Cellular Biology |
From the Vollum Institute (J.Y., H.N.W., L.A.N., E.W.M.), Oregon Health & Sciences University, Portland; Department of Integrative Physiology (J.Y.), Kyorin University School of Medicine, Tokyo, Japan; and Department of Physiology and Biophysics (L.A.N.), Federal University of Minas Gerais, Belo Horizonte, Brazil.
Correspondence to Edwin W. McCleskey, PhD, Vollum Institute, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97201-3098. E-mail mccleske{at}ohsu.edu
| Abstract |
|---|
|
|
|---|
Key Words: angina intermittent claudication acid-sensing ion channels lactic acidosis
| Introduction |
|---|
|
|
|---|
ASIC3 has 2 useful properties for sensing lactic acidosis. First, it is essentially 4 times as sensitive as a pH meter10 in the narrow range of extracellular pH (7.3 to 6.7) that occurs during cardiac and skeletal muscle ischemia.12,13 Second, it is more sensitive to lactic acid than to other forms of acid.14 However, a reason to question the hypothesis that ASIC3 is a sensor for ischemic pain is that ASIC3 rapidly desensitizes, whereas ischemic pain is persistent. ASIC3 and ASIC3/2 heteromers do generate sustained currents at pH 5 and below,15 but such extreme acidity does not occur during ischemic muscle activity (although it may during inflammation or tissue damage).
Here, we visit this issue with patch-clamp recordings during which we apply acid stimuli that are long lived and slowly progressing, meant to mimic acidosis that might occur during myocardial ischemia. We find that ASIC3 transfected into mammalian cell lines generates a perfectly persistent current at the modest extracellular pH changes typical of muscle ischemia. We characterize the kinetics and pharmacology of the current and demonstrate a pharmacologically similar current on somas of sensory neurons that innervate rat heart. The finding adds to the evidence supporting the role of ASIC3 in detecting ischemic pain in cardiac and skeletal muscle.
| Materials and Methods |
|---|
|
|
|---|
Labeling and Culture of Cardiac Dorsal Root Ganglia Neurons
Cardiac sensory neurons in rats were fluorescently labeled in vivo as previously described.9 Briefly, a left thoracotomy was performed and 25 µL of a suspension of 20 mg/mL 1,1'-dioctadecyl-3,3',3'-tetramethyl indocarbocyanine perchlorate (DiI) (Molecular Probes) was injected into the pericardial space of Sprague-Dawley rats (8 to 9 weeks old) anesthetized with 1 mL/kg rat anesthetic (in mg/mL: ketamine 55, xylazine 5.5, and acepromazine 1.1). Details of the surgery are provided in a PowerPoint file in the online data supplement, available at http://circres.ahajournals.org, or on request. Three to 5 weeks after the DiI injection, the rats were euthanized and the right and left dorsal root ganglia (DRG) from segments C8-T3 were isolated. The ganglia were dissociated in papain and collagenase/dispase solutions, successively.16 After trituration, the dissociated cells were placed on polylysine- and laminin-coated plastic in DMEM/F12 medium plus nerve growth factor (50 ng/mL) at 37°C in 5% CO2. After 4 to 6 hours, the medium was changed to L15 plus nerve growth factor, and the cells were maintained at room temperature in air. Cardiac DRG neurons were identified by fluorescence microscopy. All recordings were within 2 days of the dissection. The animals were cared for in accordance with the guidelines of the Institutional Animal Care and Use Committee of Oregon Health Sciences University.
Electrophysiology
Whole-cell patch-clamp experiments were performed on CHO cells and rat DRG neurons using an Axopatch-1C patch-clamp amplifier and PClamp 8 software (Axon Instruments). Current and voltage recordings were filtered (Bessel) at 1 to 2 kHz and 5 kHz, respectively, and digitized at 5 to 10 kHz (Digidata 1322A, Axon Instruments). Micropipettes were pulled from borosilicate glass (no. 7052; Garner Glass, Claremont, Calif) and lightly fire-polished to 1 to 4 M
resistance using a microforge (MF-83, Narishige). Voltage-clamp recordings were made at 70 mV unless otherwise stated. The series resistance was usually less than 8 M
and was compensated 70% to 90% by the Axopatch circuit. No data were included in the analysis if series resistance resulted in 12 mV (&10% of the driving force) or greater error when the holding potential was 70 mV.
