrial I_K in guinea pig heart

The potential mechanisms of adverse atrial electrical remodeling in PA+BSA injected hearts were assessed by measuring I_K, because of its role in repolarization and that dysfunction of I_K [15, 16] contributes to arrhythmias [17]. Figure 2D shows typical whole-cell atrial I_K current traces measured in guinea pig atrial myocytes from PA+BSA- and BSA alone-injected hearts (controls). Compared to controls PA-BSA injection increased I_K densities at all potentials positive to +10 mV (Figure 2E). At +50 mV, I_K peak density was increased from 4.79±0.41 pA/pF (n=6) in control to 6.88±1.13 pA/pF, (n=6, or by 30.4%, P>0.05, Figure. 2E) with PA-BSA. At +100 mV averaged I_K density was significantly increased from 11.3±3.42 pA/pF (n=6, control) to 22.1±2.34 pA/pF (or by 48.7% n=6, *P<0.05) with PA+BSA.

Effects of LY294002 on palmitic acid-induced increases in I_K density in atrial myocytes

To interpret the mechanistic bases of the adverse effects by lipotoxicity on atrial electrical properties, effects of exogenous pretreatment with inhibitors of critical metabolic pathways on I_K density were determined in atrial myocytes. PI3K, is a serine/threonine kinase which is sensitive to lipotoxicity [5] and a key regulator of cardiac ion channels [18]. Figure 3A shows typical I_K currents measured in basal conditions (i.e. untreated, BSA alone). Exogenous pretreatment with PA+BSA (1 mM) significantly increased I_K density (Figure 3B), while pre-exposure to the PI3K inhibitor, LY294002 (10μM) [19], completely reversed this effect (Figure 3C) consistent with a role for PI3K in the facilitatory effect of PA on I_K current density in atrial myocytes. On average PA+BSA significantly increased I_max, or maximal current at +30 mV from 2.59±0.44 (n=10, control) to 5.27±0.22 pA/pF (n=9, PA+BSA) (or by ~51%, *P<0.05; Figure 3D). I_K current density measured in PA+BSA-treated cells that were pre-exposed to LY294002 was 2.91±0.42 pA/pF (n=9, P>0.05) compared to 2.59±0.44 pA/pF (BSA alone, n=10, Figure 3D) measured in basal conditions.

Figure 3. Effects of PI3K and PP2A inhibition on I_K and I_Kr densities in guinea pig atrial myocytes.

Representative I_K current traces measured in untreated (n=10, A), palmitic+BSA (n=9, B) and palmitic+BSA+LY294002-treated (n=9, C) atrial myocytes by applying the voltage protocol shown at the top (A). D, Pooled I_peak-V curves for the different conditions. E, I_Kr tail current traces recorded by applying the protocol shown at the top. F, I_peak-V curves for currents recorded in untreated (Basal, n=8, Top), palmitic+BSA (n=7, Middle), and oleic+BSA (n=5, Bottom). G,H, Data for myocytes untreated (Top), oleic+BSA- (Middle), and oleic+BSA+cantharidin-treated (n=3, Bottom). H, Pooled I_peak-V curve for I_Kr in the presence of oleic+BSA+cantharidin. Data for basal (Red) and OA+BSA (Cyan) I_Kr are reproduced from F for visual comparison. *P<0.05.

Effects of exogenous palmitic acid and oleic acid on I_Kr in native atrial myocytes

Previously we demonstrated that OA prolonged atrial APD and severely depressed I_Kr density in HEK293 cells but had no effect on I_Ks, while PA shortened APD and increased both I_Kr and I_Ks densities [9]. Next we determined the effects the lipids on I_Kr currents measured in native atrial cardiomyocytes. We used the whole-cell patch clamp technique to assess the functional properties of I_Kr in atrial myocytes incubated ≥1hr with either 1 mM BSA control, PA, or OA. (Figure 3E). Compared to control or basal I_Kr (Figure 3E, Top), PA augmented (Figure 3E, Middle), whereas OA+BSA severely depressed I_Kr (Figure 3E, Bottom). PA+BSA significantly increased I_max, or maximal current at +30 mV from 0.29±0.03 (BSA alone, n=7 separate myocytes) to 0.91±0.12 (n=8 separate myocytes) (or by ~68%, *P<0.05; Figure 3F), whereas OA+BSA significantly reduced I_max by ~45% (0.16±0.01, n=5 separate myocytes; *P<0.05; Figure 3F). The data suggest that alteration of cellular lipids, either by disease, diet, or metabolic state, can potentially affect the expression of I_K channels with significant implications for cardiac function.

Effects of Cantharidin on oleic acid-induced depression of I_Kr density in atrial myocytes

Our results presented in Figure 3E and Figure 3F suggests that OA through inhibition of I_Kr, may be a key anti-arrhythmic mechanism in lipotoxicity by antagonizing the shortening of APD. Further there are reports that reduced PP2A activity is associated with AF through modulation of cardiac ion channels [8]. To elucidate the mechanism of APD prolongation and reduced I_Kr density by OA, the effect of PP2A inhibition with cantharidin (30 μM) on I_Kr was examined. Pretreatment (10-mins) of atrial myocytes with cantharidin prevented OLA+BSA-mediated depression of I_Kr (Figure 3G). At +30 mV I_Kr density was significantly (*P<0.05) increased from 0.16±0.01 pA/pF (n=9) in the presence of OA+BSA to 0.49±0.17 pA/pF (or increased by ~67%, n=3, Figure 3G,H) with cantharidin. The data suggests that activation of the OA/PP2A/I_Kr pathway may prevent or lower threshold of vulnerability to supraventricular arrhythmias.

