Resting Potential

Adv Physiol Educ 28: 139142, 2004; doi:10.1152/advan.00029.2004.
Report
REFRESHER COURSE
Cellular Homeostasis
Generation of resting membrane potential

Stephen H. Wright
Department of Physiology, College of Medicine, University of Arizona, Tucson, Arizona 85724
Received 2 July 2004; accepted in nal form 13 September 2004
Wright, Stephen H. Generation of resting membrane potential. Adv Physiol Educ 28: 139142, 2004; doi:10.1152/advan.00029.2004.This brief review is intended to serve as a refresher on the ideas associated with teaching students the physiological basis of the resting membrane potential. The presentation is targeted toward rst-year medical students, rst-year graduate students, or senior undergraduates. The emphasis is on general concepts associated with generation of the electrical potential difference that exists across the plasma membrane of every animal cell. The intention is to provide students a general view of the quantitative relationship that exists between 1) transmembrane gradients for K and Na and 2) the relative channel-mediated permeability of the membrane to these ions. Nernst equation; Goldman equation; electrical potential difference
Downloaded from ajpadvan.physiology.org on November 16, 2007
IT WOULD BE DIFFICULT TO EXAGGERATE the physiological significance of the transmembrane electrical potential difference (PD). This gradient of electrical energy that exists across the plasma membrane of every cell in the body inuences the transport of a vast array of nutrients into and out of cells, is a key driving force in the movement of salt (and therefore water) across cell membranes and between organ-based compartments, is an essential element in the signaling processes associated with coordinated movements of cells and organisms, and is ultimately the basis of all cognitive processes. For those reasons (and many more), it is critical that all students of physiology have a clear understanding of the basis of the resting membrane potential (so called to distinguish the steadystate electrical condition of all cells from the electrical transients that are the action potentials of excitable cells: i.e., neurons and muscle cells).1 How, then, does this electrical gradient arise? It is the consequence of the inuence of two physiological parameters: 1) the presence of large gradients for K and Na across the plasma membrane; and 2) the relative permeability of the membrane to those ions. The gradients for K and Na are the product of the activity of the Na-K-ATPase, a primary active ion pump that is ubiquitously expressed in the plasma membrane of (for all intents and purposes) all animal cells. This process develops and then maintains the large outwardly directed K gradient, and the large inwardly directed Na gradient, that are hallmarks of animal cells. For the purpose of this discussion, we will assume that the requisite gradients are
The PowerPoint lecture on the Generation of the Resting Potential, presented in the Cell Refresher Course at the 2003 Experimental Biology Meeting in Washington, DC, can be found at http://www.the-aps.org/education/refresher/ CellRefresherCourse.htm. Address for reprint requests and other correspondence: S. H. Wright, Dept. of Physiology, College of Medicine, Univ. of Arizona, Tucson, AZ 85724 (E-mail: shwright@u.arizona.edu).
in place (acknowledging that the mechanism of ion transport is beyond the scope of this presentation). The second parameter, the relative permeability of the plasma membrane to Na and K, reects the open versus closed status of ion-selective membrane channels. Importantly, cell membranes display different degrees of permeability to different ions (i.e., permselectivity), owing to the inherent selectivity of specic ion channels. The combination of 1) transmembrane ion gradients, and 2) differential permeability to selected ions, is the basis for generation of transmembrane voltage differences. This idea can be developed by considering the hypothetical situation of two solutions separated by a membrane permselective to a single ionic species. Side 1 (the inside of our hypothetical cell) contains 100 mM KCl and 10 mM NaCl. Side 2 (the outside) contains 100 mM NaCl and 10 mM KCl. In other words, there is an outwardly directed K gradient, and inwardly directed Na gradient, and no transmembrane gradient for Cl. For the purpose of this discussion, this ideal permselective membrane is permeable only to K. Let the experiment begin! The outwardly directed K gradient supports a net diffusive ux from inside to outside. As K diffuses from the cell it leaves behind residual negative charge, in this case, its counter anion, Cl (recall that, in our hypothetical scenario, neither Cl nor Na can cross the membrane). The resulting local imbalance of negative charge at the inside face of the membrane represents a force that can draw mobile positive charge at the external face of the membrane back into the cell; in this case, the only mobile ion capable of crossing the membrane is K. Signicantly, this electrical force can draw K from the low chemical concentration found outside the cell into the higher [K] inside the cell. As long as the chemical force of the outwardly directed K chemical gradient is larger than the oppositely oriented electrical force, there will be a net efux of K from the cell. However, as more and more K leaves, the residual negative charge within the cell, i.e., the attractive electrical force that
139
1043-4046/04 $5.00 Copyright 2004 The American Physiological Society
Report
140 GENERATION OF RESTING MEMBRANE POTENTIAL
draws K into the cell, gradually increases to a level that exactly balances the chemical force, resulting in an equilibrium condition. Whereas there may continue to be large unidirectional uxes into and out of the cell, the net ux under this equilibrium condition is zero. The equation that describes this balance of electrical and chemical forces is called the Nernst equation: VK RT Kin ln zF Kout
Table 1. Cytoplasmic and extracellular K, Na, Cl concentrations

Ion [Cytoplasm], mM [Extracellular], mM Veq, mV Pi, cm/s
K Na Cl
135 12 4
4 140 116
92 64 88
1107 1109 1108
where VK is the equilibrium electrical PD, which exactly opposes the chemical energy of the chemical gradient, the intracellular-to-extracellular K concentration ratio ([K]in/ [K]out). R is the gas constant with units of 8.31 J/(Kmol), T is absolute temperature in Kelvin (37C 310 K), F is Faradays constant at 96,500 coulombs/mol, and z is the valance of the ion question; 1 for K. It is instructive to insert the relevant values for R, T, F, and z, and to convert from the natural log to the common (base 10) log by multiplying by 2.303. The Nernst equation then becomes (at 37C) VK Kin 0.0615 Volts log 10 z Kout
These values are meant to represent only hypothetical values for a mammalian cell. Veq, calculated Nernstian equilibrium potentials for each ion gradient; Pi, typical permeability values for mammalian neurons. The listed concentrations of Cl in the cytoplasm and extracellular uid do not add up to the total concentration of K and Na. However, macroscopic electroneutrality is maintained in each compartment through the combined contribution of a diverse array of additional charged solutes (both anionic and cationic).
