Heart failure induces changes in acid‐ ... - Wiley Online Library

4575

J Physiol 593.20 (2015) pp 4575–4587

Heart failure induces changes in acid-sensing ion channels in sensory neurons innervating skeletal muscle David D. Gibbons1,2 , William J. Kutschke1 , Robert M. Weiss1 and Christopher J. Benson1,2 1 2

Department of Internal Medicine, Carver College of Medicine, University of Iowa, Iowa City, IA 52242, USA The Department of Veterans Medical Center, Iowa City, IA 52242, USA

Key points

r Heart failure is characterized by an elevated sympathetic state and exercise intolerance, which is partially driven by exaggerated autonomic reflexes triggered by skeletal muscle afferents.

The Journal of Physiology

r Acid-sensing ion channels (ASICs) are highly expressed in skeletal muscle afferents and contribute to exercise mediated reflexes.

r Here we show that ASIC currents recorded from isolated skeletal muscle sensory neurons r

display diminished pH sensitivity, altered desensitization kinetics, and faster recovery from desensitization in a mouse model of heart failure. These results indicate ASICs in muscle afferents are altered in heart failure, and may contribute to the associated sympathoexcitation and exercise intolerance.

Abstract Heart failure is associated with diminished exercise capacity, which is driven, in part, by alterations in exercise-induced autonomic reflexes triggered by skeletal muscle sensory neurons (afferents). These overactive reflexes may also contribute to the chronic state of sympathetic excitation, which is a major contributor to the morbidity and mortality of heart failure. Acid-sensing ion channels (ASICs) are highly expressed in muscle afferents where they sense metabolic changes associated with ischaemia and exercise, and contribute to the metabolic component of these reflexes. Therefore, we tested if ASICs within muscle afferents are altered in heart failure. We used whole-cell patch clamp to study the electrophysiological properties of acid-evoked currents in isolated, labelled muscle afferent neurons from control and heart failure (induced by myocardial infarction) mice. We found that the percentage of muscle afferents that displayed ASIC-like currents, the current amplitudes, and the pH dose–response relationships were not altered in mice with heart failure. On the other hand, the biophysical properties of ASIC-like currents were significantly different in a subpopulation of cells (40%) from heart failure mice. This population displayed diminished pH sensitivity, altered desensitization kinetics, and very fast recovery from desensitization. These unique properties define these channels within this subpopulation of muscle afferents as being heteromeric channels composed of ASIC2a and -3 subunits. Heart failure induced a shift in the subunit composition of ASICs within muscle afferents, which significantly altered their pH sensing characteristics. These results might, in part, contribute to the changes in exercise-mediated reflexes that are associated with heart failure. (Received 2 April 2015; accepted after revision 13 August 2015; first published online 28 August 2015) Corresponding author C. J. Benson: Department of Internal Medicine, Roy J. and Lucille A. Carver College of Medicine, University of Iowa, Iowa City, IA 52242, USA. Email: [email protected] Abbreviations ASICs, acid-sensing ion channels; BP, blood pressure; DiI, 1,1-dioctadecyl-3,3,3,3 tetramethylindocarbocyanine perchlorate; DRG, dorsal root ganglion; EPR, exercise pressor reflex; HF, heart failure; HR, heart rate.

C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

DOI: 10.1113/JP270690

4576

D. D. Gibbons and others

Introduction Heart failure is one of the most common causes of morbidity and mortality in the industrialized world (Go et al. 2013). While the most obvious manifestation of heart failure is a diminished circulatory function, it is well established that systemic neurohormonal perturbations, including increases in sympathetic nerve activity, play a major role in the pathophysiology of the disease process, as well as contributing to the overall morbidity and mortality of heart failure (Francis & Cohn, 1986). In addition, patients with heart failure are often incapacitated by their inability to exert themselves, with characteristic symptoms including fatigue and shortness of breath with exercise. Remarkably, this exercise intolerance has shown little correlation with measurements of cardiac pump or circulatory function (Franciosa et al. 1981; Sullivan & Hawthorne, 1995; Wilson et al. 1995). Instead, data suggest that the inability of patients with heart failure to exercise is related to the degree of autonomic dysfunction, and particularly to an elevation of sympathetic tone (Negrao et al. 2001; Notarius et al. 2001; Ponikowski et al. 2001). While the mechanisms that sustain this ‘dysautonomia’ are largely unknown, it has been suggested that it may be due, in part, to an overactivity of reflexes triggered by sensory nerves that innervate skeletal muscle (Piepoli et al. 1999; Sinoway & Li, 2005; Smith et al. 2006; Garry, 2011). These sensory neurons (muscle afferents) sense mechanical and metabolic changes associated with exercise and evoke reflexive increases in blood pressure (BP), heart rate (HR), and ventilation (collectively termed the ‘exercise pressor reflex’ (EPR)) (Alam & Smirk, 1937; McCloskey & Mitchell, 1972; Kaufman & Hayes, 2002). In patients with heart failure, activation of the EPR has been shown to cause exaggerated cardio-respiratory and sympathetic nerve responses during exercise (Piepoli et al. 1996; Notarius et al. 2001; Middlekauff et al. 2004). Similar results have been demonstrated in animal models of heart failure (Hammond et al. 2000; Li et al. 2004b; Smith et al. 2005; Wang et al. 2010). However, the underlying mechanisms of the exaggerated EPR in heart failure are unknown. For example, it is unclear whether this is the result of mechano- vs. metaboreflex overactivity (Middlekauff & Sinoway, 2007; Piepoli & Coats, 2007). More fundamentally, little is understood about these changes at the molecular and cellular level. Accumulating evidence suggests that acid-sensing ion channels (ASICs) are important metabolic sensors within skeletal muscle afferents. ASICs are H+ -gated channels of the degenerin (DEG)/epithelial sodium channel (ENaC) family, primarily expressed in the central nervous system and peripheral sensory neurons. In rodents, ASICs consist of six subunits (ASIC1a, -1b, -2a, -2b, -3 and -4) translated from four genes (ASIC1 and -2 have alternative splice