The standard internal solution contained (in mmol/L) 30 KCl, 110 KOH, 4 NaCl, 10 MOPS, 10 EGTA, 5 MgCl2, 2 Na2ATP, and 0.3 Na3GTP, adjusted to pH 7.0 with methanesulphonic acid. The standard external solution contained (in mmol/L) 140 NaCl, 5 KCl, 7 N-methyl-D-glucamine (NMDG), 5 HEPES, 5 MES, 2 CaCl2, 1 MgCl2, and 10 glucose, with the pH adjusted with HCl or, for pH 8 solution, NaOH. The usual control pH of the extracellular solution was 7.4. For Na/K permeability experiments, NMDG-Cl replaced NaCl. For lactate modulation experiments, lactic acid (sodium salt, Sigma) replaced NaCl. For the amiloride modulation experiments, amiloride-HCl was added before pH adjustment. Solutions were applied through an array of 10 pipets (10 µL) under 40 cm of water pressure. Solution exchanges were accomplished within 20 ms by using a computer-driven solenoid valve system. Recordings were performed at room temperature. The liquid junction potential between the internal and external solutions was 5 mV, measured in current clamp and is corrected in all data.
Activation and inactivation curves were fit by nonlinear least squares (Origin, MicroCal Software) with the Hill equation: fraction={1+(K0.5/[H+]n)}1, where K0.5 is the proton concentration that causes half the channels to open. The ratio of permeability coefficient of Na+ and K+ (PNa/PK) was calculated from reversal potentials using the Goldman-Hodgkin-Katz equation: PNa/PK={[K+]o[K+]iexp(EF/RT)}/{[Na+]o[Na+]iexp(EF/RT)}, where PNa and PK represent the permeability coefficients of Na+ and K+, respectively, E is the reversal potential, F is the Faraday constant, R is the universal gas constant, and T is the absolute temperature. Data are expressed as mean±SE.
| Results |
|---|
|
|
|---|
|
Figure 1C plots the pH dependence of the average amplitude of this sustained current. Amplitudes are divided by the peak current measured at pH 5.0, showing that sustained currents are only a small fraction (&0.002) of the maximal ASIC current. Amplitudes of sustained current at pH 7.0 and pH 5.0 (open circle) are roughly similar. Neither current occurs in nontransfected cells, so they both flow through ASIC3. However, they can differ in ionic selectivity: sustained currents at pH 7.0 are Na+ selective (PNa/PK=8.2) in both ASIC3 homomers (Figure 1D) and heteromers (see Figure 5), whereas ASIC3 heteromers make nonselective sustained current at pH 5.0.15
|
The Window of Overlap of Activation and Inactivation Underlies Sustained Current
The magnitude of the sustained ASIC3 current does not depend on the history of pH changes (star in Figure 1C); this and the bell shape of the I versus pH relation suggested that it might arise from overlap of steady-state activation and inactivation curves. Figure 2A shows a protocol that allows simultaneous measurement of activation (amplitudes near the start of the conditioning pH interval), inactivation (amplitudes evoked by the pH 6.0 step), and persistent currents (amplitudes at the end of the conditioning interval; amplified in Figure 2B). Activation and inactivation data overlap slightly (Figure 2C). Multiplying values of the two curves that are fit to the data predict the probability at each value of pH of a "window current" caused by the overlap (Figure 2D, smooth curve). This calculation closely predicts both the pH dependence and relative amplitude of the experimentally measured sustained current (triangles). Evidently, the sustained current near pH 7.0 is a simple window current.