It is convenient to simplify this equation to an adequate (and useful) approximation Kin VK 60 mV log10 Kout When we consider the K gradient of our example (100 mM inside, 10 mM outside) we nd that this outwardly directed 10-fold gradient of a monovalent cation is balanced by a 60 mV electrical PD (in this case, inside negative). This introduces an additional valuable concept evident in another term synonymous with equilibrium potential, i.e., the reversal potential. In our example, the direction of net K ux was from inside to out until the equilibrium PD of 60 mV was reached. If the potential were, by some means, to become even more negative (say, 70 mV), then the direction of net K ux would reverse, i.e., a net ux from the low chemical concentration of K outside the cell into the higher K concentration inside, with this uphill ux driven by the imposed electrical force. At this point, it reasonable to ask, If we started with 100 mM K inside and there was a net efux of K from the cell, shouldnt the intracellular K concentration now be lower? This is an important issue. As it turns out, the amount of K that leaves the cell to produce the equilibrium potential is sufciently small that it cannot be measured chemically, despite the substantial electrical effect it has (see Sidebar 1). Therefore, to a rst approximation you can assume that [K]in (and Na) remain effectively constant during the shifts in transmembrane electrical potential of the type discussed here. So, in our hypothetical cell, with the constraint that the membrane is permeable only to K, the membrane potential is precisely dened by the K chemical gradient. Although it was emphasized above that intracellular ion concentrations generally do not change as a consequence of the downhill uxes associated with transmembrane voltage changes, it is instructive to consider what would happen if the K gradient were to change. In fact, changes in the K gradient, typically the result
of changes in [K]out, can be extremely important, both physiologically and clinically; see Sidebar 1. So, go ahead, grab a calculator and determine the new equilibrium potential that would arise if [K]out were suddenly increased to 20 mM or if the internal [K] fell to 50 mM (be advised, these macroscopic changes in K concentration would be associated with parallel changes in the concentration of one or more anions). Here is the rule of thumb: any manipulation that reduces the K gradient (i.e., either decreasing intracellular K or increasing extracellular K), will decrease the equilibrium potential for K (i.e., a voltage value closer to zero). In other words, if there is less energy in the chemical gradient, it will take less energy in an electrical gradient to balance it (go ahead, do the math. . .). You will not be surprised to learn that biological membranes do not show ideal permselectivity. Real membranes have a nite permeability to all the major inorganic ions in body uids. For most cells, the only ions that can exert any significant inuence on bioelectrical phenomena are the big three (in terms of concentration): K, Na, and Cl (Ca2 also contributes to bioelectric issues in a few tissues, including the heart). The Nernst equation, which represents an idealized situation, can be modied to represent the more physiologically realistic case in which the membrane shows a nite permeability to these three major players. The new equation is called the Goldman-Hodgkin-Katz Constant Field equation; or, more typically, the Goldman equation PKKin PNaNain PClClout V m 60 mV log10 PKKout PNaNaout PClClin where Vm is the actual PD across the membrane, and Pi is the membrane permeability (in cm/s) for the indicated ion. Close inspection reveals that the Nernst equation is lurking within the Goldman equation: if the membrane were to become permeable only to K, i.e., if PNa and PCl were zero, then the equation simplies to the Nernstian condition for K.2 Note that to account for the differences in valence, the anionic Cl concentrations are presented as out over in, rather than as the in over out convention used here for cations. It is worthwhile to consider the transmembrane ion gradients and ion-specic membrane permeabilities of a typical neuron
2 The Goldman equation can also be viewed as a weighted sum of the reversal potentials for each of the three ion gradients, with the individual ion permeabilities serving as weighting factors that dene the relative inuence of each gradient on the overall membrane potential.
Advances in Physiology Education VOL
28 DECEMBER 2004
Report
GENERATION OF RESTING MEMBRANE POTENTIAL 141
(Table 1). The calculated Nernstian equilibrium potential for K, Na, and Cl establish the boundary conditions for the electrical PD across the cell membrane; i.e., our cell cannot be more negative than 92 mV or more positive that 64 mV (Fig. 1) because there are no relevant chemical gradients sufciently large to produce larger PDs. At rest, importantly, the membrane permeability of most cells, including neurons, is greatest for K, due to the activity of several distinct populations of K channels that share the general characteristic of being constitutively active under normal resting conditions. The relative contribution to the resting potential played by these channels varies with cell type, but in neurons relevant players include members of the family of inwardly rectifying K channels (KIR) and the K(2P) family of K leak channels. The combination of an outwardly directed K gradient (the product of Na-K-ATPase activity) and a high resting permeability to K makes the interior of animal cells electrically negative with respect to the external solution. However, the nite permeability of the membrane to Na (and to Cl; see Sidebar 2) prevents the membrane potential from ever actually reaching the Nernstian K potential. The extent to which each ion gradient inuences the PD is dened by the permeability of the membrane to each ion, as is evident from inspection of the Goldman equation. Even very large concentrations exert little inuence if the associated Pi value is small. However, if the membrane were suddenly to become permeable only to Na, the result would be a Nernstian condition for Na, with a concomitant change in membrane potential. Although under normal physiological conditions the concentration terms of the Goldman equation remain relatively constant, the permeability terms do not. Indeed, large, rapid changes in the ratios of permeability for different ions represent the basis for the control of bioelectric phenomena. On a molecular level, membrane permeability to ions is dened by the activity of membrane channels (the molecular basis of which, like so many other things, is outside the scope of this review). Indeed, a large increase in PNa (owing to activation of
a population of voltage-gated Na channels) is the basis of the transient depolarization of membrane potential that is associated with the neuronal action potential. In summary, the combination of an outwardly directed K gradient (the product of Na-K-ATPase activity) and a high resting permeability to K makes the interior of animal cells electrically negative with respect to the external solution. Changes in either of these controlling parameters, i.e., transmembrane ion gradients or channel-based ion permeability, can have large, and immediate consequences. Although the stability of ion gradients has been emphasized here, in fact, changes in these gradients can occur, generally with pathological consequences (see Sidebar 2). The [K]out is particularly susceptible to such changes. Because in absolute terms it is comparatively small (i.e., 4 mM), increases in [K]out of only a few millimoles per liter can have large effects on resting membrane potential (prove this to yourself using the Goldman equation and the permeability parameters listed in Table 1). Such changes can occur as a consequence of, for example, crushing injuries that rapidly release into the blood stream large absolute amounts of K (from the K-rich cytoplasm in cells of the damaged tissue). Alternatively, failure of the Na-K-ATPase during ischemia can result in local increases in the [K]out, a problem exacerbated by both the low starting concentration of K and the low volume of uid in the restricted extracellular volume of densely packed tissues (e.g., in the heart or brain). The other side of the coin, i.e., alteration in membrane permeability to ions, can arise as a consequence of an extraordinary number of pathological defects in ion channel proteins (or channelopathies). Of particular relevance to the resting membrane potential are lesions in one or more subunits of the KIR channels mentioned previously. Mutations in these channels, and the consequent changes in resting membrane potential, have been linked to persistent hyperinsulinemic hypoglycemia of infancy, a disorder affecting the function of pancreatic beta cells; Bartters syndrome, characterized by hypokalemic alkalosis, hypercalciuria, increased serum aldosterone, and plasma renin activity; and to several polygenic central nervous system diseases, including white matter disease, epilepsy, and Parkinsons disease. Sidebar 1 How many ions cross the membrane during bioelectric activity? Although the PD at equilibrium is the result of the physical separation by the membrane of cations from anions, the actual number of ions separated to produce this PD is very small. At time zero in our example, there were equal concentrations of cations and anions on both Side 1 and Side 2 (i.e., 0.11 M each). On a macroscopic scale, this is always the case and is the basis for the presumption that electroneutrality must be maintained in any system; you cant have a jar that contains only cations! However, as K diffuses from Side 2 to Side 1, there will be a net increase in positive charge along the Side 2 (outside) surface of the membrane, producing a microscopic imbalance in charge with respect to the residual negative charge (i.e., Cl) left behind along the inside surface of the membrane. The net movement of K produces a current, and the storage of charge on the membrane surface is analogous to the action of a capacitor. When the net movement of K is zero (i.e., at equilibrium), the current ow will be zero and maxi28 DECEMBER 2004
Fig. 1. Graphical representation of the Nernstian equilibrium potentials (Vi) for Na (VNa), K (VK), and Cl (VCl); and the resting membrane potential (Vm) calculated by using the Goldman equation. The relevant ion concentrations and permeabilities are listed in Table 1.