J Physiol 593.20

transcripts) (Wemmie et al. 2013). Functional channels are composed of three subunits; individual subunits form homotrimers, and two or more subunits can assemble to form heterotrimers, and each combination of subunits forms channels with unique biophysical properties (Benson et al. 2002; Hesselager et al. 2004; Jasti et al. 2007). For example, each ASIC channel has a different pH dose–response relationship, displays different activation and desensitization kinetics, and each is differentially modulated. Several observations indicate that ASICs play a particularly important role in skeletal muscle afferents. ASICs are expressed in muscle afferents at higher levels than in cutaneous afferents (Molliver et al. 2005). They are activated in the narrow ranges of extracellular pH (pH 7.2–6.8) that occur during muscle ischaemia and exercise, and are potentiated by other metabolites that are released from exercising muscle, including lactic acid and ATP (Immke & McCleskey, 2001; Cushman et al. 2007; Light et al. 2008). We recently found ASIC-like currents in a high percentage of isolated muscle afferents, and by genetic and pharmacological interventions we determined that the ASIC subunit composition in muscle afferents consists primarily of ASIC1a and -3 subunits, with a lesser contribution from ASIC2 subunits (Gautam & Benson, 2013). ASICs are required for the development of normal muscle pain. Either genetic or pharmacological inhibition of ASICs attenuates hyperalgesia in mouse models of muscle pain (Price et al. 2001; Sluka et al. 2003, 2007; Walder et al. 2011). Lastly, ASICs contribute to the EPR (Li et al. 2004a; Gao et al. 2006; Hayes et al. 2007, 2008). Hayes et al. showed that a selective ASIC antagonist, A-317567, injected into cat hindlimb muscles inhibited the pressor response to lactic acid injection by 75%, and to static muscle contraction by 60% (Hayes et al. 2008). Since ASICs contribute to the EPR during normal physiological conditions, and the EPR response is significantly altered in heart failure, we tested if ASIC function in skeletal muscle afferents is altered in a mouse model of heart failure. Methods Generation of heart failure mice

All animal procedures were followed in accordance with, and were approved by, the Institutional Animal Care and Use Committee of the University of Iowa. C57BL6/J mice (8 weeks of age) obtained from Harlan Laboratories (Indianapolis, IN, USA) underwent myocardial infarctions as previously described (Zhang et al. 2005). All surgical procedures were performed observing aseptic techniques. Briefly, animals were anaesthetized with a ketamine/xylazine mixture (0.1 ml (20 g body weight)−1 , 17.5 mg ml−1 ketamine/2.5 mg ml−1 xylazine, C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

J Physiol 593.20

Heart failure alters ASICs in muscle afferents

I.P.) and then prepared and ventilated. Hearts were accessed via a thoracotomy and infarcted by placing a permanent ligature on the mid-left anterior descending coronary artery. The ribcages were then closed with the air evacuated and the incision closed. Mice were then monitored continuously until awake and analgesia was administered immediately and daily for 2 days postoperatively. Mice were monitored daily for 5 days and then twice weekly until the muscle injection surgery. At 9 weeks postinfarction, conscious mice underwent transthoracic echocardiography to determine cardiac size and function, as previously described (El Accaoui et al. 2014). Cardiac output was calculated from the stroke volume and HR.

Maximal treadmill exercise testing

Control and heart failure mice (20 weeks old) were acclimated to a 6-track rodent treadmill (Columbus Instruments, Columbus, OH, USA) for 3 days prior to maximal exercise testing by placing them on a non-moving track for 15 min followed by 15 min at a velocity of 3.5 m min−1 . A shock grid at the back of the treadmill delivered a mild painful stimulus to enforce running. The exercise–stress test protocol consisted of stepwise increases in velocity and/or incline at 3 min intervals. If mice attained 20.8 m min−1 and 15 deg incline, exercise was continued at these parameters until they were exhausted. The point of dropout was defined by failure to sustain performance at a given workload level, despite receiving a shock. Workload was defined as the sum of kinetic (Ek = (mv2 )/2) and potential (Ep = mgvt(sinϕ)) energy of the mice on the treadmill, where m is animal mass, v is running velocity, g is acceleration due to gravity, t is elapsed time at a given protocol level, and ϕ is the angle of incline. Total workload is the sum of Ek (kinetic) and Ep (potential). The test was administered in a blinded fashion (Koganti et al. 2015). Mice were weighed 24 h after the exhaustion protocol was completed in order to allow them to rehydrate. Labelling of muscle sensory neurons

Sensory neurons innervating skeletal muscle were fluorescently labelled (at 8 weeks following myocardial infarction) using the retrograde tracer 1,1-dioctadecyl3,3,3,3 tetramethylindocarbocyanine perchlorate (DiI; 17 mg ml−1 dissolved in 20% v/v ethanol and suspended in 80% v/v sterile saline; Invitrogen, Eugene, OR, USA). Mice were anaesthetized with 2–5% inhaled isoflurane, with adjustments made as needed to ensure the appropriate depth of anaesthesia as determined by paw pinch. A small incision was made in the skin over both gastrocnemius muscles, and 10 μl of DiI solution was injected into the


4577

muscles as described previously (Gautam et al. 2010). After injection, saline-soaked sterile gauze was placed on the open incision for 10 min before the skin was sutured to prevent the dye from leaking to the overlying skin. The mice were allowed to recover from surgery and were monitored daily for 5 days and then twice a week until termination. Mice were killed by 5% inhaled isoflurane anaesthesia induction followed by decapitation. Culture of dorsal root ganglion (DRG) neurons

At 2 weeks post DiI injection the mice were killed and the DRGs (L4–L6) were collected and dissociated as described previously (Benson et al. 2002). DRGs were successively treated with papain and collagenase–dispase and then gently triturated to isolate neurons. Neuron suspensions were plated on 35 mm Petri dishes coated with poly-D-lysine and laminin. Cells were cultured in F12 medium supplemented with 10% heat-inactivated serum, penicillin–streptomycin, and 50 ng ml−1 nerve growth factor. Muscle afferents were identified by fluorescence microscopy (see Fig. 1A) and were studied 18–48 h after plating. Cell diameters were measured in all studied cells. Electrophysiology

Whole-cell patch-clamp recordings (at −70 mV) of DiI-labelled muscle sensory neurons were performed at room temperature with an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA) and were acquired with PATCHMASTER 2.71 (HEKA Electronics, Lambrecht, Germany) and analysed with FITMASTER 2.71 (HEKA Electronics) software. Currents were filtered at 1 kHz and sampled at 2 kHz. Micropipettes (3–5 M) were filled with internal solution: 100 mM KCl, 10 mM EGTA, 40 mM Hepes, and 5 mM MgCl2 , pH 7.4 with KOH. External solution contained 120 mM NaCl, 5 mM KCl, 1 mM MgCl2 , 2 mM CaCl2 , 10 mM Hepes, and 10 mM 2-(N-morpholino)ethane sulfonic acid. The pH was adjusted with tetramethylammonium hydroxide, and osmolarity was adjusted with tetramethylammonium chloride. Rapid extracellular solution exchanges were made within 20 ms by using a computer-driven solenoid valve system (Benson et al. 1999). The kinetics of desensitization and recovery from desensitization data were fitted with single exponential equations, and time constants (τ) are reported. Statistical analysis

All data are presented as means ± SEM. Statistical significance was assessed using an unpaired Student’s t test between control and heart failure at individual pH values.