|
Lactate Decreases Threshold for the Window Current
Acidosis during ischemia occurs because of accumulation of lactic acid. ASIC3 is more sensitive to lactic acid than to other forms of acid because lactate subtly shifts the ASIC3 activation curve to more basic pH.14 We tested lactate on the sustained current and found that it increases sustained ASIC3 current but not because of a simple shift of the bell-shaped activation curve (Figure 3). Rather, the bell shape is broadened: lactate increases sustained current at pH 7.2 and 7.1 but does not alter current at pH 7.0 and below. In other words, lactate decreases the threshold for detecting sustained ASIC3 current. Figure I in the online data supplement confirms that the mechanism for lactate modulation of this sustained current is the same as that previously shown for transient ASIC current.14 In brief, lactate acts by decreasing the free concentration of extracellular divalent ions, and this affects the activation curve slightly more than the inactivation curve, leading to a broadened bell-shape curve.
|
Paradoxically, Amiloride Enhances Sustained ASIC3 Currents
To our surprise, amiloride enhanced the sustained current evoked in cells transfected with ASIC3 (Figure 4A). Amiloride is the classic blocker of ASICs,6 although the block is weak, occurring at &100-fold higher concentrations than block of epithelial Na channels.17 Mutation of the putative amiloride binding site on ASIC2 unmasks another site that allows amiloride to increase ASIC current.18 Thus, the prior literature reports complexity to amiloride action on ASICs.
|
All previously reported amiloride data were obtained by stimulating ASICs with transient pH changes that maximally activate the channels. The stimulus in Figure 4A differs by being both persistent and at a submaximal pH. Is it the persistence or the pH value that leads to enhancement? In Figure 4B, transient stimuli are applied to pH 7.0 (a) or 6.0 (b); 100 µmol/L enhances at pH 7.0 and blocks at pH 6.0. Thus, amiloride action depends on pH. At still higher amiloride concentrations, a block does develop at pH 7 (Figure 4Ba, right).
Figure 4C gives the relative current in the presence and absence of 100 µmol/L amiloride for different stimuli. The sustained current (S7.0) increases more than 2-fold, which is similar to the transient current evoked at pH 7.0. Amiloride block becomes evident at slightly more acidic pH. Figure 4D gives dose-response curves for amiloride actions on current evoked by brief pulses to pH 7.0 or 6.0 from a holding pH of 7.4. The data implies that amiloride enhances ASIC3 at the foot of the activation curve and blocks farther up the curve.
ASIC2a Confers Greater Sustained Current on ASIC3
Acid-gated currents are created by three ASIC genes, among which are 4 splice variants: ASIC1a, -1b, -2a, and -2b.7,19 ASICs 1a, 1b, 2b, 3 are certainly expressed in sensory neurons, but reports disagree about sensory expression of ASIC2a.15,20 We tested whether other subtypes generate sustained currents as does ASIC3. ASIC1a and -2a do not (Figure 5A, b and e), and coexpression of ASIC1a with ASIC3 eliminates the sustained current expected from ASIC3 homomers (Figure 5Ag). In contrast, coexpression of ASIC3 with either variant of ASIC2 generates sustained currents. Heteromeric ASIC2a/3 gives much larger currents that activate at slightly more acidic pH (Figure 5A, h and i); ASIC2b/3 is no different from ASIC3 alone (Figure 5A, j and k).
The activation curves (Figure 5B) out to pH 6.7 are bell shaped for ASIC3 and ASIC3/2b but not for ASIC3/2a. Tests of ASIC3/2a over a wider range of pH revealed a bell-shaped region around pH 6.7 followed by the larger, nonselective sustained current previously shown15 to occur at extreme pH (Figure 5C and 5D). The window current of ASIC3/2a heteromers is so acid-shifted and large that there is no pH that fails to express some sustained current. Window currents through the heteromeric channels were Na+ selective, whereas the sustained currents evoked at pH 5.0 did not distinguish Na+ from K+ (see Figure 5 legend for PNa/PK). Amiloride (100 µmol/L) increased sustained window currents evoked at pH 7.0 in both heteromers (ASIC2a/3 by 5.9-fold and ASIC2b/3 by 2.3-fold; n=4 for each; not shown).