Report
142 GENERATION OF RESTING MEMBRANE POTENTIAL
mum storage of charge in the capacitor has occurred. The amount of charge stored is proportional to the net movement of K and is given by q cV where q is charge (coulombs), c is the capacitance of the membrane (in farads), and V is the applied voltage (the PD). For the purpose of this discussion, we will use the capacitance of a generic biological membrane, 1 106 F/cm2. As shown previously, the equilibrium voltage for a 10-fold gradient of a monovalent ion 0.060 V. Thus q 1 106 F/cm20.060V 6.0 108 coulombs/cm2 There are about 1 105 mol of monovalent ion per coulomb (recall the Faraday constant). Thus if we consider the specic instance of our example, i.e., K efux across a K permselective membrane, this charge separation corresponds to 6.0 108 coulombs/cm21 105 mol K/coulomb 6.0 1013 moles of K/cm2 So, to move the membrane potential from zero to a PD of 60 mV, the necessary efux of K amounts to a minimum of 6 1013 mol/cm2 of membrane, leaving behind its negatively charged counter ion (in this case, Cl). It is instructive to consider the potential impact of moving that many ions into or out of cellular-sized compartments. Consider, for example, an axon 10 m in diameter. Taking into account the appropriate surface-to-volume relationship, the ux associated with a 60 mV PD shift should change the axoplasmic concentration of an ion, such as K by less than one part in 40,000! Moreover, the modest imbalance of charged particles arising from the net efux of K is effectively localized to the region very near the membrane, with the excess positive charge (in this case, K) in the extracellular compartment held electrostatically along the outer surface of the membrane by the excess negative charge (in this case, Cl) that is arrayed along the inner surface of the membrane. Importantly, conductive K uxes can, at least under pathophysiological conditions, produce substantial elevations in [K]out, owing to 1) the low starting concentration of K in the extracellular uid; and 2) the restricted volume of the extracellular compartment of some tissues (including some neuronal tissues). Sidebar 2 What about Cl? A common question is: What about the inuence of Cl? The resting PD is as close (or closer) to VCl as it is to VK. Why dont we conclude that Cl is the dominant ion in dening the resting membrane potential? The answer lies in the fact that the cell is spending its energy, via the Na-K-ATPase, in establishing the gradients for K and Na, not Cl. Indeed, to a rst approximation, the observed inwardly directed Cl gradient, with its calculated Nernstian potential of 89 mV, arises in large part through the simple passive distribution of Cl in response to the electrical gradient
(i.e., the resting PD) that is effectively dened by the outwardly directed K gradient. In other words, 1) the cell actively builds transmembrane gradients of K and Na; 2) the outward ux of K down its gradient that reects the large PK of the resting cell (relative to PNa), shifts the PD toward the equilibrium (Nernstian) potential for K; 3) Cl, which is high in the blood, moves into the cell in response to its chemical gradient; 4) but the inside negative potential established by K serves as a force to limit the buildup of intracellular Cl. Importantly, this is the case even in those cells (e.g., some skeletal muscle cells) in which the channel-mediated permeability to Cl exceeds that for K. The fact that the Nernstian Cl potential is usually not exactly equal to the resting PD of the cell means that there are one or more active transport processes that keep Cl away from an equilibrium distribution (e.g., secondary active Cl/HCO 3 exchange). Whereas in neurons Cl is a minor player in establishing the resting membrane potential, there are several situations in which membrane permeability to Cl (as inuenced by the activity of Cl channels) is very important. An increase in Cl permeability is an effective way to stabilize the resting membrane potential by opposing changes in PD that would otherwise be produced by uxes of K or Na. Thus Cl permeability is modulated as one means to inuence synaptic transmission. Conversely, decreases in PCl make it easier for the PD to shift away from its resting value. Thus, in the disease myotonia congenita, the observed hyperexcitability of skeletal muscle cells is the result of a decrease in PCl (arising from defects in the ClCN1 Cl channel) and the concomitant decrease in the stability of the resting membrane potential. Sidebar 3 Traps students fall into. The material presented here is aimed at rst-year medical students and graduate students. Although graduate students may need selected core concepts presented in greater depth, I have found that they must rst be comfortable with the interrelation that exists between ion gradients and membrane permeability. In my experience, the two largest misconceptions that students cling to are 1) that changes in PD are associated with large changes in ion concentrations (see Sidebar 1); and 2) that changes in ion concentration that are frequently referred to in discussions of manipulating membrane potential inuence membrane potential because they represent a net addition of charge (e.g., increasing external K from 4 to 10 mM results in adding positive charge to the extracellular solution because K is a cation!). As teachers, it is easy to feed these misconceptions through statements like Na rushes in (implying to some that a lot of Na crosses the membrane), and add K to the extracellular solution (implying to some the addition of only a positively charged species). Dealing with such issues requires attention to detail in the description of these events, and substantial repetition. In the nal analysis, however, the best way for students to become comfortable with the concepts associated with establishment of the membrane potential is to work with the Nernst and Goldman equations.
28 DECEMBER 2004
Adv Physiol Educ 28: 143154, 2004; doi:10.1152/advan.00046.2004.
Report
REFRESHER COURSE
Cellular Homeostasis
Ion homeostasis, channels, and transporters: an update on cellular mechanisms

George R. Dubyak
Department of Physiology and Biophysics, Case Western Reserve University, School of Medicine, Cleveland, Ohio 44106
Submitted 13 September 2004; accepted in nal form 17 September 2004
Dubyak, George R. Ion homeostasis, channels, and transporters: an update on cellular mechanisms. Adv Physiol Educ 28: 143154, 2004; doi:10.1152/advan.00046.2004.The steady-state maintenance of highly asymmetric concentrations of the major inorganic cations and anions is a major function of both plasma membranes and the membranes of intracellular organelles. Homeostatic regulation of these ionic gradients is critical for most functions. Due to their charge, the movements of ions across biological membranes necessarily involves facilitation by intrinsic membrane transport proteins. The functional characterization and categorization of membrane transport proteins was a major focus of cell physiological research from the 1950s through the 1980s. On the basis of these functional analyses, ion transport proteins were broadly divided into two classes: channels and carrier-type transporters (which include exchangers, cotransporters, and ATP-driven ion pumps). Beginning in the mid-1980s, these functional analyses of ion transport and homeostasis were complemented by the cloning of genes encoding many ion channels and transporter proteins. Comparison of the predicted primary amino acid sequences and structures of functionally similar ion transport proteins facilitated their grouping within families and superfamilies of structurally related membrane proteins. Postgenomics research in ion transport biology increasingly involves two powerful approaches. One involves elucidation of the molecular structures, at the atomic level in some cases, of model ion transport proteins. The second uses the tools of cell biology to explore the cell-specic function or subcellular localization of ion transport proteins. This review will describe how these approaches have provided new, and sometimes surprising, insights regarding four major questions in current ion transporter research. 1) What are the fundamental differences between ion channels and ion transporters? 2) How does the interaction of an ion transport protein with so-called adapter proteins affect its subcellular localization or regulation by various intracellular signal transduction pathways? 3) How does the specic lipid composition of the local membrane microenvironment modulate the function of an ion transport protein? 4) How can the basic functional properties of a ubiquitously expressed ion transport protein vary depending on the cell type in which it is expressed?
ION TRANSPORT PROTEINS
are most broadly categorized as being either channels or transporters. From the functional perspective, transporters have been classically viewed as vectorial enzymes whose catalytic cycle involves 1) a relatively selective recognition/binding of the transported ion(s), 2) conformational changes in the transporter protein itself due to binding of the ion(s), and 3) the coupling of these conformational changes to physical movement of the ion(s) across the membrane bilayer. In contrast, channels have been viewed as
Address for reprint requests and other correspondence: G. R. Dubyak, Dept. of Physiology and Biophysics, Case Western Reserve Univ., School of Medicine, Cleveland, OH 44106 (E-mail: george.dubyak@case.edu).