4578


Comparisons between multiple groups were assessed using either one-way (Dunnett’s post hoc test) or two-way (Sidak’s post hoc test) ANOVAs. Correlation comparisons were performed using the Pearson correlation calculation, with coefficients of determination (R2 ) reported.

Results Mice with heart failure have reduced maximal exercise capacity

To generate heart failure in mice, we created myocardial infarctions by coronary ligation. Compared to control mice, heart failure mice displayed left ventricular hypertrophy (Fig. 1A and B), chamber dilatation (Fig. 1C and D), and reduced ejection fraction (Fig. 1E). However, resting stroke volume and cardiac output were not significantly different (Fig. 1F–H). We then tested if mice with heart failure have altered exercise capacity. Figure 2 shows that compared to control mice, mice with heart failure displayed significantly diminished endurance running capacity measured by treadmill testing.

A

C

F

J Physiol 593.20

Heart failure does not alter the population of muscle afferents that express ASIC-like currents

To test if ASICs are altered in skeletal muscle afferents in mice with heart failure, we first studied pH 5-evoked currents recorded from isolated muscle afferents that had been labelled by injection of a retrograde tracer dye (DiI) into the gastrocnemius muscles (Fig. 3A). Figure 3B demonstrates typical ASIC-like currents from four separate muscle afferent neurons in dissociated culture: the currents rapidly activated and then desensitized in the continued presence of pH 5 solution (transient current), followed by a variable sustained current. We previously demonstrated that these transient currents in mouse muscle afferents are blocked by the ASIC antagonist amiloride and are absent in mice that lack ASICs (Gautam & Benson, 2013). Other muscle afferents responded to pH 5 with only a sustained current, which is not typical of ASIC channels (Hesselager et al. 2004), or showed no measurable current (Fig. 3C). We found that the percentage of muscle afferents that had ASIC-like currents was not different in heart failure mice compared to those from control mice (Fig. 3D), nor was the percentage of

B

D

G

E

H

Figure 1. Coronary ligation induces cardiac structural alterations A, echocardiographic short axis images of the left ventricle during systole and diastole of control mice or mice that had undergone coronary ligation 10 weeks previously. Epicardium (Epi.), endocardium (Endo.) and infarct area are identified. B–H, left ventricular (LV) mass, end-diastolic volume (LVEDV), end-systolic volume (LVESV), ejection fraction, stroke volume, heart rate and cardiac output (CO) in control and heart failure (HF) mice (∗ P < 0.01 compared to control; n = 11 for control and 21 for HF mice). C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society


J Physiol 593.20

muscle afferents that displayed only sustained currents (Fig. 3E). ASICs are expressed in small- and medium-sized sensory neurons that primarily respond to noxious stimuli, and are also expressed in larger neurons that are purported to be low threshold mechanoreceptors (Lingueglia, 2007). Figure 3F shows histograms of the cell diameters of ASIC-like and non-ASIC-like responsive muscle afferents from control and heart failure mice. We found that ASIC-like responsive cells had larger diameters than non-ASIC-like cells (35.0 ± 0.4 μm ASIC-like, 31.36 ± 0.5 μm non-ASIC-like, P < 0.01). However, the cell diameters of the ASIC-like and non-ASIC-like responsive cells were not different between control and heart failure mice (Fig. 3F). In summary, these data indicate that heart failure did not alter the population of muscle afferents that express ASIC-like currents, nor did it alter the population of muscle afferents that express other pH-sensitive ion channels.

ASIC-like current properties are altered in mice with heart failure

We next tested if heart failure altered the biophysical properties of ASIC channels in muscle afferents. Figure 4A shows typical currents evoked by acidic solutions of increasing H+ concentration in a cell with ASIC-like currents. The peak current amplitudes evoked at each pH tested were not different between control and heart failure animals (Fig. 4B). By normalizing current amplitudes to the pH 5-evoked peak amplitude, we found that the pH dose–response of activation of muscle afferents from heart failure mice was not different compared to those from control mice (Fig. 4C). Additionally, the sustained current amplitudes evoked at pH 5 were not

A

altered in muscle afferents from heart failure compared to control animals (Fig. 4D). Notably, the sustained currents were significantly larger in cells with ASIC-like currents (0.47 ± 0.02 nA, n = 108) compared to those with non-ASIC-like currents (0.38 ± 0.03 nA, n = 69; P < 0.05; Fig. 4D). By fitting the desensitization phase of the currents to a single exponential and recording the time constants of desensitization (τ), we found that ASIC-like currents from muscle afferents from both control and heart failure mice desensitize very fast (Fig. 5). Given our previous data, these fast kinetics indicate that the ASIC channels in muscle afferents from both groups of mice are heteromeric channels; only heteromeric channels that contain ASIC3 as one of the subunits possess such fast desensitization kinetics (τ < 0.2 s) for currents evoked by pH 6 (Benson et al. 2002; Hattori et al. 2009). At a couple of pH values tested, the desensitization kinetics were altered in heart failure mice. pH 6.8-evoked currents in muscle afferents from heart failure mice desensitized at a significantly slower rate than those from control mice (Fig. 5A and B). Conversely, pH 6-evoked currents in afferents from heart failure mice desensitized at a significantly faster rate than the control cells (Fig. 5A and B). Lastly, we measured the rate of recovery from desensitization. After acid-evoked desensitization, ASIC channels require exposure to a more alkaline pH for some time period before they can again be activated by acidic solution. We found that ASIC-like currents recovered from desensitization significantly faster in afferents from heart failure mice compared to control (Fig. 6A–C). Fitting the recovery data to a single exponential curve and plotting the time constants (τ) of individual cells reveals a population of muscle afferents from heart failure mice with very fast recovery from desensitization (Fig. 6D and E). Given the variability of the current amplitudes between cells, we

B

Figure 2. Mice with heart failure have reduced maximal exercise capacity A, illustration of the treadmill exercise protocol featuring step-wise increases in speed and/or inclination at 3 min intervals. B, maximal exercise workload in control (n = 11) and heart failure mice (n = 8; ∗ P = 0.01).


4579

4580


tested if the amplitude of the pH 6-evoked current might correlate with the recovery kinetics (τ), and found no correlation for the control (R2 = 0.18, P = 0.08) or heart failure (R2 = 0.01, P = 0.55) groups.