Discussion

Lipotoxicity is an independent risk factor for arrhythmias in obese patients; however the mechanisms are poorly understood. The goal of the present study is to determine whether we could examine the effects of HFD-induced lipotoxicity on atrial electrophysiology in guinea pig independent of other obesity co-morbidities. Short-term HFD diet feeding in guinea pig separates hyperlipidemia from hyperglycemia and inflammation. Lipotoxic hearts also displayed atrial electrical dysfunction (Figure 4A). The data suggests that short-term HFD diet feeding in guinea pig permits investigation of lipotoxic mechanisms and atrial electrical remodeling prior to supraventricular arrhythmias.

Figure 4. Hypothesized mechanisms linking palmitic acid and oleic-acid activated pathways to atrial electrical remodeling.

A, The data suggests that high-fat diet feeding will promote cardiac lipotoxicity, leading to shortened APD and increase vulnerability to atrial arrhythmogenesis. B, Proposed PA-induced paradigm leading to atrial arrhythmias. This involves activation of the PI3K signaling pathway leading to altered I_K channel biophysical properties and posttranslational modifications. Alternatively, increasing dietary OA is anti-arrhythmic, possibly through increased PP2A activity and depression of I_Kr density that may contribute to normal electrocardiogram (C).

Therapeutic effects of PI3K inhibition on short-term HFD-induced modulation of I_K in guinea pig myocytes

Shortening of atrial APD contributes to mechanisms (refractory periods, reentry circuits) [20], that lead to arrhythmias. Therefore, enhanced I_K density will be expected to accelerate repolarization and contribute to supraventricular arrythmias. The relevance of I_K mechanisms is highlighted by recent studies in AF patients [21, 22]. Mechanistically, increased I_K may be due to altered posttranslational modifications [22, 23]. As shown in this study, inhibition of the lipid kinase PI3K [24], inhibited I_K density and suggests that myocardial PI3K inhibition has the potential to be anti-arrhythmic by prolonging APD and reducing vulnerability to supraventricular arrhythmias (Figure 4B). This is supported by a previous report in mouse hearts lacking PI3K p110α catalytic subunit, demonstrating prolonged APD and QT interval [25]. Diabetic mice which lacks the PI3K p110γ or expressing the catalytically inactive p110γ prevented systolic and diastolic dysfunction [26]. Therefore, it would also be valuable to explore the link between PI3K, I_K and atrial arrhythmias.

Beneficial effect of unsaturated oleic acid in limiting atrial arrhythmogenesis due to short-term HFD feeding in guinea pig heart

Our finding that exogenous OA selectively inhibited I_Kr density in atrial myocytes, is in-line with its role in atrial APD prolongation in obese myocytes [9]. Electrophysiological studies have shown that fish oil diets prevented cardiac hypertrophy in a rat model of pressure overload [27] and a rabbit model of heart failure [28] through modulation of potassium channels [28]. Therefore, dietary supplementation of OA may be a valuable anti-arrhythmic mechanism in lipotoxicity by antagonizing the shortening of APD and raising the threshold for inducibility of atrial tachycardia (Figure 4C).

One of the well-accepted mechanisms underlying initiation and maintenance of supraventricular arrythmias is impaired posttranslational regulation due to reversible phosphorylation by protein phosphatases [29]. Furthermore, reduced PP2A activity is associated with AF [8], suggesting therapeutic potential of PP2A activators. Evidence shows that decreased I_Ca,L current density associated with AF [22] is due to increased calcium channel dephosphorylation by increased activity of PP2A [8]. The findings of this study further show that PP2A inhibition prevented decreases in I_Kr current density due to OA and suggests that PP2A activation contributed to OA-induced APD.

These results suggest that in lipotoxic atrial myocytes: 1) activation of endogenous of PI3K may underlie increased densities of I_Kr and I_Ks; and 2) activation of native PP2A may mediate the selective down regulation of atrial I_Kr seen in the presence OA. Overall, the DATA preview the new mechanistic insights likely to be gathered from future studies (Figure 4).

Taking together these results and the findings of Aromolaran et al [9], short-term HFD feeding in guinea pig leads to significant weight gain, lipotoxicity and adverse remodeling of atrial I_K. These findings have potential implications for atria electrical activities, possibly through significant elevations in intramyocardial saturated palmitic acid. It is intriguing to speculate that timely interventions that either prevent: 1) ectopic lipid accumulation (anti-lipotoxic drugs that divert lipid to adipose stores or decrease systemic lipid levels); 2) increased I_Ks function (dietary unsaturated FFAs) may be anti-arrhythmic, and therefore beneficial to obese patients with cardiac events (Figure 4A).

Highlights.

1. Short-term high-fat diet feeding selectively induced lipotoxicity in guinea pig.

2. Lipotoxic hearts showed shortened atrial action potential and increased I_K density.

3. PI3K inhibition prevented increases in I_K due to palmitic acid.

4. PP2A inhibition reversed decreases in the rapid I_K current due to oleic acid.

5. Lipotoxicity is a key mechanism in obesity-related arrhythmias.

Nonstandard Abbreviations:

HFD

High-fat diet

FFA

Free-fatty acid

PA

Palmitic acid

OA

Oleic acid

IK

Delayed rectifier K current

PI3K

Phosphoinositide 3- kinase

PP2A

Serine threonine protein phosphatase-2