transport proteins that facilitate the physical translocation of ions by mechanisms that involve relatively little energetic interaction between the channel protein and the transported ion(s). Rather, extrinsic factors, such as changes in membrane potential or the binding of small regulatory molecules (e.g., extracellular neurotransmitters or intracellular second messengers) dictate whether the channel protein is an open or gated state (capable of ion transport) or a closed state (incapable of ion transport). Thus the major conformational changes in channels are produced when the various extrinsic factors regulate those domains of the channel protein that act as gates for controlling the accessibility of the transported ion to a pore domain. In turn, the pore domain acts as the pathway or conduit for ions moving from one side of the membrane to the other. Because the actual transport of ions through channels does not require a dened sequence of energetic interactions between the transport protein (the channel) and the transported ions, the rate of ion transport through channel proteins is usually many times faster than the rate of transport through carrier-type proteins. This constitutes one of the major functional criteria classically used to distinguish channels vs. transporters. Another classic distinction is how the function of an ion transport protein is related to the electrochemical gradient(s) for the ion(s) being transported. The driving force for the transmembrane ux of an ion is dictated by the electrochemical gradient, which reects the difference in the concentrations of that ion on each side of the membrane as well as any electrical potential across the membrane. All channels mediate the movement of ions down their respective chemical or electrochemical gradients; this is termed passive transport. Thus the major function of ion channels is to facilitate, usually for only very short durations, the movement of ions down the electrochemical gradients previously established across either the plasma membrane or the membranes of intracellular organelles; this transient dissipation of the ionic gradient comprises a perturbation of cell function and underscores the predominant use of ion channels for rapid signal transduction or information transfer. A prototypical example is the role of voltage-gated Na channels in rapidly increasing the Na permeability of the plasma membrane of excitable cells during the rising phase of the action potential. In contrast, most transporters (but no channels) can catalyze the active transport of certain ions against their electrochemical gradients, ultimately through a net expenditure of cellular energy. Primary active transporters directly couple the movement of the transported molecule to the binding and hydrolysis of ATP. Such ATPase-coupled ion pumps are typied by the Na-K-ATPase, which functions to
143
1043-4046/04 $5.00 Copyright 2004 The American Physiological Society
Report
144 NEW PERSPECTIVES ON ION TRANSPORT PROTEINS
actively transport Na from the intracellular compartment (with low [Na]) into the extracellular compartment (with high [Na]) in exchange for the movement of K from its low concentration in extracellular uids into the high K compartment of the cytosol. Secondary active transporters couple the movement of one ion against its electrochemical gradient to the movement of another ion down its electrochemical gradient; this mechanism is used in both cotransporters (e.g., Na-K2Cl cotransporter) or exchange transporters (e.g., Na/H exchanger). Thus a major function of carrier-type transporters is to establish and maintain the transmembrane electrochemical gradients of the physiologically signicant inorganic cations and anions. It is important to appreciate that most ion transport proteins exhibit certain common structural features regardless of whether they function as channels or transporters. The vast majority of membrane transport proteins possess several transmembrane-spanning domains (usually -helices) that are physically juxtaposed to create the amino acid-lined pore, which acts as the physical conduit for the ion as it passes from one side of the membrane to the other. The composition of this amino acid-lined pore creates a microenvironment, with suitable electrostatic and polar characteristics, that greatly reduces the energetic constraints that effectively prevent the direct passage of an inorganic ion through a lipid bilayer. The relative rates at which ions enter the pore from one side of the membrane and exit the pore at the other side determines the effective net transmembrane ux of the ion. Channels facilitate only the net transmembrane ux of an ion from a cis compartment with higher concentration to a trans compartment with lower concentration. Thus the major mechanistic issue for channel-type transporters is merely limiting the overall accessibility of the ion to the pore by using a single gate
(another protein domain of the channel itself) to prevent entry of cis ions into the pore until an appropriate gating signal is generated by the cell (Fig. 1A). Once ions from the cis compartment have entered the pore domain, it is much more probable that they will exit the pore at the trans side (and remain there) due to the lower ion concentration in that compartment. However, it should be stressed that ion channels also contain particular arrangements of amino acid residues (usually within the permeability pore itself) that favor the entry of ions with particular charges and sizes and will retard ions with the reverse charge or the wrong size. Although channels are most broadly categorized as cation or anion specic, many cation and anion channels exhibit marked selectivity for particular species of ions. For example, certain voltage-gated cation channels select either Na over K, whereas others transport K rather than Na. In contrast, the effective function of a carrier-type transporter raises a more complex set of mechanistic issues, particularly for transporters, such as the Na-K-ATPase, that catalyze the active transmenbrane ux of an ion from the cis compartment with lower concentration to the trans compartment with higher concentration. In this case, at least two gates are required to sequentially control accessibility to the pore domain of ions within both the trans and cis compartments (Fig. 1B). Although the opening of a single gate at the cis side will permit entry/exit of ions from that compartment with lower concentration into the pore region, a second gate needs to be simultaneously closed to prevent entry/exit of ions from the trans compartment with higher concentration. Once ions have entered the pore domain from the cis compartment, the cis gate must be closed before the trans gate opens to permit accessibility of ion within the pore domain to the trans compartment (and vice versa). This simultaneous closing of both
Fig. 1. Fundamental differences between ion channels and ion transporters in terms of gating and occlusion states [adapted from Gadsby (14)].
28 DECEMBER 2004
Report
NEW PERSPECTIVES ON ION TRANSPORT PROTEINS 145
accessibility gates effectively traps or occludes ions that have entered the pore from the cis compartment. Although subsequent opening of the second gate will allow ions to move between the pore domain and trans compartment, this gating needs to be coupled to some other conformational change within the transporter that effectively favors the exit of the ions previously occluded within the pore domain while minimizing the entry (or reentry) of the same species of ion from the trans compartment. This can be accomplished by changing the afnity and/or selectivity of the pore domain for particular ions depending on whether the pore is opened to the cis or trans compartment. This is accomplished in the Na-K-ATPase by conformational changes that reduce the association of Na ions with the permeability pore while increasing the afnity of the pore cavity for extracellular K ions. By thermodynamically coupling the sequential opening and closing of the two accessibility gates to the sequential conformational changes that alter the ionic selectivity of the pore domain, carrier-type transporters can catalyze the net ux of ions against their electrochemical gradients. The required conformational changes that modulate the ionic selectivity and afnity of carriers can be energized by either the direct binding of transported ions to specic sites within the pore domain or parallel chemical reactions (e.g., ATP hydrolysis, protein phosphorylation/dephosphorylation) catalyzed by the carrier-type transporter.
ION TRANSPORT PROTEINS AS CHANNELS VS. TRANSPORTERS: NOT AS DIFFERENT AS WE THINK
Concept: Both channel-like activity and transporter-type activity can be accommodated within the basic structures of most ion transport proteins. Model Example: The conversion of the Na-K-ATPase transporter, an ATP-powered ion exchange pump, into a cation channel by palytoxin (PTX), a marine coral toxin that selectively binds to the Na-K-ATPase. Most physiologists will readily agree that the Na-KATPase pump is one of the most important and extensively characterized ion transporters. The Na-K-ATPase is an archetypal active transporter that catalyzes the net efux of
three Na ions in exchange for the net accumulation of two K ion per reaction cycle. Recent analyses of the interactions of the Na-K-ATPase with PTX, a marine coral poison, indicate that this prototypical carrier-type transporter may continue to provide paradigmatic insights regarding the fundamental function and structure of such transport proteins (3, 4, 15, 16, 20, 22). Signicantly, these PTX studies suggest that the basic functional and structural features that dene ion transporters vs. ion channels may be much more similar than expected. Full appreciation of the insights and implications of these PTX experiments requires a review of the generally accepted mechanism of the Na-K-ATPase as an ion transporter. Figure 2 illustrates this cyclic interplay of chemical reactions, gating movements, and progressive ion transport as it occurs in the Na-K-ATPase. The initial reactions correspond to the entry of cytosolic Na into the permeability pore followed by occlusion of now-resident Na ions within the cavity dened by simultaneous closing of both the cis and trans gates. The initial binding of cytoplasm-derived Na ions within the cis-gated open state of the Na-K-ATPase triggers the ATP-driven phosphorylation of a regulatory aspartate within the carrier (i.e., the carrier acts as a Na-activated protein kinase). This autophosphorylation drives the conformational change that closes the cis gate to trap the bound Na ions within the now occluded permeability pore. Closing of the cis gate also enables the opening of the trans gate. This opening of the trans gate is accompanied by yet another conformational change that greatly reduces the Na afnity of the pore domain while increasing the relative afnity for K. After the sequential exit of Na ions and entry of K ions, the binding of the K ions triggers a conformational switch that results in both reclosing of the trans gate to trap the bound K in an occluded state of the pore and the dephosphorylation of the regulatory aspartate (i.e., the carrier now acts as a K-activated protein phosphatase). Finally, the binding of ATP to low-afnity sites (distinct from the high-afnity sites required for aspartate phosphorylation) accelerates the opening of the cis gate with consequent deocclusion of the trapped K ions and their release into the cytosolic compartment.
Fig. 2. Cyclic interplay of chemical reactions, gating movements, and progressive ion transport as it occurs in the Na-K-ATPase [adapted from Artigas and Gadsby (3)].