Heart failure causes a shift in ASIC subunit composition in a subset of muscle afferents

The altered desensitization kinetics and the rapid recovery from desensitization of ASIC channels within a population of muscle afferents in the heart failure mice (13 of 32 cells (41%) with ASIC-like currents had a recovery time constant (τ) 0.12 s, whereas only 3 of 18 cells (17%) had a τ 0.12 s in the control mice), suggested to us that the ASIC subunit composition was altered in a subpopulation of muscle afferents in heart failure mice. In particular, we recognized from our previous work that this fast recovery from desensitization is a unique property of ASIC2a/3 heteromultimeric channels; only ASIC2a/3 channels have

J Physiol 593.20

a recovery rate (τ) 0.12 s, and in fact all other ASIC homo- and heteromeric channel combinations have rates of recovery that are at least three times slower (Benson et al. 2002; Hattori et al. 2009; Gautam & Benson, 2013). To further test this hypothesis, we measured other properties of this population of ‘fast recovery’ (τ 0.12 s) muscle afferents and compared them to the known properties of heterologously expressed ASIC2a/3 heteromultimeric channels. Compared to our data from control mice, the pH dose–response of activation of the ‘fast recovery’ group was shifted to the right (Fig. 7A), indicating that these channels were less pH sensitive. Figure 7B shows the Hill fit curve of the ‘fast recovery’ group superimposed on the fit of our previously published data for heterologously expressed ASIC2a/3 heteromeric channels (Hattori et al. 2009). In addition, we compared these results to our previous data from muscle afferents from ASIC1a−/− mice (Gautam & Benson, 2013), where we concluded that the resultant current was consistent with ASIC2a/3 heteromeric channels. These

Figure 3. Skeletal muscle afferent responses to application of acidic solution A, corresponding phase (left panel) and fluorescence (right panel) micrographs of a labelled skeletal muscle afferent in primary dissociated culture of DRG neurons 2 weeks after injection of fluorescent tracer dye (DiI) into the gastrocnemius muscles. B, typical ASIC-like currents recorded from 4 separate labelled muscle afferents evoked by stepping to a pH 5 solution from a control pH 7.4 solution. C, representative pH 5-evoked responses of 3 muscle afferents that did not have ASIC-like currents. D, the percentage of muscle afferents with ASIC-like currents in response to pH 5 (transient current >50 pA) were calculated for each animal and then the mean was determined (46.98 ± 0.04% control, n = 8 mice, 17–26 cells per mouse; 49.05 ± 0.04% HF, n = 8 mice, 13–21 cells per mouse). E, the percentage of cells with non-ASIC-like cells with sustained currents (>50 pA) was calculated for each animal (37.05 ± 0.06% control; n = 8 mice, 8–16 cells per mouse; 46.51 ± 0.06% HF; n = 8 mice, 6–14 cells per mouse). F, diameter–frequency distribution of ASIC-like responsive and non-responsive muscle afferents from control and heart failure mice. C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

J Physiol 593.20


data demonstrate that the pH sensitivity of the ‘fast recovery’ muscle afferents from heart failure mice is consistent with that of ASIC2a/3 heteromers. Next we measured the desensitization rates of acid-evoked currents from the heart failure ‘fast recovery’ muscle afferent group and found that currents evoked at several pH values desensitized significantly slower than those from control mice (Fig. 7C). These results were very similar to the published desensitization rates of ASIC2a/3 heteromers and our previous data from muscle afferents from ASIC1a−/− mice (Fig. 7C) (Hattori et al. 2009; Gautam & Benson, 2013). In the remaining 60% of muscle afferents from heart failure mice that displayed ASIC-like currents with ‘slow recovery’, the properties of ASIC-like currents were remarkably similar to those from control mice. The rates of recovery from desensitization were the same (τ = 0.49 ± 0.04 s for the ‘slow recovery’ afferents from heart failure mice vs. τ = 0.52 ± 0.09 s for control), as were the pH dose–response relationships (Fig. 7A), and the kinetics of desensitization (Fig. 7C). In summary, these results suggest heart failure induced a population of muscle afferents to undergo a phenotypic switch in their expression of ASICs: from channels composed of ASIC1a/2a/3 and ASIC1a/3 to ASIC2a/3 heteromeric channels.

4581

Discussion ASICs are highly expressed in sensory nerves that innervate skeletal muscle, where they serve as mediators of pain and contribute to exercise-mediated reflexes. Previous work has shown that ASICs are required for the metabolic component of the EPR (Hayes et al. 2008; McCord et al. 2009). In the setting of heart failure, the EPR is overactive. This exaggeration of the EPR contributes to sympathoexcitation, which is a major contributor to the disease pathogenesis and exercise intolerance of heart failure patients. Thus, we tested if ASIC function is altered in skeletal muscle afferents in mice with heart failure. Heart failure did not induce a dramatic up- or down-regulation in ASIC expression in skeletal muscle afferents; the percentages of cells that displayed ASIC-like currents and current amplitudes were unchanged. However, careful analysis of the biophysical properties of the ASIC-like currents showed they were altered in a subpopulation of cells. In 40% of muscle afferents from mice with heart failure that displayed ASIC-like currents, the currents displayed very fast recovery from desensitization (τ 0.12) – a characteristic seen in very few cells from control mice. In fact, in our previous study characterizing the composition of ASIC channels

A pH 7.0

pH 6.8

pH 6.5

pH 6.0

pH 5.0

pH 4.0

0.5 nA 1 sec

3

Control

1

HF 7

6

5

4

pH

D

HF 0.4 0.2 0.0 ASIC-like Cells


0.8 Control

0.4

HF

0.0 7

6

5 pH

Control

0.6

Peak l/l pH 5

1.2

2

0

pH 5 lSustained(nA)

Figure 4. ASIC-like current amplitudes and pH sensitivity are not altered in skeletal muscle afferents in heart failure mice A, representative currents evoked by the indicated pH solutions in a labelled skeletal muscle afferent. B, mean peak amplitudes of pH-evoked currents recorded from muscle afferents from control and heart failure mice (n = 21–56). C, data from B was normalized to the pH 5 current amplitude for each cell to compare pH dose responses. Lines are fits of the Hill equation of the means (control pH50 = 6.01; HF pH50 = 5.95). D, mean sustained current amplitudes evoked by pH 5 solution in muscle afferents that showed ASIC-like and non-ASIC-like currents from control and heart failure animals (P < 0.05 between ASIC-like and non ASIC-like cells, n = 34–55).

C Peak Current (nA)

B

Non ASIC-like Cells

4

4582


within muscle afferents we found no ASIC-like currents with such fast recovery kinetics; all cells recovered with a τ > 0.33 s (Gautam & Benson, 2013). Moreover, these afferents with ‘fast recovery’ kinetics were less pH sensitive (Fig. 7A), and their desensitization kinetics differed significantly from those of muscle afferents from control mice (Fig. 7C). The properties of the ASIC-like currents from the remaining 60% of muscle afferents from the heart failure mice – those with ‘slow recovery’ kinetics – were nearly identical to those from control mice. Together, these data demonstrate that heart failure induced an alteration in ASIC properties within a subpopulation of muscle afferents. Very fast recovery from desensitization is a unique characteristic of ASIC2a/3 heteromeric channels (Benson et al. 2002; Hattori et al. 2009). The other properties of this ‘fast recovery’ group of cells from heart failure mice also matched the properties of ASIC2a/3 heteromers. Thus, we conclude that heart failure induced a phenotypic switch in ASIC composition within a subpopulation of skeletal muscle afferents, from channels primarily composed of ASIC1a/2a/3 and -1a/3 heteromers, to channels consisting of ASIC2a/3 heteromers. While our data are consistent with a shift in ASIC subunit composition, we also cannot

A

pH 6.8

pH 6.0 Control HF

0.2 nA

0.5 nA

0.5 nA

1 sec

τ desensitization (ms)

B

0.4 nA 1 sec

500

Control HF

400 300 200 100 0

7.0

6.8

6.5 pH

6.0

5.0

Figure 5. The rates of desensitization of the ASIC-like currents were altered in heart failure mice at selective pH values A, superimposed currents evoked from muscle afferents from control and heart failure mice by the indicated pH solutions. Current amplitudes were normalized to demonstrate differences in the desensitization kinetics. B, mean time constants of desensitization (τ ) as measured from single exponential fits to the falling phase of the currents evoked by the indicated pH solutions (∗ P < 0.05 compared to control at the same pH; n = 27–54).

J Physiol 593.20

exclude the possibility that the channels were modulated by other post-translation mechanisms.

Dynamic expression of ASIC subunits

What are the mechanisms driving this change in the composition of ASICs in muscle afferents in the setting of heart failure? Increasing evidence indicates that ASIC expression can be dynamic in various disease conditions. Inflammatory mediators such as nerve growth factor (NGF), serotonin, interleukin-1 and bradykinin increase ASIC transcription in DRG neurons (Mamet et al. 2002). Accordingly, ASIC mRNA levels and/or ASIC-like currents from sensory neurons are altered in animal models of paw inflammation (Voilley et al. 2001), arthritis (Ikeuchi et al. 2008) and gastric ulcers (Sugiura et al. 2005). Regarding skeletal muscle, we previously found that injection of carrageenan to induce muscle inflammation caused an increase in ASIC-like currents in labelled muscle afferents. Muscle inflammation also caused a shift in the pH dose–response of activation to the right (less pH sensitive), similar to our findings here with heart failure (Gautam et al. 2010). These functional changes in ASIC currents after muscle inflammation correlated with 10-fold increases in ASIC2 and ASIC3 mRNA levels, but not ASIC1a or -1b (Walder et al. 2010). It is interesting that this differential regulation of ASIC subunits seen with muscle inflammation – a relative up-regulation in ASIC2 and -3 compared to ASIC1 mRNA – correlates well with our data here suggesting a functional increase in ASIC2a/3 heteromers within muscle afferents in mice with heart failure. Similar increases in ASIC3 have been observed in muscle afferents after femoral artery occlusion to induce limb ischaemia, and this increase in ASIC3 correlated with potentiated reflex increases in BP and sympathetic nerve activity in response to muscle injection of lactic acid (Liu et al. 2010). However, our findings here cannot be explained by a similar reduction in limb perfusion since stroke volume and cardiac output were preserved in our model. While the changes in ASIC channel composition that we observed here in muscle afferents in the setting of heart failure might also reflect a differential expression of ASIC subunits, it is also possible that these changes occurred due to altered protein translation, channel assembly, and/or channel trafficking to the cell surface. Altered expression of other sensory receptors within muscle afferents has been observed in the setting of heart failure. The transient receptor potential vanilloid 1 (TRPV1) is a channel that is activated by heat, protons and the pungent compound produced by pepper plants, capsaicin. In a rat model of heart failure, TRPV1 mRNA levels are reduced in the DRG corresponding to


J Physiol 593.20


diminished haemodynamic reflex responses to capsaicin injection (Smith et al. 2005). How might altered ASIC channel composition affect muscle afferent excitability?

One might assume that since ASIC-like current amplitudes were unaltered, neuronal excitability would not be significantly changed. However, the observed alterations in the biophysical properties of the channels can have profound effects on neuronal excitability. For example, the diminished pH sensitivity (shift in dose–response to the right) of a population of muscle afferents from the heart failure mice would have a marked effect on H+ and lactic acid sensing of muscle afferents. Such a shift (Fig. 7A) could render these afferents insensitive to the pH changes that occur within muscle during exercise, and might, in part, contribute to a diminished metaboreflex component of the EPR described in some settings of heart failure (Middlekauff et al. 2001; Smith et al. 2005; Wang et al. 2010; Garry, 2011; Xing et al. 2015). On the other hand, the altered kinetics of desensitization, and in particular the faster recovery from desensitization is likely to alter

Figure 6. Recovery from desensitization is faster in skeletal muscle afferents from heart failure mice A and B, overlay of current traces demonstrating recovery from desensitization of a control (A) and heart failure (B) muscle afferent. Current was desensitized with a 4 s application of pH 6 (only the first second is shown). Cells were then exposed to a pH 7.4 solution for varying time periods before they were then exposed to a second pH 6 application. Recovery is the percentage of the current evoked by the second pH 6 application compared with the first. C, the mean percentage recovery for each time period of pH 7.4 application for heart failure and control muscle afferents (n = 9–32). Lines are fits of single exponentials of the means. D, the time constants (τ ) of fits of the recovery data of individual cells and means ± SEM for heart failure and control muscle afferents (P < 0.05). E, distribution of the time constants (τ ) of fits of the recovery data of individual cells from heart failure and control mice. Dashed lines are fits of 4th order polynomial equations of the distributions. C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

4583

the capacity of muscle afferents to respond to fluctuating pH conditions. These changes in properties could cause the neurons to be more sensitive to pH changes, and would be consistent with reports describing an increased metaboreceptor sensitivity associated with heart failure, and that buffering the acidosis associated with exercising muscle blunts this hyperactive metaboreflex (Scott et al. 2002, 2003). The importance of desensitization kinetics to neuronal excitability is illustrated by the link of an inheritable form of epilepsy with mutant nicotinic acetylcholine receptors that have faster desensitization and slower recovery from desensitization (Kuryatov et al. 1997). Future work will be necessary to test the effect of these altered ASIC biophysical characteristics (pH sensitivity and desensitization kinetics, particularly at physiological/pathophysiological pH changes) on muscle afferent excitability.