Sustained Currents in Native Cardiac Sensory Neurons
We labeled sensory neurons that innervate the heart using a retrogradely transported fluorescent dye and then dissected and dissociated the appropriate dorsal root ganglia (Figure 6A). Figure 6B shows records from a representative cell. A very large current evoked by a brief pulse of pH 5 (left) was seen in almost all (28 of 30) cardiac afferents tested (supplemental Figure II shows a record from each cell.). A prolonged pH protocol (right) was applied under both voltage clamp (middle record) and current clamp (lower record) to 15 cardiac afferents. Sustained currents were seen in all cells but it triggered sustained firing of action potentials in only 4 (supplemental Figure III shows data from these 4). Figure 6C shows the pH dependence of current amplitude (all cells) and firing frequency (4 cells that fired repetitively). As a general rule, sustained current inputs and also sustained pH stimuli21 cause adaptation rather than repetitive firing in nociceptive sensory neurons, so it is unsurprising that only a minority of cells fired repetitively.
|
The current versus pH curve for native cardiac afferents (Figure 6C) is unlike any of the individual ASIC curves (Figure 5B): current is larger at pH 6.8 than at 7.0, unlike ASIC3 homomers, and current is evident at pH 7.2, unlike ASIC 3/2a heteromers. Other pH-sensitive channelsTRP channels or TASK channels, for examplemight underlie the current, as might a mix of different ASICs. Figure 6D shows that both amiloride and zinc, a coactivator of ASICs,22 enhance the sustained current evoked by pH 7.0. Thus, the sustained current in native cardiac afferents is pharmacologically like ASICs. This shows that ASICs contribute to the current, but it does not rule out a role for other ion channels as well.
| Discussion |
|---|
|
|
|---|
Importance of Small Sustained Currents in Sensory Transduction
Cardiac sensory afferents in dorsal root ganglia express ASIC3 at extraordinarily high levels: maximal currents at 70 mV are 10 to 30 nA,9,10 as high as the largest inward currents in the rat nervous system. Such a large current cannot be practical for sensory transduction because it depolarizes cells to levels that inactivate voltage-gated Na channels. We suggest that the small, sustained currents (10 to 50 pA) at the foot of the ASIC3 activation curve are the relevant signal for sensory transduction because they should depolarize only by 10 mV or less.
There are other examples where a physiologically relevant current is a small fraction of the maximal possible amplitude. Small, persistent currents through voltage-gated Na channels underlie burst firing in several neural tissues.2325 The hyperpolarization-gated current (also called Ih and If) that contributes to cardiac diastolic depolarization is a minute fraction of the total current possible through the HCN channels expressed in sinoatrial node cells.26 Excitability theory argues that an excess number of sensors, each with a very low probability of activation, increases the likelihood of reaching threshold for an action potential.27 The arrangement enhances electrical noise, which increases the chance of reaching threshold with low-level stimuli.28 Extracellular acidosis during muscle ischemia is relatively mild, dropping only to approximately pH 7.0.12,13 ASIC3-containing channels seem appropriate for detecting this subtle signal. The more extreme pH of inflammatory pain and bone cancer29 should fully desensitize ASIC3 channels; ion channels that are less pH sensitiveTRPV1 and 2-transmembrane K channelsare more likely sensors for these conditions.3032
The Complexity of Amiloride Action on ASICs
We found that amiloride enhances ASIC currents when they are evoked at the foot of the activation curve. Because amiloride and its analogs are the only commercially available blockers of ASICs, they are used to diagnose whether ASICs play roles in physiological studies. Our data suggest caution in such interpretations because amiloride might cause block, enhancement, or neither depending on the exact value of the pH at sensory nerve endings, a variable that is unmeasurable in complex tissue. This ambiguity compounds with the fact that amiloride affects multiple molecules (epithelial Na channels, Na/H exchangers, and T-type Ca channels) at the high concentrations used to block ASICs. The discovery of ASIC-specific peptide toxins offers important new pharmacology for studying ASICs in physiological preparations.3335
Enhancement of ASICs by amiloride under certain conditions has been previously reported. The sustained current carried by ASIC3 evoked at extreme acidity is enhanced by amiloride.6,15 Point mutation of an amino acid thought to be part of the intrapore blocking site for amiloride did not just eliminate amiloride block of an ASIC, it caused amiloride to increase ASIC current at all pH tested.18 The authors suggested there are 2 amiloride binding sites on ASICs, 1 that blocks and 1 that enhances. Our results may be explained if action of the enhancing site is more pronounced at the foot of the activation curve and the blocking site more important at the peak.