28 DECEMBER 2004
Report
It is easy to appreciate how this stepwise and highly coupled sequence of gating movements, conformational changes, and enzymatic reactions contributes to the much slower rates of net ionic transport facilitated by transporters relative to channels. Indeed, the rate of ion ux through an open channel can be up to 106 faster than the ux catalyzed by transporter/pumps. A corollary of this difference in ux rate is that there must be an exceedingly low probability (on the order of 1/105 to 106) that both gates of a transporter/pump might open at the same time (3, 14). These signicant differences in physiological function might also portend major differences in the structures of the pore domains and gating domains in channels vs. transporters. Structures of the pore and gating domains of several different ion channels have now been described at high resolution through extensive functional characterization of mutant proteins and X-ray crystallographic analysis (10, 12). In addition to providing detailed molecular information regarding the crystallized channels per se, these structures have provided a framework for directing new functional analyses of related channel proteins that have not yet been crystallized. For example, Estevez et al. (13) used the crystal structure of the bacterial Ec-ClC and St-ClC Cl transporters (see below for why these are called transporters rather than channels), to direct mutagenesis studies of the mammalian ClC-1 Cl channel. Based on the Cl-coordinating sites of Ec-CLC and St-ClC, these investigators mutated relevant residues of ClC-1 to map the interaction site of 9-anthracene carboxylic acid (9-AC), a known blocker of Cl ux through ClC-1. Crystal structures have also been described for several transporters, including the Ca2-bound and Ca2-free states of the sarcoplasmic/endoplasmic reticulum Ca2-ATPase (SERCA) pump, another extensively studied ion transporter protein (37, 38) Although transmembrane (TM) helices 4, 5, and 6 contain the key residues for binding the two Ca2 ions transported per each reaction cycle of a SERCA pump, the precise structures of the entire permeability pathway and the gating domains still need to established. At present, there are no crystal structures of the Na-K-ATPase. However, modeling of the Na-KATPase based on the SERCA1a pump lattices suggests that the homologous TM4, 5, and 6 of the Na-K-ATPase contain key residues suitable for binding two of the three Na ions and the two K ions (30). The third Na may be coordinated by a pocket formed by apposition of TM 6, 8, and 9. However, the nature of the ion translocation pore domain and the structure/ function of the regulatory gates remain key issues for a more complete mechanistic description the Na-K-ATPase. In this regard, the recent studies of PTX-modied Na-K-ATPase transporters are likely to provide a valuable starting point for generating new and experimentally testable hypotheses. PTX is a marine toxin rst isolated 33 years ago from the Palythoa toxica coral (27). It is a large (3,000 Da) nonpeptide molecule that consists of a chain of 100 carbons with a complex variety of organic side groups (20). PTX was identied on the basis of its striking lethality when injected into mice; this lethality in intact organisms is due to rapid disruption of cardiac function together with severe vasoconstriction. However, studies with isolated cells have revealed that PTX readily affects most cell types by depolarizing the plasma membrane potential as a secondary consequence of inducing a small-conductance, nonselective cation channel activity. In excitable cells, such as neurons or myocytes, this is manifested
as an increase in the resting Na conductance (23, 28). In osmotically fragile cells, such as eryrthrocytes, PTX elicits Na accumulation, K loss, and swelling that can culminate in lysis (2, 15). Early studies by Habermann et al. (15) revealed that these effects of PTX on erythrocyte ion homeostasis were substantially attenuated by the presence of ouabain, the classical and highly selective inhibitor of the Na-K-ATPase. Subsequent biochemical and physiological analyses over the past 20 years have identied the Na-K-ATPase as the major molecular target of PTX (20). The take-home message from these studies indicates that bound PTX directly affects the Na-K-ATPase by converting this archetypal active ion transporter into a non-selective cation channel that acts to dissipate the transmembrane Na and K gradients which are generated by the Na-K-ATPase acting in its normal (i.e., non-PTX occupied) transporter mode. PTX induces this remarkable functional transformation by overriding the fundamental mechanism that distinguishes pumps and transporters from channels: the stringent maintenance of at least one gate in the closed position at every phase of the catalytic transport cycle. Thus PTX binding appears to greatly increase the probability for simultaneous opening of both gates of the Na-K-ATPase protein. Recent electrophysiological analyses, together with the mutagenesis studies, have begun to dene the mechanistic characteristics of the PTX-liganded Na-K-ATPase protein. Earlier studies of yeast cells expressing normal or mutated versions of mammalian Na-K-ATPase pumps revealed that PTX can induce ouabain-sensitive cation uxes even when the heterologous Na-K-ATPase protein lacked residues critical for ATP hydrolysis and autophosphorylation (34). This indicated that PTX could independently stabilize the open gates that regulate ionic access to the permeability pathway or pore in the absence of the coupled chemical reactions that regulate gating in the normal transport protein. However, these studies left open the reverse mechanistic question: how is the interaction of PTX with the Na-K-ATPase channel modulated quantitatively or qualitatively by the ligands, such as intracellular ATP or extracellular K, that regulate conformational changes in the native transporter? A converse set of questions is concerned with how the binding of PTX to the Na-KATPase may modulate the ability of the protein to bind ATP and to be autophosphorylated. In a recent series of elegant studies, Artigas and Gadsby (3, 4) have addressed these questions by using classic patchclamp electrophysiological methods to analyze the interactions of PTX with native Na-K-ATPase transporters in excised outside-out patches from HEK293 cells or guinea pig ventricular myocytes. High concentrations (2 nM) of the toxin could induce channel activity, albeit with a low open-state probability (Po), in the absence of ATP (at the cytosolic side). Conversely, in the presence of internal ATP, even low concentrations (25 pM) of PTX rapidly induced Na channels with high Po. Other experiments showed that poorly hydrolyzable analogs of ATP were equally effective in potentiating the Po of PTX-induced channels. These ndings indicated that the PTXliganded channels can be allosterically modulated by the simultaneous binding of ATP, which is a principal physiological regulator and substrate of Na-K-ATPase conformation. Other experiments showed that the presence of extracellular K markedly altered the kinetics of channel opening/closing
28 DECEMBER 2004
Report
by the PTX-modied Na-K-ATPase by increasing the rate and probability of channel closure. Recall that the entry of extracellular K ions into sites of the Na-K-ATPase permeability pore acts to close the trans (extracellular) gate and so induce occlusion of the bound K. Thus these various electrophysiological data indicated that PTX-liganded Na-KATPase channels retain this sensitivity to allosteric modulation by extracellular K, another physiological regulator and substrate of the Na-K-ATPase. Interestingly, other recent studies by Horisberger and colleagues (15, 16, 22) have used cysteine-scanning mutagenesis of the Na-K-ATPase to determine sites in the transmembrane segments of the PTX-modied protein that are accessible to small hydrophilic sulfhydryl reagents. This approach, commonly used by channel biophysicists, relies on the ability of the small sulfhydryl reagent to enter the permeability pore of an open-gated channel, covalently react with any cysteine residues (native or purposely substituted for the native, noncysteine amino acids at particular positions) that line the pore, and thereby attenuate ionic ux through the permeability pore. Signicantly, cysteine substitution of residues in the TM4, TM5, and TM6 segments of the Na-K-ATPase become accessible in the PTX-bound state. Recall that these TM segments contain the likely binding sites for the three Na ions and two K ions that are normally transported by the NaK-ATPase in its physiological transporter mode (30). Thus the observations of Artigas and Gadsby (3, 4), together with those of Horisberger and colleagues (15, 16, 22), suggest that PTX acts to stabilize intermediate conformational states, which are normally extremely short lived, of the Na-KATPase-mediated ion transport process (Fig. 2). This indicates that PTX may prove to be a highly useful tool for dening the structures and regulation of the gates that control the entry, occlusion, and release of Na and K during the physiological cycling of the Na-K-ATPase. Moreover, these ndings indicate that the fundamental structures of channels and transporters may share a highly related molecular architecture
derived from common ancestral membrane proteins (1, 14). In this regard, Accardi and Miller (1) have recently demonstrated that bacterial (Escherichia coli) homologs (12) of the mammalian ClC-family Cl channels actually function as Cl/H exchange transporters rather than high-ux Cl channels. For channels, the ancestral protein architecture has evolved to retain only one major gating domain so as to favor rapid ionic ux when the gate is the open position. Conversely, in transporters, this ancestral structure may have evolved to retain (or possibly to gain) at least two gating domains whose movements are allosterically regulated by the ion substrate(s) of the transport reaction and/or biochemical reactions directly catalyzed by the transport protein.
INTERACTIONS OF ION TRANSPORT PROTEINS WITH ADAPTER PROTEINS: NO TRANSPORTER IS AN ISLAND
Concept: Many channels and transporters physically associate with adapter proteins that regulate the subcellular localization of the transport protein. Additionally, the associated adapter protein may control the direct interaction of the transport protein with signal transduction complexes that include receptors, second messenger-producing effector enzymes, and protein kinases. Model Example: The role of NHERF-family adapter proteins in the localization and acute regulation of the type 2 Na-phosphate cotransporter (NPT2) within apical signaling complexes of renal epithelial cells. Over the past decade or so, the study of ion transport proteins has increasingly utilized the tools of biochemistry and cell biology to describe the subcellular localization of ion transport proteins and their regulation by various intracellular signaling cascades. Such analyses have been facilitated by two types of reagents: 1) specic antibodies directed against native ion transport proteins and 2) cDNA expression vectors (plasmids or viruses) encoding chimeric versions of ion transport proteins with various protein tags, such as green uorescent
Fig. 3. Schema of how adapter proteins can regulate the trafcking of ion transport proteins and the assembly of the signaling complexes that include ion transport proteins, receptors, effector enzymes, and cytoskeletal proteins. PLC, phospolipase C; GPCR, G protein-coupled receptor.
28 DECEMBER 2004
Report
protein or smaller peptide epitopes. The use of these reagents with standard immunoprecipitation and immunohistological methods has demonstrated that many transport proteins interact, via direct protein-protein association, with a variety of regulatory proteins that have been generically termed adapter proteins. By denition, adapter proteins contain several different types of highly conserved modular protein-protein interaction domains (e.g., the src homology 2 domains that recognize phosphotyrosine motifs) that allow adapters to act as molecular bridges between two or more different proteins; these can include signaling proteins, membrane transport proteins, and cytoskeletal elements. The number and variety of adapters continues to increase and includes both soluble and intrinsic membrane proteins. As schematically depicted in Fig. 4, adapter proteins affect the function of ion transport proteins in two, nonmutually exclusive ways. 1) Adapters can modulate the steady-state levels of an ion transport protein at the plasma membrane by affecting trafcking to, or retention within, the plasma membrane compartment. 2) Adapters can concentrate an ion transport protein within subdomains (punctate patches, hot spots, clusters) of the plasma membrane that are also enriched in the receptors or other signaling proteins known to acutely regulate the trafcking or activity of the transport protein. In these ways, adapter proteins can comprise the central nexus of multiprotein complexes that greatly increase the selectivity, subcellular localization, and temporal control of critical ion transport responses to extrinsic stimuli. Adapter proteins that interact with ion transport proteins (channels or transporters) often contain so-called PDZ interaction domains (reviewed in Ref. 29). The PDZ domain was named for three proteins in which the motif was rst recognized: PSD-95 (postsynaptic density-95, a synaptic protein), Dlg (Discs-large, a protein in Drosophila septate junctions), and ZO-1 (zona occludins-1, a protein in the tight junctions of epithelia). PDZ-containing proteins are widespread, with over 400 identied in humans alone. NHERFs comprise a family of PDZ-containing adapter proteins that play important roles in the regulation of several ion transport proteins (36). Indeed, the term NHERF is an acronym for Na/H exchanger (NHE) regulatory factor. Two distinct NHERF proteins, NHERF-1 and NHERF-2, have been extensively characterized at the
Fig. 4. Three major protein-protein interaction domains of Na/H exchanger-regulatory factor (NHERF) adapter proteins and some of the proteins that can interact with each domain [adapted from Shenolikar and Weinman (36)]. See text for denitions.
biochemical, physiological, and cell biological levels. NHERFs are 55-kDa proteins that contain two adjacent PDZ domains: PDZ-I near the NH2 terminus and PDZ-II in the center. COOH-terminal to these PDZ domains is an ERM (ezrin, radixin, moesin, and merlin) motif, a protein-protein interaction domain found in adapter proteins that bridge membrane and cytoskeletal elements. Given these three distinct protein interaction domains, it is not surprising that NHERFs have been shown to bind about 50 different proteins (39) including the 16 illustrated in Fig. 4. In addition to the cytoskeletal proteins mentioned previously, these encompass G protein-coupled receptors (GPCR) such as the 2-adrenergic receptor, P2Y1-nucleotide receptor, and parathyroid hormone (PTH) receptor; ion transport proteins, such as the cystic brosis transmembrane regulator (CFTR) Cl channel, the H-ATPase pump, the type 3 Na/H exchanger (NHE3), and the type 2 Na-phosphate cotransporter (NPT2); and second messenger-generating enzymes (phospholipase C-1). Identifying and characterizing the physiological consequences of these various NHERF-based interactions has become an active and exciting area of study in ion homeostasis biology (18, 36, 39, 41). A particularly salient and informative example is provided by the role of NHERFs in orchestrating the regulation of NPT2-mediated phosphate transport by G protein-coupled PTH receptors in the kidney. NPT2 in the proximal tubules is responsible for most reuptake of phosphate from the glomerular ltrate. Increases in serum phosphate (due to diet and other factors) elevate PTH levels, which, in turn, act to decrease the NPT2-mediated recovery of ltered phosphate. This PTH-induced decrease in NPT2 transport activity involves a rapid decrease in the density of NPT2 transporters in the apical membranes due to acute changes in the steady-state trafcking of NPT2 between cell surface and intracellular membrane pools (26). Thus activation of PTH receptors predominantly acts to alter the subcellular localization of NPT2 rather than to decrease the total cellular content of NPT2. The G protein-dependent signaling cascades that trigger these changes in steady-state NPT2 trafcking involve both the phospholipase C (PLC) and adenylyl cyclase (AC) pathways. In this regard, it is important to note that the PTH receptor belongs to a subgroup of GPCR that can efciently couple to the activation of PLC and/or AC depending on cellular context. Segre and colleagues (2426) have dissected the mechanisms by which PTH triggers these changes in NPT2 in the recent series of studies that underscores the critical role of NHERFs in the directly coordinating regulation of both NPT2 transporters and PTH receptors. Their studies utilized an established cell line (OK) from opossum kidney that retains high expression of NPT2 and PTH receptors as well as the ability to rapidly decrease Na-dependent phosphate accumulation in response to PTH. In contrast, OKH cells, a clonal variant of the OK cell line, exhibited little change in phosphate accumulation during PTH stimulation despite expressing levels of functional NPT2 and PTH receptors similar to those in the parental cells. This inability of PTH to appropriately regulate NPT2 was correlated with a hypoexpression of NHERF1 in the OKH cells, and stable expression of NHERF1 in these cells (OKH-N1 clone) completely rescued the ability of PTH to repress phosphate accumulation (24). Conversely, expression of a truncated version of NHERF1 lacking the ERM domain
28 DECEMBER 2004
Report
for cytoskeletal interaction was unable to restore regulation of phosphate transport. Confocal microscopic analysis of NPT2 subcellular localization revealed similar levels of NPT2 at the apical cell surface of both OKH-N1 and OKH cells in the absence of PTH stimulation. However, the NPT2 was diffusely distributed over the apical surface in OKH cells but clustered into punctuate patches on the OKH-N1 cell surface. Signicantly, activation of PTH receptors in the OKH-N1 cells triggered redistribution of NPT2 out of the apical plasma membrane and into an intracellular membrane pool while NPT2 remained at the apical surface of PTH-treated OKH cells. Other experiments (25, 26) showed that the presence of NHERF is necessary for effective coupling of PTH receptors to Gi-protein-regulated PLC- effector enzymes that drive increases in cytosolic Ca2. In the absence of NHERF, PTH receptors couple only to the AC effector enzyme, which, by itself, appears insufcient to generate optimal signals for the redistribution of cell surface NPT2 into intracellular pools. These ndings of Segre and colleagues (2426) strongly suggest that direct interaction with NHERF proteins appears mandatory for the coordinated physiological functions of both NPT2, a key phosphate transport protein, and the PTH receptor, a critical endocrine element for homeostatic regulation of phosphate. A physiological role for NHERFs in the regulation of NPT2 function is also consistent with the phenotype of knockout mice, in which the NHERF1 gene has been deleted in collaborative research directed by Shenolikar and Weinman and their colleagues (9, 35). Although total NPT2 protein content is similar in the kidneys of NHERF1/ and NHERF1/ mice, the steady-state level of NPT2 at the apical surface of the proximal tubules is much lower in the knockout animals. This reduced apical content of NPT2 is matched by elevated levels of internalized NPT2. Moreover, the NHERF1/ animals exhibit 1) signicant phosphate wasting into the urine under normal steady-state conditions and 2) an inability to adequately match renal phosphate transport to diet-induced changes in total phosphate intake. The altered NPT2 phenotype of the NHERF1 knockout mouse is more complex and nuanced than might be expected from the tissue culture analyses of NHERF1-NPT2 interactions described above. Recall that NPT2 is present mainly in the plasma membrane of the (NHERF1-low) OKH cells under basal conditions. This contrasts with the predominantly intracellular location of NPT2 in the NHERF1-null proximal tubule cells under basal conditions. This disparity likely reects differences in the relative rates of steady-state NHERF1 trafcking to, and retention within, apical membranes under in vivo conditions, with the complex contribution of multiple endocrine and other regulatory factors, vs. the more rigidly controlled cell culture environment. Nonetheless, both the in vivo and ex vivo observations support the same fundamental implication: that autonomous expression and function of NPT2 as a transporter island, disconnected from the contextual whole of proximal tubule cells wherein it is natively expressed, fails to reveal the true physiology of this membrane protein. This theme is becoming the rule, rather than the exception, as the functions of cloned channels and transporters are increasingly analyzed in a similar way. It should be stressed that the particular interactions between NHERFs, the PTH receptor, and the NPT2 transporter described above represent only a subset of the various ways in
which adapter proteins have been shown to regulate ion transport processes. First, interaction of the 2-adrenergic receptor with NHERF allows that receptor to modulate Na/H exchange activity without the classical mediation by G proteindependent processes (18). Second, in addition to NHERFs, two other PDZ-containing proteins, i.e., PDZK1 and PDZK2, play important roles in regulated various ion transporters in the proximal tubule (19). Third, dimeric PDZ proteins can effectively cross-link and thereby induce dimerization of some ion transport proteins with signicant functional consequences (33, 40). Fourth, because some cytoskeletal proteins, e.g., ezrin, also act as AKAPs (protein kinase A-anchoring proteins), they can colocalize such protein kinases in the immediate vicinity of ion transporters within NHERF-based protein ensembles; this can facilitate the highly localized phosphorylation and functional modication of the transporter (11, 41).
INTERACTIONS OF ION TRANSPORT PROTEINS WITH LOCAL LIPIDS: THE BILAYER AS MORE THAN A LOW DIELECTRIC PERMEABILITY BARRIER
Concept: The function of many channels and transporters is directly modulated by the specic binding of phosphatidylinositol 4,5-bisphosphate (PIP2). This binding permits rapid modulation of ion transport activity by highly localized changes in PIP2 synthesis or degradation. Model Example: Hypersensitization of the activity of nociceptive vanillanoid receptor channels (VR1) by inammatory mediators for GPCRs that activate PLC and PIP2 degradation. The detailed biophysical study of ion channels now routinely involves the use of patch clamp electrophysiological methods to analyze single-channel current in excised samples of plasma membrane. This allows investigators to systematically and precisely manipulate the composition of the solutions bathing either side of the membrane in which the ion channel resides. A common observation in such studies is that one or more basic indexes of the ion channels function (open or closed probability, single-channel conductance, ion selectivity, relative stability or lability) can be markedly changed by the presence or absence of ATP at the cytoplasmic surface of the membrane patch. This ATP-dependent perturbation of channel function can reect at least three distinct types of regulation. ATP binding can exert direct allosteric effects on channel conformation and function; in these cases, the modulatory actions of ATP can readily be mimicked by nonhydrolyzable analogs of ATP (i.e., there is no requirement for a transfer of phosphate from ATP). Alternatively, ATP can act as a substrate for a protein kinase colocalized with the channel in a membrane-associated signaling complex such as those organized by the PDZ adapter proteins described above. Here, the modulatory effects of ATP cannot mimicked by nonhydrolyzable analogs but can be attenuated by pharmacological inhibitors of the colocalized protein kinase (if known). The kinase may directly phosphorylate the channel or a distinct modulatory protein that directly associates with the channel. However, regulation of channel activity by ATP can often be unequivocally dissociated from ether a direct allosteric action of the nucleotide or the phosphorylation of the channel (or channelassociated protein) by a protein kinase. Rather, the effects of ATP reect its ability to also act as a substrate for a variety of phosphatidylinositol kinases (i.e., lipid kinases) that can seri28 DECEMBER 2004
Report
ally phosphorylate phosphatidylinositol (the most abundant inositol lipid) to phosphatidylinositol 4-monophosphate (PIP) to phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP3). In turn, the higher-order phosphoinositides (PIP2 in particular) act as the allosteric modulators of channel activity. Moreover, biochemical approaches, involving reconstitution of various ion transporters in lipid vesicles of known composition, have indicated that the activities of some ion pumps, cotransporters, and exchangers are also strongly modulated by anionic phospholipids such as PIP2. Thus the lipid membrane in which a channel or transporter resides is not simply an inert permeability barrier and support matrix, but also acts as a dynamic regulator of the conformation and function of many ion transport proteins. As reviewed by Hilgemann et al. (21), the activity of some ion transport proteins increases as the local PIP2 is elevated, whereas other channels or transporters exhibit reduced activity in the presence of PIP2. Ion transport proteins positively regulated by PIP2 include plasma membrane Ca2-ATPase pumps (PMCA), Na/Ca2 exchangers (NCX), Na/H exchangers (NHE), inward rectier K channels (IRK), G protein-activated inward rectifer K channels (GIRK), ATP-inhibited K channels (KATP), and the ryanodine-sensitive Ca2 release channels (RyR) of endoplasmic reticulum. PIP2 negatively regulates the cyclic nucleotide-gated channels (CNG) of the mammalian visual system, the transient receptor potentiallike channels (TRPL) of the Drosophila visual system, the 1,4,5-inositol trisphosphate (IP3)-gated Ca2 release channels of the endoplasmic reticulum, and, as discussed in detail below, the vanillanoid receptor channels (VR1) that mediate perception of noxious stimuli. The ability of PIP2 to act as a positive or negative modulator of ion transport proteins underscores the versatility of this relatively low-abundance phospholipid as a central player in signal transduction at the cellular level. Most readers are more familiar with the role of PIP2 as the substrate for the PLC effector enzymes that hydrolyze PIP2 to generate the second messengers IP3 and diacylglycerol (DAG). PLC enzymes include the PLC- isoforms that are regulated by certain trimeric
G proteins and their upstream GPCR, and the PLC- enzymes that can be regulated by the receptor tyrosine kinases (RTK) targeted by growth factors, or nonreceptor tyrosine kinases. Thus exposure of cells to neurotransmitters, local mediators, or growth factors that couple to PLC enzymes can trigger very rapid decreases in PIP2 levels in the immediate vicinity of an ion transport protein. This negative action on PIP2 levels is opposed by the inositol lipid kinases that replenish the PIP2 pool (Fig. 5). As discussed in the previous section, PLC enzymes can be intimately colocalized with ion transport proteins via adapter proteins such as NHERF. Exactly how local PIP2 modulates the activity of particular ion transport proteins remains an area of active investigation. In some cases, this reects direct binding of PIP2 to intracellular domains of the transport protein (21, 32). In other cases, PIP2 seems to act indirectly by affecting the interactions of the transport protein with other regulatory proteins. The role of PIP2 as a physiological modulator of an important ion transport protein is provided by the elegant work of Julius and colleagues (5, 6, 32) on the vannilanoid receptor channel VR1. VR1, which is highly expressed in sensory nerve endings, belongs to the TRP (transient receptor potential) superfamily of nonselective cation channels (31). VR1 was originally identied as the receptor for capsaicins, the active ingredients in hot peppers and other spicy foods. The binding of lipid-permeable capsaicins, or endogenous lipids, such as arachidonic acid, can gate VR1 to the open state to produce depolarization of the sensory neuron and consequent perception (by the organism) of a noxious or painful sensation. Signicantly, gating of VR1 to its open conformation is directly produced by elevated temperature or an elevated concentration of extracellular H. Thus VR1 also mediates the sensation of pain-producing burns elicited by high heat or exposure to acid (Fig. 6). As anyone who has experienced a wound, skin cut, or other tissue damage appreciates, the area in the immediate vicinity of damage is usually characterized by a much heightened sensitivity to touch, heat, cold, and tissue swelling. The Julius group (6, 32) was able to show that, at the cellular level, this
Fig. 5. Schema of how local changes in phosphatidylinositol 4,5-bisphosphate (PIP2) levels can regulate the functional characteristics of a nearby ion transport protein.