Patho-/physiological implications of our findings

What is the physiological function of this population of muscle afferents in which we observed an alteration in ASICs? Muscle afferents that participate in the

4584


EPR are composed of type III (thinly myelinated) and type IV (unmyelinated) nerves that are principally responsive to mechanical and metabolic (chemical) stimuli, respectively. Previous work from the Kaufman group has demonstrated that ASICs primarily contribute to the metabolic component of the EPR (Hayes et al. 2008; McCord et al. 2009). Thus, it is likely that the cells that displayed ASIC-like currents are metaboreceptive afferents. Evidence suggests that there are subpopulations of metaboreceptive muscle afferents. For example, Light et al. found a population of muscle afferent neurons that were activated by low concentrations of protons, ATP and lactic acid, and another population that responded to higher concentrations of these metabolites (Light et al. 2008). Additionally, Xing et al. found that ASIC-like currents were larger in labelled muscle afferent neurons from slow-twitch oxidative muscle compared to fast-twitch glycolytic muscle (Xing et al. 2008). The gastrocnemius muscle contains both fast-twitch and slow-twitch fibres, and our labelling technique invariably labelled

J Physiol 593.20

neurons innervating both fibre types. In the setting of heart failure, there is a shift in muscle fibre types from oxidative to glycolytic (Drexler et al. 1992), and this could potentially drive some of the changes we found in ASIC function. Further studies labelling muscle afferents that innervate specific fast-twitch (e.g. extensor digitorum longus) and slow-twitch (e.g. soleus) muscles in mice with heart failure will be informative. It is well established that dysautonomia, characterized by increased sympathetic tone and diminished parasympathetic tone, plays a major role in the pathophysiology of the disease process, as well as contributing to the morbidity and mortality of heart failure. Increasing evidence suggests that overactivity of the EPR contributes to this dysautonomia and the exercise intolerance associated with heart failure. However, the underlying mechanisms that trigger this abnormal reflex are not well understood. The EPR is activated by receptors on muscle afferents that sense mechanical and chemical changes within exercising muscle. These afferents then synapse

Figure 7. A subset of muscle afferents from heart failure mice demonstrate altered properties consistent with those of ASIC2a/3 heteromers Muscle afferents from heart failure mice were divided into 2 groups as determined by the rate of recovery data: fast recovery (τ 0.12 ms) and slow recovery (τ > 0.12 ms). A, pH dose–response data for muscle afferents from control mice and the fast and slow recovery groups from heart failure mice normalized to currents evoked by pH 5 (n = 8–56; ∗ P < 0.05 and ∗∗ P < 0.01 for fast recovery compared to control). Lines are fits of the Hill equation (control pH50 = 6.24; slow recovery HF pH50 = 6.28; fast recovery HF pH50 = 6.01). B, fits of pH dose–response curves for control and the HF fast recovery group were compared to previously reported data for skeletal muscle afferents recorded from ASIC1a knockout (KO) mice (pH50 = 5.94) (Gautam & Benson, 2013), and heterologously expressed ASIC2a and -3 in cells from the CHO cell line (pH50 = 5.94) (Hattori et al. 2009). C, comparison of the time constants of desensitization muscle afferents from control, slow and fast recovery HF groups, and ASIC1a KO mice, (Gautam & Benson, 2013) as well as heterologously expressed ASIC2a/3 (Hattori et al. 2009) (n = 11–51; ∗ P < 0.05 and ∗∗ P < 0.01 for fast recovery compared to control groups. C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

J Physiol 593.20


in the spinal cord, where the signals are relayed to the autonomic control centres in the brain, which in turn adjust sympathetic/parasympathetic efferent output to control BP, HR, respiration and regional blood flow. In heart failure, while maladaptive changes probably occur at multiple levels within the reflex arc, evidence suggests that changes in the receptors within muscle afferents are important contributors. In support of this it has been shown that buffering muscle pH abolishes the exaggerated hyperventilatory response to exercise in patients with heart failure (Sterns et al. 1991). Moreover, because muscle pH is decreased to the same extent with exercise in heart failure patients compared to normal subjects, this suggests that this exaggerated ventilatory response might be due to increased sensitivity of the muscle afferents (Scott et al. 2003). Here we show that ASIC channels in muscle afferents are functionally altered in the setting of heart failure, and may contribute to altered autonomic reflexes associate with the disease.

References Alam M & Smirk RF (1937). Observations in man upon a blood pressure raising reflex arising from the voluntary muscles. J Physiol 89, 372–383. Benson CJ, Eckert SP & McCleskey EW (1999). Acid-evoked currents in cardiac sensory neurons: A possible mediator of myocardial ischemic sensation. Circ Res 84, 921–928. Benson CJ, Xie J, Wemmie JA, Price MP, Henss JM, Welsh MJ & Snyder PM (2002). Heteromultimers of DEG/ENaC subunits form H+ -gated channels in mouse sensory neurons. Proc Natl Acad Sci USA 99, 2338–2343. Cushman KA, Marsh-Haffner J, Adelman JP & McCleskey EW (2007). A conformation change in the extracellular domain that accompanies desensitization of acid-sensing ion channel (ASIC) 3. J Gen Physiol 129, 345–350. Drexler H, Riede U, Munzel T, Konig H, Funke E & Just H (1992). Alterations of skeletal muscle in chronic heart failure. Circulation 85, 1751–1759. El Accaoui RN, Gould ST, Hajj GP, Chu Y, Davis MK, Kraft DC, Lund DD, Brooks RM, Doshi H, Zimmerman KA, Kutschke W, Anseth KS, Heistad DD & Weiss RM (2014). Aortic valve sclerosis in mice deficient in endothelial nitric oxide synthase. Am J Physiol Heart Circ Physiol 306, H1302–H1313. Franciosa JA, Park M & Levine TB (1981). Lack of correlation between exercise capacity and indexes of resting left ventricular performance in heart failure. Am J Cardiol 47, 33–39. Francis GS & Cohn JN (1986). The autonomic nervous system in congestive heart failure. Annu Rev Med 37, 235–247. Gao Z, Henig O, Kehoe V, Sinoway LI & Li J (2006). Vanilloid type 1 receptor and the acid-sensing ion channel mediate acid phosphate activation of muscle afferent nerves in rats. J Appl Physiol (1985) 100, 421–426. Garry MG (2011). Abnormalities of the exercise pressor reflex in heart failure. Exerc Sport Sci Rev 39, 167–176. C 2015 The Authors. The Journal of Physiology C 2015 The Physiological Society