ASIC Subtypes
Our prior studies showed that most (but not all) of ASIC current in cardiac sensory neurons used ASIC3-containing ion channels.9,10 Many sensory neurons that innervate skeletal muscle have the same phenotype, whereas relatively few such cells innervate skin.11 However, ASIC3 may not often be expressed alone.7 The sea anemone toxin, APETx2, blocks ASIC3-like current in sensory neurons at concentrations that are higher than necessary for ASIC3 homomers, suggesting the presence of ASIC3 heteromers.34
ASIC3 homomeric channels differ from heteromers in a number of ways.6,15,22,36 Here we report a further kinetic difference between homomers and heteromers: the sustained window current at modest acidity is much larger in ASIC3/2a than in either ASIC3/2b or ASIC3 homomers. The ASIC3/2a current is so acid-shifted that the window current overlaps with the extreme acid-evoked sustained current, leaving no pH without some sustained current from ASIC3/2a heteromers. ASIC2b is certainly expressed in sensory neurons,15 but there is disagreement about ASIC2a: no ASIC2a mRNA was found in rat sensory neurons,15 but an antibody to 2a was reported to label sensory neurons.20 Given the rather extreme action of ASIC2a on window current and also the ability of kinases to modulate ASIC3/2 heteromers,37,38 it becomes important to clarify this controversy.
| Acknowledgments |
|---|
Sources of Funding
Research for this study was supported by NIH grants NS37010 and HL64840. J.Y. was supported by a grant-in-aid from the Promotion and Mutual Aid Corporation for Private Schools of Japan and a Research Project of Kyorin University. H.N.W. was supported by a National Research Service Award from the NIH; L.A.N. by a scholarship from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brazil.
Disclosures
None.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
Related Article:
This article has been cited by other articles:
![]() |
Z.-Y. Tan, Y. Lu, C. A. Whiteis, C. J. Benson, M. W. Chapleau, and F. M. Abboud Acid-Sensing Ion Channels Contribute to Transduction of Extracellular Acidosis in Rat Carotid Body Glomus Cells Circ. Res., November 9, 2007; 101(10): 1009 - 1019. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Chen, G. Polleichtner, I. Kadurin, and S. Grunder Zebrafish Acid-sensing Ion Channel (ASIC) 4, Characterization of Homo- and Heteromeric Channels, and Identification of Regions Important for Activation by H+ J. Biol. Chem., October 19, 2007; 282(42): 30406 - 30413. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Lingueglia Acid-sensing Ion Channels in Sensory Perception J. Biol. Chem., June 15, 2007; 282(24): 17325 - 17329. [Full Text] [PDF] |
||||
![]() |
K. A. Cushman, J. Marsh-Haffner, J. P. Adelman, and E. W. McCleskey A Conformation Change in the Extracellular Domain that Accompanies Desensitization of Acid-sensing Ion Channel (ASIC) 3 J. Gen. Physiol., March 26, 2007; 129(4): 345 - 350. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. K. Bubien and D. J. Benos Molecular pH Probes: Mediators of Angina and Ischemic Preconditioning? Circ. Res., September 1, 2006; 99(5): 453 - 454. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Research Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |