Action Potential Shortening and Impairment of Cardiac Function by Ablation of Slc26a6

Abstract

Background

Intracellular pH (pH_i) is critical to cardiac excitation and contraction; uncompensated changes in pH_i impair cardiac function and trigger arrhythmia. Several ion transporters participate in cardiac pH_i regulation. Our previous studies identified several isoforms of a solute carrier Slc26a6 to be highly expressed in cardiomyocytes. We show that Slc26a6 mediates electrogenic Cl⁻/HCO₃⁻ exchange activities in cardiomyocytes, suggesting the potential role of Slc26a6 in regulation of not only pH_i, but also cardiac excitability.

Methods and Results

To test the mechanistic role of Slc26a6 in the heart, we took advantage of Slc26a6 knockout (Slc26a6^−/−) mice using both in vivo and in vitro analyses. Consistent with our prediction of its electrogenic activities, ablation of Slc26a6 results in action potential (AP) shortening. There are reduced Ca²⁺ transient and sarcoplasmic reticulum Ca²⁺ load, together with decreased sarcomere shortening in Slc26a6^−/− cardiomyocytes. These abnormalities translate into reduced fractional shortening and cardiac contractility at the in vivo level. Additionally, pH_i is elevated in Slc26a6^−/− cardiomyocytes with slower recovery kinetics from intracellular alkalization, consistent with the Cl⁻/HCO₃⁻ exchange activities of Slc26a6. Moreover, Slc26a6^−/− mice show evidence of sinus bradycardia and fragmented QRS complex, supporting the critical role of Slc26a6 in cardiac conduction system.

Conclusions

Our study provides mechanistic insights into Slc26a6, a unique cardiac electrogenic Cl⁻/HCO₃⁻ transporter in ventricular myocytes, linking the critical roles of Slc26a6 in regulation of pH_i, excitability, and contractility. pH_i is a critical regulator of other membrane and contractile proteins. Future studies are needed to investigate possible changes in these proteins in Slc26a6^−/− mice.

Introduction

Acid-base balance is critical for maintaining normal cardiac function. An uncompensated shift of intracellular pH (pH_i) generates abnormal electrical activities, contractile disorder, and altered intracellular Ca²⁺ signaling.^{1, 2} The bicarbonate (HCO₃⁻) / carbon dioxide (CO₂) buffering system plays a central role in regulation of pH_i. Therefore, transport of HCO₃⁻ across the plasma membrane is critical in maintaining cellular pH homeostasis. Several HCO₃⁻ transporters have been identified in cardiomyocytes, including Na⁺/HCO₃⁻ (NBC) co-transporter and Cl⁻/HCO₃⁻ exchanger (AE or CBE).¹ NBC mediates acid extrusion, while AE mediates acid loading.

Solute carrier Slc26a6 is a versatile anion exchanger, first identified in epithelial cells, with abundant expression in kidneys, pancreas, intestines, and placenta.^3–5 It mediates Cl⁻/oxalate, Cl⁻/HCO₃⁻, Cl⁻/OH⁻, Cl⁻/SO₄²⁻, and Cl⁻/formate exchange activities. It was previously reported that Slc26a6 is the predominant Cl⁻/HCO₃⁻ and Cl⁻/OH⁻ exchanger in the mouse heart, based on the transcript levels.⁶ Our recent functional study identified four cardiac Slc26a6 isoforms, which mediate Cl⁻/HCO₃⁻ exchange in atrial and ventricular myocytes⁷. Importantly, we found that Cl⁻/HCO₃⁻ exchange by cardiac Slc26a6 is electrogenic. Indeed, cardiac Slc26a6 represents the first electrogenic exchanger with only anionic substrates and provides active transport of Cl⁻ into the cells to regulate intracellular Cl⁻ homeostasis. In addition, activities of Slc26a6 are predicted to alter the pH_i by transporting HCO₃⁻ or OH⁻. Therefore, we hypothesize that cardiac Slc26a6 may play a unique role in the regulation of not only pH_i, but also cardiac excitability and function.

To directly test the hypothesis, we took advantage of a Slc26a6 knockout (Slc26a6^−/−) mouse model⁸ using in vivo and in vitro analyses. Under physiological conditions, Slc26a6 activities are predicted to coincide with the dynamic range of cardiac action potential (AP), generating mainly inward currents, with outward current beyond ~ +36 mV⁷. Indeed, our study reveals that ablation of Slc26a6 results in AP shortening, consistent with electrogenic activities of Slc26a6. The shortened AP durations (APDs) result in reduced Ca²⁺ transient and sarcoplasmic reticulum (SR) Ca²⁺ load, along with decreased sarcomere shortening in Slc26a6^−/− cardiomyocytes. These abnormalities translate into reduced fractional shortening and cardiac contractility at the in vivo level. Importantly, pH_i is elevated in Slc26a6^−/− cardiomyocytes with slower recovery kinetics from intracellular alkalization, consistent with Cl⁻/HCO₃⁻ exchange activities of Slc26a6. Moreover, Slc26a6^−/− mice show evidence of sinus bradycardia and fragmented QRS complex, supporting the critical role of Slc26a6 in cardiac conduction system.

Materials and Methods

All animal care and procedures were performed in accordance with the protocols approved by Institutional Animal Care and Use Committee of the University of California, Davis and in accordance with National Institutes of Health guidelines. De-identified human ventricular specimens were obtained from a commercial source, in accordance with the approved UC Davis Institutional Review Board (IRB) protocol. Please refer to the Online Supplemental Data for additional details.

We used 129S6/SvEv wild type (WT) and Slc26a6^−/− mice previously generated and reported⁸ (a generous gift from Dr. Peter S. Aronson, Yale University) in our study. All experiments described in the study were conducted in a blinded fashion.

Cardiac tissue preparation and cardiomyocyte isolation

Mice were anesthetized by intraperitoneal injection of 80 mg/kg of ketamine and 5 mg/kg of xylazine. Cardiomyocytes were isolated as previously described⁷

Genotyping, western blot, and histological analyses

Histological and western blot analyses of cardiac tissue were performed as we have previously described⁹. Genotyping analyses of Slc26a6^−/− mice are presented in Supplemental Figure 1.

Electrocardiographic Recordings

ECG recordings were performed as previously described^{10, 11}.

Analysis of cardiac function by echocardiography

Echocardiograms to assess systolic function were performed in conscious animals as we have previously described.¹⁰

Hemodynamic monitoring

Mice were anesthetized by intraperitoneal injection of 80 mg/kg of ketamine and 5 mg/kg of xylazine and maintained at 37 ºC. Recording of pressure and volume was performed by using Millar Pressure-Volume System MPVS-300 (Millar, Inc., Houston, TX), Power Lab, and Lab Chart 6.0 software (AD Instruments, Colorado Springs, CO).

Patch-clamp recordings

Whole-cell current⁷ and AP recordings¹⁰ were performed as we have previously described. For current recordings, the clamped and suspended whole-cell was alternately exposed to two capillary tubes dispensing control and test solutions at a potential of 0 mV.⁷ For the outward current recording, we used the same solutions as we previously documented⁷.

For the inward current recording, the pipette solution contained (in mM): 24 NaHCO₃, 116 Na Glutamate, 1 EGTA, 1.4 NaCl, and gassed with 5% CO₂ and 95% O_2. The bath control solution was the same as pipette solution. The bath test solution with 140 mM Cl⁻ contained (in mM): 140 NaCl, 10 HEPEs, 1 EGTA, and pH 7.4; the bath test solution with 14 mM Cl⁻ contained (in mM): 14 NaCl, 126 Na Glutamate, 10 HEPEs, 1 EGTA, and pH 7.4. Current recordings were performed at the room temperature and APs were recorded at 36 °C.

Measurement of sarcomere shortening and Ca²⁺ transient (CaT)

We used IonOptix sarcomere detection (IonOptix LLC., Westwood, MA) and the fast Fourier transform (FFT) method¹². Contraction was measured using a high-speed camera (MyoCam-S, 240 to 1000 frames/s) to record sarcomere movement. The sarcomere pattern was used to calculate the sarcomere length using an FFT algorithm. The fractional shortening was calculated as the percentage change in sarcomere length during contraction. Simultaneous Ca²⁺ transients (CaT) were recorded using Fura-2 dual-wavelength ratiometric method¹², which is more precise than the Fluo-4 single-wavelength method.

The recording solution contained (in mM): 145 NaCl, 4 KCl, 1 CaCl₂, 0.33 NaH₂PO₄, 1 MgCl₂, 10 Glucose, 10 HEPES, with pH 7.4. In a separate set of experiments, HCO₃⁻ was used as a buffer, and the recording solution contained (in mM): 120 NaCl, 4 KCl, 1 CaCl₂, 0.33 NaH₂PO₄, 1 MgCl₂, 10 Glucose, 24 NaHCO₃, gassed by 5% CO₂ and 95% O₂. To reduce the pH_i, but keep pH_o constant, sodium acetate was applied to clamp the pH_i^{13, 14}. Sodium acetate was used to replace an equal concentration of NaCl, and 5 μM 5-(N-Ethyl-N-isopropyl) amiloride (EIPA) was added to inhibit the Na⁺/H⁺ exchanger for acetate application.

Intracellular pH (pH_i) measurement

The pH_i of cardiomyocytes was measured using carboxy-SNARF-1 fluorescent pH indicator. Isolated cardiomyocytes were loaded with 10 μM SNARF-1 AM in Tyrode’s solution. The measurement was performed using Na⁺-free solutions, buffered by either HEPES or HCO₃⁻. HEPES-buffered solution contained (in mM): 144 NMG-Cl, 4 KCl, 1 MgCl₂, 2 CaCl₂, 10 Glucose, 10 HEPES, and pH 7.4. HCO₃⁻-buffered solution contained (in mM): 120 NMG-Cl, 4 KCl, 1 MgCl₂, 2 CaCl₂, 10 Glucose, 24 Choline-HCO₃, gassed by 5% CO₂ and 95% O₂. Sodium acetate was used to clamp pH_i, and 5 μM 5-(N-Ethyl-N-isopropyl) amiloride (EIPA) was added to inhibit the Na⁺/H⁺ exchanger for acetate application.

The SNARF emission ratio (F₅₈₀/F₆₄₀) was converted to pH_i using standard calibration^15–17. SNARF emission ratios were measured while cells were perfused by calibration solutions with five different pH values. The calibration solutions contained (in mM): 140 KCl, 1 MgCl₂, 20 HEPES (or MES at pH 5.5), with pH 5.5, 6.5, 7.0, 7.5, and 8.0. Negericin (10 μM, a K⁺/H⁺ antiporter ionophore) was added to the calibration solution before use.

Molecular cloning from human cardiac tissues

De-identified human ventricular specimens were obtained from a commercial source (T Cubed), in accordance with the approved UC Davis Institutional Review Board (IRB) protocol. Total RNA and mRNA were extracted, and similar cloning strategy was used as previously described⁷.

Heterologous expression in Chinese hamster ovary (CHO) cells

Human cardiac SLC26A6 isoforms were expressed in CHO cells following the protocol we used previously⁷.

Statistical analyses

Data are presented as mean ± S.E.M. Statistical comparisons were analyzed by student’s t-test. Statistical significance was considered to be achieved when p<0.05.

Results

Ablation of Slc26a6^−/− results in cardiac AP shortening

To test the functional roles of Slc26a6, we took advantage of an Slc26a6^−/− mouse model.⁸ We first confirmed the lack of Slc26a6 expression in ventricular myocytes isolated from Slc26a6^−/− mice (Fig. 1A), and the two bands in WT lane suggest different mouse cardiac slc26a6 isoforms we reported before⁷. To test the role of Slc26a6 in Cl⁻/HCO₃⁻ and Cl⁻/oxalate exchange, we recorded outward currents using fast solution exchange and suspended whole-cell recording methods, as we have previously described.⁷ Fig. 1B&C show the solution exchange configuration and corresponding currents, respectively, recorded from WT and Slc26a6^−/− ventricular myocytes. The outward Cl⁻/HCO₃⁻ and Cl⁻/oxalate exchange currents were completely abolished in Slc26a6^−/− ventricular myocytes (Fig. 1C, right).

Figure 1. Ablation of Slc26a6^−/− results in cardiac AP shortening.

A. Western blot analysis confirms the absence of Slc26a6 expression in Slc26a6^−/− mouse hearts. B. Schematic diagram of the fast solution exchange of the suspended whole-cell recording (left), and the ionic configuration to activate outward exchange current through Slc26a6. For recordings of the outward Slc26a6 currents, cardiomyocytes were first clamped at 0 mV, suspended, and moved to the outlet of the solution exchange capillary tube dispensing the bath control solution. Then the myocyte was switched to the capillary tube dispensing bath test solution, containing oxalate or HCO₃⁻, for the activation of outward currents. The fast solution switch was performed by SF-77 solution exchanger controlled by pClamp10 software. The compositions of the bath control and test solutions have been described in Methods. C. Representative recording traces of outward currents recorded from WT ventricular myocytes (left) and Slc26a6^−/− ventricular myocytes (right). The black and blue bars under the current traces represent the solution exchange to bath test solutions to activate Cl⁻/Oxalate and Cl⁻/HCO₃⁻ exchange currents, respectively. D. Representative traces of action potentials (APs) from WT and Slc26a6^−/− ventricular myocytes. Dashed lines represent 0 mV. E. Comparisons of resting membrane potentials (RMPs). F. Summary data of the action potential durations at 50% (APD₅₀) and 90% repolarization (APD₉₀). The AP and RMP were recorded in HCO₃⁻ buffered bath solution. G. Inward Cl⁻/HCO₃⁻ exchange current elicited from WT ventricular myocytes using fast solution exchange, as indicated in the diagram on the left. The red and blue bars above the current traces represent the solution exchange to bath test solutions containing 14 or 140 mM Cl⁻ to activate Cl⁻/HCO₃⁻ exchange currents, respectively.

Since cardiac Slc26a6 is an electrogenic Cl⁻/HCO₃⁻ exchanger, we predict that activities of Slc26a6 will affect cardiac APDs. There were no significant differences in the resting membrane potentials (RMPs), however, APD at 90% repolarization (APD₉₀) was significantly shorter in Slc26a6^−/− ventricular myocytes compared to WT (Fig. 1D, E and F). It has previously been reported that the stoichiometry of cardiac Slc26a6 for HCO₃⁻:Cl⁻ is 2 or greater, therefore, the estimated reversal potential of the exchange is ~+36 mV under physiological condition. Our findings are consistent with the prediction that Slc26a6 generates inward currents by electrogenic Cl⁻/HCO₃⁻ exchange through most of the AP, due to the positive reversal potential for Cl⁻/HCO₃⁻ exchange under physiological condition in cardiomyocytes.⁷ Indeed, a prominent inward current can be generated in cardiomyocytes via Cl⁻/HCO₃⁻ exchange with 24 mM HCO₃⁻ inside the cells (Fig. 1G).

The APD shortening may also be caused by the remodeling of other major ion channels such as Ca²⁺ and K⁺ channels due to the knockout of Slc26a6. We therefore recorded the Ca²⁺ and total K⁺ currents from WT and Slc26a6^−/− ventricular myocytes. As shown in supplemental Figure 2, we did not find significant differences in the current density between WT and Slc26a6^−/− myocytes.

Slc26a6^−/− cardiomyocytes show reduced sarcomere shortening, Ca²⁺ transient, and SR Ca²⁺ load

The shortening of cardiac APs in Slc26a6^−/− is predicted to result in decreased Ca²⁺ entry and possible impairment of cardiomyocyte contractility. We, therefore, tested the Ca²⁺ transient (CaT), sarcomere shortening, and SR Ca²⁺ load in WT and Slc26a6^−/− ventricular myocytes using HEPES (Cl⁻/OH⁻ exchange, Fig 2A–D) or HCO₃⁻ (Cl⁻/HCO₃⁻ exchange, Fig. 2E–H) buffered solutions. Compared to WT cardiomyocytes, Slc26a6^−/− cardiomyocytes demonstrated significantly reduced fractional shortening and CaT amplitude (Fig. 2A&B). To test whether the decrease in CaT was due to reduced SR Ca²⁺ content, we measured the SR Ca²⁺ content in WT and Slc26a6^−/− cardiomyocytes (Fig. 2C&D). Indeed, the SR Ca²⁺ load was significantly reduced in Slc26a6^−/− cardiomyocytes. Similar findings are shown for HCO₃⁻ buffered solution (Fig. 2E–H).

Figure 2. Slc26a6^−/− cardiomyocytes show reduced sarcomere shortening, Ca²⁺ transient, and SR Ca²⁺ load.

Experiments were performed using HEPES (panels A–D) or HCO₃⁻ (panels E–H) as buffers. A, E. Representative traces and summary data for percentages of sarcomere shortening, and B, F. Ca²⁺ transient (CaT) measured using Fura-2 ratio (R_FURA) from WT and Slc26a6^−/− cardiomyocytes. C, G. Representative traces of sarcoplasmic reticulum (SR) Ca²⁺ load measurement of WT and Slc26a6^−/− ventricular myocytes. The myocytes were first paced to steady state, and then pacing was stopped for 15 s followed by a rapid application of 20 mM caffeine to induce SR Ca²⁺ release. D, H. Summary data for SR Ca²⁺ load from WT and Slc26a6^−/− ventricular myocytes. Numbers within the bar graphs represent the sample sizes.

Roles of Slc26a6 in pH_i regulation

To test whether the Cl⁻/HCO₃⁻ exchange, mediated by Slc26a6, contributes to pH_i regulation in ventricular myocytes, we directly measured the pH_i of cardiomyocytes by using SNARF-1 AM pH dye, as shown in Fig. 3A and B. The pH_i in resting Slc26a6^−/− cardiomyocytes was 7.42 ± 0.02, significantly higher than that of 7.16 ± 0.02 in WT cardiomyocytes (n=5, *p<0.05). To test the function of Slc26a6 in acid loading, we first induced intracellular acidification by wash in of 40 mM acetate followed by an immediate wash out, and monitored the pH_i recovery from alkalization. The pH_i in WT cardiomyocytes recovered to the baseline level over time (Fig. 3C), whereas there was a delay in the recovery of pH_i in Slc26a6^−/− cardiomyocytes, over the same period of time (Fig. 3D). The results suggested an impairment of the acidification process in Slc26a6^−/− cardiomyocytes and participation of Slc26a6 in the acid loading in cardiomyocytes.

Figure 3. Slc26a6^−/− cardiomyocytes show higher intracellular pH (pH_i).

A. Measurement of pH_i in ventricular myocytes by SNARF-1 AM using confocal microscopy (scale bar: 10 μm). Loaded cells were excited by 488 nm laser and images were acquired by using two band-pass filters with center wavelengths of 580 nm and 640 nm, respectively. The ratio of the emission signals at 580 nm and 640 nm was calculated and converted to pH value following the standard methods described before^15–17. A bright field (BF) image of the cardiomyocyte was also shown at the lower panel. B. Comparisons of pH_i between WT and Slc26a6^−/− cardiomyocytes. Alternations and recoveries of pH_i when 40 mM acetate was applied in WT (C) and Slc26a6^−/− (D) ventricular myocytes. EIPA (5 μM) was added to inhibit the Na⁺/H⁺ exchanger during acetate application. E. Effects of 40 mM acetate on the CaT and sarcomere length in WT and Slc26a6^−/− ventricular myocytes. EIPA (5 μM) was added to inhibit the Na⁺/H⁺ exchanger during acetate application. The recovery process after wash out of acetate for Slc26a6^−/− myocytes is much slower than that of WT myocytes. The black traces showed the CaT, and the blue traces showed the sarcomere length.

We reason, based on its exchange activities, that Slc26a6 contributes to acid loading in cardiomyocytes under normal physiological condition, and that ablation of Slc26a6 may delay the acid loading process and affect the sarcomere shortening and CaT. We perfused ventricular myocytes with a solution containing 40 mM sodium acetate to acidify the cytosol, and monitored the sarcomere shortening and CaT, as shown in Fig. 3E. Acid loading resulted in marked reduction of sarcomere shortening in both WT and Slc26a6^−/−. The CaT was initially enhanced slightly, and then inhibited by acetate-induced intracellular acidification. Moreover, after wash out of acetate, the recovery process of sarcomere shortening and CaT was much slower in Slc26a6^−/− cardiomyocytes, consistent with Slc26a6 exchange activities (please note that different scales were used for upper and lower panels). The immediate alkalization during the wash out of acetate also inhibited contraction of the cardiomyocytes.

A decrease in fractional shortening in Slc26a6^−/− mice

The above cellular data strongly support the contributions of Slc26a6 to cardiac pH_i, APD, and contractility. To test the effects of reduced CaT and sarcomere shortening in Slc26a6^−/− on in vivo cardiac function, we performed echocardiography in conscious mice. There was no evidence of cardiac hypertrophy or dilation, with no significant differences in heart/body weight ratios between WT and Slc26a6^−/− mice (Fig. 4A&B). However, Slc26a6^−/− mice showed evidence of sinus bradycardia (Fig. 4C). Histological analyses, using wheat germ agglutinin and Picrosirius Red stain, of cardiac sections demonstrated no evidence of cardiac fibrosis, hypertrophy, or dilation in Slc26a6^−/− mice (Fig. 4D). However, there was a significant reduction in the fractional shortening in Slc26a6^−/− mice compared to WT controls (Fig. 4E).

Figure 4. Slc26a6^−/− mice demonstrate worsening fractional shortening and sinus bradycardia.

A. Photomicrographs of whole heart of WT and Slc26a6^−/− mice (Scale bar,1 mm). B. Summary for heart/body weight ratio in mg/g. C. Heart rate. (BPM: beat per minute). D. Wheat germ agglutinin (left two panels) and Picrosirius Red (right two panels) staining of cardiac sections from WT and Slc26a6^−/− mice demonstrating no evidence of increase fibrosis in the knockout mice (scale bar, 0.5 mm). E. 2D and M-mode echocardiography from WT and Slc26a6^−/− mice. The right panel shows summary data for fractional shortening (%FS), illustrating a significant decrease in %FS in KO mice compared to WT.

Indeed, hemodynamic monitoring demonstrated a significant decrease in cardiac contractility in Slc26a6^−/− mice compared to WT controls. Shown in Fig. 5A are representative recordings of left ventricular pressure, volume, and developed pressure (dP/dt) in WT and Slc26a6^−/− mice. There was a significant right and downward shift of the pressure-volume (P-V) loops in Slc26a6^−/− mice, indicating reduced end systolic pressure and relatively larger end systolic and diastolic volumes, with reduced stroke volume (Fig. 5B). To determine the end systolic P-V relationship, we obtained a series of P-V loops by altering the preload and derived the slope of the end systolic P-V relationship (Ees), a load independent measure of cardiac contractility (Fig. 5C&D for WT and Slc26a6^−/−, respectively). Consistent with the cellular data and echocardiography data, the end systolic pressure (Fig. 5E), maximum dP/dt (Fig. 5F), and Ees (Fig. 5G) in Slc26a6^−/− mice were significantly reduced compared to those of WT mice, supporting a decrease in cardiac contractility in Slc26a6^−/− mice.

Figure 5. Hemodynamics monitoring from Slc26a6^−/− and WT mice demonstrates the reduced cardiac contractility in Slc26a6^−/− mice.

A. Recording traces of left ventricular pressure, volume, and derivative of pressure with respect to time (dP/dt) from WT and Slc26a6^−/− mice. B. P-V loop analysis. The pressure and volume have been calibrated. The volume calibration was performed using cuvette (P/N 910-1049, Millar Inc.) filled with fresh heparinized 37° C mouse blood. A series of P-V loops when the preload was altered in WT (C) and Slc26a6^−/− mice (D). E, F, G. Summary data for end systolic pressure (ESP), maximum dP/dt, and the slope of the end systolic P-V relationship (Ees).

Fragmented QRS complex in Slc26a6^−/− mice

Cellular electrophysiology and pH_i measurement of Slc26a6^−/− cardiomyocytes demonstrated the shortening of APD₉₀ and an elevated pH_i (Fig. 1 & 3). We, therefore, compared the ECG between WT and Slc26a6^−/− mice (Fig. 6A&B). Slc26a6^−/− mice showed evidence of fragmented QRS complex with prolonged PR interval. The QRS complex duration of Slc26a6^−/− mice was significantly prolonged compared to that of WT mice (22.8 ± 0.8 ms vs 20.3 ± 0.5 ms; n=8, p< 0.05). Mice were challenged with subcutaneous isoproterenol (ISO, 25 mg/kg, Fig. 6C&D). ISO injection increased heart rates, but fragmented QRS complexes remained in Slc26a6^−/− mice. Summary data are shown in Fig. 6E&F. Of note, even though the heart rate measured by echocardiography was significantly reduced in unanesthesized Slc26a6^−/− compared to WT mice (Fig. 4C), only the PR intervals were significantly prolonged in the anesthetized Slc26a6^−/− mice during ECG recordings (Fig. 6E). These in vivo findings suggest critical roles of Slc26a6 in cardiac conduction systems that warrant additional future investigations.

Figure 6. ECG recordings from age-matched WT and Slc26a6^−/− mice reveal evidence of fragmented QRS complexes in the knockout mice.

A to D. Representative ECG traces. The insets showed the enlarged traces marked by the rectangular boxes demonstrating fragmented QRS complexes observed only in Slc26a6^−/− mice. E. ECG parameters in control condition and with isoproterenol (ISO) (F).

Molecular identification of human cardiac SLC26A6 isoforms

Previous studies suggest that Cl⁻/oxalate exchange of mouse Slc26a6 is electrogenic, while the human SLC26A6 mediates electroneutral Cl⁻/oxalate exchange.^{18, 19} To directly test whether there are indeed species differences, we performed RT-PCR to clone full-length SLC26A6 from human heart (Fig. 7).

Figure 7. Molecular identification and functional analysis of human cardiac SLC26A6 isoforms.

A. The cloning strategy. Two different forward primers, For-long and For-short, were designed, corresponding to the genomic DNA sequence in chromosome 3 (NC_000003.12), one on 5′-UTR of exon 1 and the other on the intron sequence between exon 1 and exon 2, to verify isoforms having distinct 5′-UTR and/or 5′-end of coding sequence. The exon structure of the cloned isoforms was shown below. B. RT–PCR amplification of the full-length cDNA of human cardiac SLC26A6 isoforms (indicated by an arrow) using two sets of primers. C. Patch-clamp recordings and fast solution exchange to activate two heterologously expressed human cardiac SLC26A6 isoforms that we have obtained. The outward current was activated by the solution exchange diagram shown in Fig.1B, and the inward current was activated by the solution exchange diagram shown in Fig.1G. Recordings on non-transfected CHO cells were shown at the bottom. Dotted lines represent zero current.

Six different transcript variants of human SLC26A6 have previously been described. During the cloning of human cardiac SLC26A6, two major transcript variants were repetitively obtained, suggesting the presence of alternatively spliced variants of SLC26A6 in human heart. We used a For-long and Rev primer set, as well as a For-short and Rev primer set (Fig. 7A); the PCR products were shown in Fig. 7B. 12 clones were randomly selected for each primer set, compared by restriction enzyme digestion, and full-length sequence confirmed by sequencing. Sequence analysis revealed that the two identified isoforms correspond to previously described transcript variant 2 (NM_134263.2) and transcript variant 4 (NM_001040454.1) for primer set 1 and 2, respectively. Variant 4 is 21 amino acids shorter in the N-terminus compared to transcript variant 2 (see protein sequence alignment in Supplementary Figure 3).

Electrogenic Cl⁻/oxalate and Cl⁻/HCO₃⁻ exchange activities mediated by human cardiac SLC26A6 isoforms

To test the function of the human cardiac SLC26A6, we expressed the human cardiac SLC26A6 isoforms in CHO cells and recorded the current generated by the anion exchanger, using patch-clamp coupled with a fast solution exchange technique, as described in Fig. 1B &G. Membrane potential was held at 0 mV. Fig. 7C shows the outward currents generated by electrogenic Cl⁻/oxalate and Cl⁻/HCO₃⁻ exchanges, and the inward currents generated by Cl⁻/HCO₃⁻ exchanges. Both variants identified in human hearts are functional electrogenic Cl⁻/oxalate and Cl⁻/HCO₃⁻ exchangers.

Discussion

Slc26a6 has been proposed to be the predominant Cl⁻/HCO₃⁻ exchanger and a specific Cl⁻/OH⁻ exchanger in the mouse heart, based on transcript levels.⁶ Our group has since identified four cardiac Slc26a6 isoforms in the mouse heart, and importantly, we demonstrated that mouse cardiac Slc26a6 mediates eletrogenic Cl⁻/HCO₃⁻ exchange activities in cardiac myocytes.⁷ We predict, based on its electrogenic activities, that Slc26a6 may participate not only in the regulation of pH_i, but also in cardiac excitability. To directly test this hypothesis, we took advantage of Slc26a6^−/− mice. Consistent with our prediction, ablation of Slc26a6 significantly shortens APD, resulting in reduced sarcomere shortening, CaT, and SR Ca²⁺ load, in addition to an elevated pH_i. Moreover, our in vivo studies demonstrate decreased fractional shortening and cardiac contractility, as well as sinus bradycardia in the knockout mice. For the first time, our results uncover new insights into the critical roles of cardiac electrogenic anion transporter—linking the regulation of not only pH_i, but also cardiac excitability and contractility.

Roles of bicarbonate in cardiac pH regulation

HCO₃⁻/CO₂ buffer is the major component of the cellular pH buffering system. Bicarbonate is the byproduct of mitochondrial respiration; therefore, its concentration is highly affected by physiological and pathological conditions. HCO₃⁻ is critical for pH regulation, acid/base secretion, and body fluid secretion.^{20, 21} Although CO₂ can diffuse across the plasma membrane, HCO₃⁻ transport across the plasma membrane requires facilitation by HCO₃⁻ transporters and anion channels. HCO₃⁻ transporters are encoded by 14 genes belonging to SLC4A and SLC26A gene families.²⁰

The heart is an organ with a high metabolic rate, due to its continuous mechanical activities. Therefore, HCO₃⁻ is a critical anion in the regulation of cardiac function^{1, 2}. Unlike Na⁺, K⁺ and Ca²⁺ ions, which directly affect the excitability and contractility of the heart, HCO₃⁻ participates in the regulation of cardiac function through its coupled transport with other cations and anions, as well as its direct control of cardiac pH. Therefore, it is a unique anion in the heart with dynamic regulation and a wide range of function. HCO₃⁻ participates in acid extruding and acid loading processes in cardiomyocytes, which are coupled with Na⁺ and Cl⁻ transport, respectively.¹ Our studies provided strong evidence to support the role of Slc26a6 as an acid loader contributing to the acid loading process in cardiomyocytes. Since cardiac Slc26a6 mediates Cl⁻/HCO₃⁻ and Cl⁻/OH⁻ exchange, our study also demonstrated the critical roles of Cl⁻ in regulation of cardiac function through Slc26a6.

Electrogenic properties of human SLC26A6

In contrast to our current findings, previous studies suggest that human SLC26A6 may not be electrogenic.^{18, 19} These studies, however, used an oocyte expression system and indirect measurement of changes in reversal potentials, wherein intracellular ion concentrations were unknown and could not be precisely controlled during the recordings. In contrast, we used patch-clamp recordings with fast solution exchange to precisely set the ion gradient across the plasma membrane and monitor the dynamic changes of the currents. We observed electrogenic Cl⁻/HCO₃⁻ and Cl⁻/oxalate exchange activities of both mouse and human cardiac SLC26A6 (Fig. 7).

Critical roles of electrogenic transporters in cardiac excitability

Electrogenic transporters contribute significantly to cardiac excitability. The well described Na⁺/K⁺ pump contributes to the generation of resting membrane potentials and repolarization of cardiac action potentials.²² Another electrogenic transporter, Na⁺/Ca²⁺ exchanger (NCX), is not only important in intracellular Ca²⁺ homeostasis and cardiac contractility, but also in the regulation of cardiac excitability and arrhythmogenesis. Similar to Slc26a6^−/−, cardiac-specific NCX knockout mice show an abbreviated APD in cardiomyocytes.^23–25 Additionally, the electrogenic Na⁺/HCO₃⁻ cotransporter was reported to modulate resting membrane potentials and cardiac APs.²⁶

Slc26a6 mediates eletrogenic Cl⁻/HCO₃⁻ exchange in the heart, thus regulating cardiac APD, as we demonstrate for the first time in this study. Based on the predicted stoichiometry of Slc26a6, at least two HCO₃⁻ are exchanged for one Cl⁻ ion, with estimated reversal potential of the exchange of ~+36 mV. Under physiological condition, the activation of Slc26a6 generates inward currents, which depolarize the membrane potential.⁷ Consistently, Slc26a6^−/− cardiomyocytes show shortened cardiac APs.

Functional roles of Slc26a6 in acid loading process in cardiomyocytes

The elevated pH_i in Slc26a6^−/− cardiomyocytes supports the essential role of Slc26a6 in the acid loading process.^{1, 2, 6, 7} Under physiological conditions, Slc26a6 mediates Cl⁻/HCO₃⁻ or Cl⁻/OH⁻ exchange, maintaining a relatively lower pH_i and higher intracellular Cl⁻ concentration in cardiomyocytes. Here, we directly monitored the acidification process during the application and wash out of acetate. Slc26a6^−/− cardiomyocytes show delayed acidification, supporting the role of Slc26a6 in cellular acid loading (Fig. 3).

In addition to Slc26a6, it has been reported that Slc4a families are expressed in cardiomyocytes, including AE1 (Slc4a1), AE2 (Slc4a2), and AE3 (Slc4a3), which are acid loaders. Other Slc4a members are mostly Na⁺/HCO₃⁻ cotransporters, mediating acid extrusion. However, the expression levels of AE1, AE2, AE3, as well as Slc26a3 are relatively low compared to Slc26a6.⁶ Nonetheless, it is possible that there is compensation from AE and other Slc26a families during the pH_i recovery process. The dominant molecule for acid loading is Slc26a6 because its expression level is nearly one hundred fold higher than that of AEs and Slc26a3.

Decreased cardiac contractility in Slc26a6^−/− mice

The shortened cardiac APs are predicted to decrease Ca²⁺ influx through L-type Ca²⁺ channels, leading to reduced CaT and sarcomere fractional shortening in Slc26a6^−/− cardiomyocytes. Additionally, the elevated pH_i may contribute to changes in CaT and contractility. Indeed, pH_i is a critical regulator of cardiac ion channels and transporters, as well as other membrane and contractile proteins, and, therefore, exerts significant influences on cardiac Ca²⁺ signaling, contractility, and excitability.^{1, 2} Effects of acidosis on cardiac function have been extensively investigated due to its pathological significance during ischemia-reperfusion.^27–34 Cellular acidification reduces CaT and contraction in cardiomyocytes.^{1, 27, 29, 31, 35} Our results showed that CaT was initially enhanced and later inhibited by cellular acidification, induced by wash in of acetate (Fig. 3E) when Na⁺/H⁺ exchanger was inhibited, similar to that observed by Vaughan-Jones et al.¹ The reduced sarcomere shortening observed in our study from acidification (Fig. 3E), agrees with previously reported findings.

On the other hand, effects of intracellular alkalization on cardiac contractility and CaT are less understood. Our findings in Fig. 3E demonstrate that acid loading had a relatively small influence on CaT compared to cell shortening in both WT and Slc26a6^−/− myocytes, however, alkalization during wash out dramatically decreased CaT and cell shortening. The findings suggest that lower pH_i had a minor effect on CaT, but may affect the Ca²⁺ affinity of troponin C.¹ Taken together, the impairment of cardiac contractility in Slc26a6^−/− mice is likely due to multiple mechanisms, including alterations in pH_i, in addition to reduced APDs, CaT, and SR Ca²⁺ load.

Future studies

One previous study using a different strain of Slc26a6^−/− mice reported normal body weight, heart rate, and blood pressure in Slc26a6^−/− mice.³⁶ Our studies demonstrate comparable heart/body weight ratio between WT and Slc26a6^−/− mice, however, there is evidence of sinus bradycardia and fragmented QRS complex in Slc26a6^−/− mice with prolonged PR interval. Fragmented QRS complexes have been shown to be common in patients post myocardial infarction; they may represent a marker for cardiovascular diseases^37–39, and predict arrhythmic events and mortality in patients with cardiomyopathy.^{40, 41} Fragmented QRS was also reported to be associated with prognosis in patients with Brugada syndrome, supporting the association of fragmented QRS and arrhythmia substrate.⁴² Slc26a6^−/− mice showed sinus bradycardia and fragmented QRS, however, there is no significant cardiac fibrosis in Slc26a6^−/− mice. Our findings of fragmented QRS complexes and prolonged PR interval in Slc26a6^−/− mice support the critical role of Slc26a6 not only in ventricular myocytes, but also in the cardiac conduction system. Sinus bradycardia in Slc26a6^−/− mice further suggests functional roles of Slc26a6 in pacemaking cells.

In addition, the elevated pH_i may result in changes in gene expression profiles, as well as the function of other membrane and cytosolic proteins. For example, pH_i may affect APDs of rabbit and guinea pig ventricular myocytes by H⁺-induced changes in late Ca²⁺ entry through the L-type Ca²⁺ channel.⁴³ Therefore, we expect a spectrum of cellular proteins which need to be further evaluated. However, these extensive investigations will be performed in our future experiments.

The global knockout of Slc26a6 was reported to affect the kidney and duodenum epithelial transport³⁶ and induce Ca²⁺ oxalate urolithiasis.⁸ The alteration in metabolic profiles in Slc26a6^−/− may affect the observed in vivo cardiac function. Therefore, we took advantage of not only in vivo physiological measurement, but also in vitro analyses in our studies to circumvent these possible confounding factors. Indeed, our in vivo findings are consistent with in vitro analyses.

In conclusion, the in vivo and in vitro studies unravel novel mechanistic insights into the newly described and unique cardiac electrogenic Cl⁻/HCO₃⁻ transporter in ventricular myocytes. Activities of Slc26a6 contribute not only to the regulation of pH_i, but also cardiac AP and contractility. Ablation of Slc26a6 shortens the APDs, impairs cardiac function, and results in fragmented QRS complexes and elevated pH_i. Additional studies are required to further understand the functional roles of Slc26a6 not only in ventricular myocytes, but also in pacemaking cells and the cardiac conduction system.

WHAT IS KNOWN?

• Intracellular pH (pH_i) is critical to cardiac excitation and contraction. Uncompensated changes in pH_i impair cardiac function and trigger arrhythmias, however, cardiac pH_i regulation mechanisms are not completely understood.

• Our previous studies identified several isoforms of a solute carrier, Slc26a6, to be highly expressed in mouse cardiomyocytes. We demonstrated that Slc26a6 mediates electrogenic Cl⁻/HCO₃⁻ exchange activities in cardiomyocytes supporting the role of Slc26a6 in the regulation of pH_i and cardiac function.

WHAT THE STUDY ADDS?

• Ablation of Slc26a6 resulted in action potential (AP) shortening, reduced Ca²⁺ transient and sarcoplasmic reticulum Ca²⁺ load, decreased sarcomere shortening, and elevated pH_i in mouse cardiomyocytes.

• At in vivo level, Slc26a6 knockout mice showed reduced fractional shortening and cardiac contractility, as well as fragmented QRS complexes, supporting the critical roles of Slc26a6 in the regulation of cardiac excitability and contractility.

• For the first time, we identified human cardiac SLC26A6 isoforms and demonstrated their electrogenic Cl⁻/HCO₃⁻ exchange activities.