4585

Gautam M & Benson CJ (2013). Acid-sensing ion channels (ASICs) in mouse skeletal muscle afferents are heteromers composed of ASIC1a, ASIC2, and ASIC3 subunits. FASEB J 27, 793–802. Gautam M, Benson CJ & Sluka KA (2010). Increased response of muscle sensory neurons to decreases in pH after muscle inflammation. Neuroscience 170, 893–900. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Huffman MD, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Magid D, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Nichol G, Paynter NP, Schreiner PJ, Sorlie PD, Stein J, Turan TN, Virani SS, Wong ND, Woo D & Turner MB; American Heart Association Statistics Committee and Stroke Statistics Subcommittee (2013). Heart disease and stroke statistics–2013 update: a report from the American Heart Association. Circulation 127, e6-e245. Hammond RL, Augustyniak RA, Rossi NF, Churchill PC, Lapanowski K & O’Leary DS (2000). Heart failure alters the strength and mechanisms of the muscle metaboreflex. Am J Physiol Heart Circ Physiol 278, H818–H828. Hattori T, Chen J, Harding AM, Price MP, Lu Y, Abboud FM & Benson CJ (2009). ASIC2a and ASIC3 heteromultimerize to form pH-sensitive channels in mouse cardiac dorsal root ganglia neurons. Circ Res 105, 279–286. Hayes SG, Kindig AE & Kaufman MP (2007). Blockade of acid sensing ion channels attenuates the exercise pressor reflex in cats. J Physiol 581, 1271–1282. Hayes SG, McCord JL, Rainier J, Liu Z & Kaufman MP (2008). Role played by acid-sensitive ion channels in evoking the exercise pressor reflex. Am J Physiol Heart Circ Physiol 295, H1720–H1725. Hesselager M, Timmermann DB & Ahring PK (2004). pH dependency and desensitization kinetics of heterologously expressed combinations of acid-sensing ion channel subunits. J Biol Chem 279, 11006–11015. Ikeuchi M, Kolker SJ, Burnes LA, Walder RY & Sluka KA (2008). Role of ASIC3 in the primary and secondary hyperalgesia produced by joint inflammation in mice. Pain 137, 662–669. Immke DC & McCleskey EW (2001). Lactate enhances the acid-sensing Na+ channel on ischemia-sensing neurons. Nat Neurosci 4, 869–870. Jasti J, Furukawa H, Gonzales EB & Gouaux E (2007). Structure of acid-sensing ion channel 1 at 1.9 A˚ resolution and low pH. Nature 449, 316–323. Kaufman MP & Hayes SG (2002). The exercise pressor reflex. Clin Auton Res 12, 429–439. Koganti SR, Zhu Z, Subbotina E, Gao Z, Sierra A, Proenza M, Yang L, Alekseev A, Hodgson-Zingman D & Zingman L (2015). Disruption of K channel expression in skeletal muscle by targeted oligonucleotide delivery promotes activity-linked thermogenesis. Mol Ther 23, 707–716. Kuryatov A, Gerzanich V, Nelson M, Olale F & Lindstrom J (1997). Mutation causing autosomal dominant nocturnal frontal lobe epilepsy alters Ca2+ permeability, conductance, and gating of human α4β2 nicotinic acetylcholine receptors. J Neurosci 17, 9035–9047.

4586


Li J, Maile MD, Sinoway AN & Sinoway LI (2004a). Muscle pressor reflex: potential role of vanilloid type 1 receptor and acid-sensing ion channel. J Appl Physiol (1985) 97, 1709–1714. Li J, Sinoway AN, Gao Z, Maile MD, Pu M & Sinoway LI (2004b). Muscle mechanoreflex and metaboreflex responses after myocardial infarction in rats. Circulation 110, 3049–3054. Light AR, Hughen RW, Zhang J, Rainier J, Liu Z & Lee J (2008). Dorsal root ganglion neurons innervating skeletal muscle respond to physiological combinations of protons, ATP, and lactate mediated by ASIC, P2X, and TRPV1. J Neurophysiol 100, 1184–1201. Lingueglia E (2007). Acid-sensing ion channels in sensory perception. J Biol Chem 282, 17325–17329. Liu J, Gao Z & Li J (2010). Femoral artery occlusion increases expression of ASIC3 in dorsal root ganglion neurons. Am J Physiol Heart Circ Physiol 299, H1357–H1364. McCloskey DI & Mitchell JH (1972). Reflex cardiovascular and respiratory responses originating in exercising muscle. J Physiol 224, 173–186. McCord JL, Tsuchimochi H & Kaufman MP (2009). Acid-sensing ion channels contribute to the metaboreceptor component of the exercise pressor reflex. Am J Physiol Heart Circ Physiol 297, H443–H449. Mamet J, Baron A, Lazdunski M & Voilley N (2002). Proinflammatory mediators, stimulators of sensory neuron excitability via the expression of acid-sensing ion channels. J Neurosci 22, 10662–10670. Middlekauff HR, Chiu J, Hamilton MA, Fonarow GC, Maclellan WR, Hage A, Moriguchi J & Patel J (2004). Muscle mechanoreceptor sensitivity in heart failure. Am J Physiol Heart Circ Physiol 287, H1937–H1943. Middlekauff HR, Nitzsche EU, Hoh CK, Hamilton MA, Fonarow GC, Hage A & Moriguchi JD (2001). Exaggerated muscle mechanoreflex control of reflex renal vasoconstriction in heart failure. J Appl Physiol (1985) 90, 1714–1719. Middlekauff HR & Sinoway LI (2007). Increased mechanoreceptor stimulation explains the exaggerated exercise pressor reflex seen in heart failure. J Appl Physiol (1985) 102, 492–494. Molliver DC, Immke DC, Fierro L, Pare M, Rice FL & McCleskey EW (2005). ASIC3, an acid-sensing ion channel, is expressed in metaboreceptive sensory neurons. Mol Pain 1, 35. Negrao CE, Rondon MU, Tinucci T, Alves MJ, Roveda F, Braga AM, Reis SF, Nastari L, Barretto AC, Krieger EM & Middlekauff HR (2001). Abnormal neurovascular control during exercise is linked to heart failure severity. Am J Physiol Heart Circ Physiol 280, H1286–H1292. Notarius CF, Atchison DJ & Floras JS (2001). Impact of heart failure and exercise capacity on sympathetic response to handgrip exercise. Am J Physiol Heart Circ Physiol 280, H969–H976. Piepoli M, Clark AL, Volterrani M, Adamopoulos S, Sleight P & Coats AJ (1996). Contribution of muscle afferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure: effects of physical training. Circulation 93, 940–952.

J Physiol 593.20

Piepoli M, Ponikowski P, Clark AL, Banasiak W, Capucci A & Coats AJ (1999). A neural link to explain the ‘muscle hypothesis’ of exercise intolerance in chronic heart failure. Am Heart J 137, 1050–1056. Piepoli MF & Coats AJ (2007). Increased metaboreceptor stimulation explains the exaggerated exercise pressor reflex seen in heart failure. J Appl Physiol (1985) 102, 494–496. Ponikowski PP, Chua TP, Francis DP, Capucci A, Coats AJ & Piepoli MF (2001). Muscle ergoreceptor overactivity reflects deterioration in clinical status and cardiorespiratory reflex control in chronic heart failure. Circulation 104, 2324–2330. Price MP, McIlwrath SL, Xie J, Cheng C, Qiao J, Tarr DE, Sluka KA, Brennan TJ, Lewin GR & Welsh MJ (2001). The DRASIC cation channel contributes to the detection of cutaneous touch and acid stimuli in mice. Neuron 32, 1071–1083. Scott AC, Davies LC, Coats AJ & Piepoli M (2002). Relationship of skeletal muscle metaboreceptors in the upper and lower limbs with the respiratory control in patients with heart failure. Clin Sci (Lond) 102, 23–30. Scott AC, Wensel R, Davos CH, Georgiadou P, Kemp M, Hooper J, Coats AJ & Piepoli MF (2003). Skeletal muscle reflex in heart failure patients: role of hydrogen. Circulation 107, 300–306. Sinoway LI & Li J (2005). A perspective on the muscle reflex: implications for congestive heart failure. J Appl Physiol (1985) 99, 5–22. Sluka KA, Price MP, Breese NM, Stucky CL, Wemmie JA & Welsh MJ (2003). Chronic hyperalgesia induced by repeated acid injections in muscle is abolished by the loss of ASIC3, but not ASIC1. Pain 106, 229–239. Sluka KA, Radhakrishnan R, Benson CJ, Eshcol JO, Price MP, Babinski K, Audette KM, Yeomans DC & Wilson SP (2007). ASIC3 in muscle mediates mechanical, but not heat, hyperalgesia associated with muscle inflammation. Pain 129, 102–112. Smith SA, Mitchell JH & Garry MG (2006). The mammalian exercise pressor reflex in health and disease. Exp Physiol 91, 89–102. Smith SA, Williams MA, Mitchell JH, Mammen PP & Garry MG (2005). The capsaicin-sensitive afferent neuron in skeletal muscle is abnormal in heart failure. Circulation 111, 2056–2065. Sterns DA, Ettinger SM, Gray KS, Whisler SK, Mosher TJ, Smith MB & Sinoway LI (1991). Skeletal muscle metaboreceptor exercise responses are attenuated in heart failure. Circulation 84, 2034–2039. Sugiura T, Dang K, Lamb K, Bielefeldt K & Gebhart GF (2005). Acid-sensing properties in rat gastric sensory neurons from normal and ulcerated stomach. J Neurosci 25, 2617–2627. Sullivan MJ & Hawthorne MH (1995). Exercise intolerance in patients with chronic heart failure. Prog Cardiovasc Dis 38, 1–22. Voilley N, de Weille J, Mamet J & Lazdunski M (2001). Nonsteroid anti-inflammatory drugs inhibit both the activity and the inflammation-induced expression of acid-sensing ion channels in nociceptors. J Neurosci 21, 8026–8033.


J Physiol 593.20


Walder RY, Gautam M, Wilson SP, Benson CJ & Sluka KA (2011). Selective targeting of ASIC3 using artificial miRNAs inhibits primary and secondary hyperalgesia after muscle inflammation. Pain 152, 2348–2356. Walder RY, Rasmussen LA, Rainier JD, Light AR, Wemmie JA & Sluka KA (2010). ASIC1 and ASIC3 play different roles in the development of hyperalgesia after inflammatory muscle injury. J Pain 11, 210–218. Wang HJ, Li YL, Gao L, Zucker IH & Wang W (2010). Alteration in skeletal muscle afferents in rats with chronic heart failure. J Physiol 588, 5033–5047. Wemmie JA, Taugher RJ & Kreple CJ (2013). Acid-sensing ion channels in pain and disease. Nat Rev Neurosci 14, 461–471. Wilson JR, Rayos G, Yeoh TK, Gothard P & Bak K (1995). Dissociation between exertional symptoms and circulatory function in patients with heart failure. Circulation 92, 47–53. Xing J, Lu J & Li J (2015). ASIC3 contributes to the blunted muscle metaboreflex in heart failure. Med Sci Sports Exerc 47, 257–263. Xing J, Sinoway L & Li J (2008). Differential responses of sensory neurones innervating glycolytic and oxidative muscle to protons and capsaicin. J Physiol 586, 3245–3252. Zhang R, Khoo MS, Wu Y, Yang Y, Grueter CE, Ni G, Price EE Jr, Thiel W, Guatimosim S, Song LS, Madu EC, Shah AN, Vishnivetskaya TA, Atkinson JB, Gurevich VV, Salama G, Lederer WJ, Colbran RJ & Anderson ME (2005). Calmodulin kinase II inhibition protects against structural heart disease. Nat Med 11, 409–417.


4587

Additional information Competing interests None declared.

Author contributions D.D.G. and C.J.B. conceived the project and designed the experiments. All authors performed the research, analysed and interpreted the data, contributed to the writing, and approved the final version of the manuscript. All experiments were carried out at the University of Iowa. Funding This study was supported by a NIH T32 Institutional Training grant (5T32HL007121-39) awarded to D.D.G., NIH Shared Instrumentation Grants (RR26293 and OD019941) to R.M.W. and the Department of Veterans Affairs Merit Award (5I01BX000776) to C.J.B. Acknowledgements We thank Anne Harding, Mamta Gautam and the University of Iowa Echocardiography Core Facility for technical assistance.