28 DECEMBER 2004
Report
Fig. 6. Noxious stimuli that regulate the gating of the vannilanoid receptor channel VR1 [adapted from Caterina and Julius (5)].
sensitization reected the actions of extracellular mediators, such as bradykinin, that accumulate at sites of tissue damage as part of the protective inammatory response. Signicantly, many of these local inammatory mediators act as agonists for GPCR that activate PLC- effector enzymes. Chuang et al. (6) demonstrated that VR1 channel activity is negatively regulated by the ambient PIP2 levels present in the membranes of nonstimulated cells. Thus only small membrane currents were elicited by extracellular acid or caspsaicin pulses in a voltage-
clamped HEK293 cell expressing recombinant VR1. However, in cells briey pretreated with bradykinin, the same pulses of acid or capsaicin triggered very large depolarizing inward currents. In an extensive series of experiments, Chiang et al. conclusively showed that this effect was due to the ability of bradykinin to transiently decrease PIP2 levels and thereby derepress the direct, tonic effects of the lipid on VR1 gating by the primary noxious stimuli. Prescott and Julius (32) further demonstrated that mutation of selected amino acids on the
Fig. 7. Schema of how tissue-specic expression of a modulator protein may alter transport phenotype of given ion transport protein in different tissues.
28 DECEMBER 2004
Report
Fig. 8. Cell biology of the Na-K-ATPase in terms of the tissue-specic synthesis of and -subunits and modulation by coexpressed FXYD proteins [adapted from Crambert and Geering (8)].
intracellular COOH-terminal tail of the VR1 signicantly increased the sensitivity of the channels to gating by low concentrations of acid or capsaicin, even in the absence of bradykinin, while reducing the potentiation of VR1 gating by bradykinin and other stimuli that activate PIP2 breakdown. This identied the VR1 COOH terminus as the likely site of highafnity PIP2 interaction.
INTERACTIONS OF ION TRANSPORT PROTEINS AND MODULATOR PROTEINS: CELL-SPECIFIC CONTEXT EXPLAINS ALL
Concept: An ion transport protein can exhibit tissue-specic differences in function that reect its direct association with modulator proteins that are expressed in a tissue-specic or stimulus-specic fashion. Model Example: The role of FXYD-family membrane proteins in the tissue-specic modulation of Na-K-ATPase pump activity. Detailed structure-function analysis of ion channels or transporters routinely involves heterologous expression of cDNAs (or cRNAs) encoding the normal or purposefully altered sequence of the transport protein in various recipient cell types (COS, CHO, HEK293, Xenopus oocytes). Sometimes, significant differences are observed in the fundamental transport properties of the expressed channel or transporter depending on the cell type used for heterologous expression. In other cases,
properties of the ion transport protein are similar when heterologously expressed in various cell types but are quite different from the functional properties of the transporter as natively expressed in specialized cells or tissues. Additionally, the properties of ion transport proteins in native tissues may be altered (sometimes in unexpected or surprising ways) when the expression of other proteins is genetically ablated in knockout animal models. Thus the detailed function or phenotype of a given transport protein, even one that that is widely or ubiquitously expressed, may be signicantly altered by tissuespecic or cell-specic context due to the presence or absence of particular modulator proteins (Fig. 7). A particularly striking example of how modulator proteins can direct the phenotypic characteristics of an ion transport protein is provided by the interactions of the Na-K-ATPase with members of the FXYD protein family (reviewed in Refs. 7 and 8). Figure 8 illustrates the overall cell biology of the Na-KATPase in terms of synthesis, trafcking, and association between its - and -subunits. As already noted, Na-KATPase is one of the extensively studied and characterized ion transport proteins. Although its basic function and catalytic cycle is invariant regardless of tissue source, signicant differences in the kinetic details of the Na pump from different tissues have been widely reported. These differences include variations in its turnover rate or apparent afnity (K) for its
Table 1. Tissue distribution, associated Na pump -subunit isoform, and functional effects of selected FXYD proteins
FXYD Subtype (alternative name) Tissue Distribution Interacting -Subunit Vmax KNa KK KATP
FXYD1 (PLM) FXYD2 FXYD4 (CHIF) FXYD7
Heart and muscle Kidney Kidney and colon Brain
1 and 2 1 1 1
7 2 7 2
1 1 2 7
1 1 1 or 7 1
ND 2 7 ND
Information is adapted from Cornelius and Mahmoud (7) to show the differential tissue distribution of selected FXYD proteins, their association with Na pump -subunit subtypes, and their effects on kinetic constants of associated Na pump, Vmax, maximal velocity; KNa , afnity constant for intracellular Na; KK , afnity constant for extracellular K; KATP , afnity constant for intracellular ATP; ND, not determined. Advances in Physiology Education VOL
28 DECEMBER 2004
Report
physiological substrates (Na, K, and ATP). Although these differences have sometimes been linked to the tissue-specic expression of the four different isoforms of the -subunit or three distinct -subunits, there remain examples of tissuespecic differences in function despite the similar expression of - and -subunit isotypes. Recent studies have indicated that these latter differences reect the presence of particular modulator proteins from the so-called FXYD family. The FXYD family comprises a group of seven, relatively small (66178 residues) intrinsic membrane proteins that share a common structure of 1) an extracellular amino terminus containing a juxtamembrane stretch of four amino acids with the sequence (-F-X-Y-D-) that gives the family its name, 2) a single transmembrane domain, and 3) a short cytosolic COOH termimus that is quite divergent except for a conserved juxtamembrane domain containing a serine adjacent (or almost adjacent) to positively charged amino acids. Most of the FXYD proteins were rst identied and characterized as small plasma membrane proteins of uncertain function but distinctive tissue expression. For example, FYXD1 was originally described as plasmalemman, a muscle-derived protein that induced an unusual channel-like activity when ectopically expressed in Xenopus oocytes. FXYD4 was initially characterized as CHIF (corticosteroid hormone-induced factor) a gene predominantly expressed in Na-conserving segments of the colon and kidney (medullary collecting duct). However, subsequent research has demonstrated that the various FXYD proteins predominantly act by directly associating with the Na-K-ATPase and thereby modulating the kinetic properties of this vectorial enzymes. Table 1 summarizes the tissue-specic expression and functional consequences of the best characterized FXYD proteins. The take-home message is that FXYD proteins modulate Na-K-ATPase pump activity in a way that matches the particular rates of Na removal and K reaccumulation that are dictated by the specic functions of a tissue or cell. This is reasonable from a purely teleological view, given the very signicant quantitative variations in Na-K-ATPase activity among different tissues (i.e., reect on the differences in overall ionic uxes in an actively ring neuron, an epithelial cell specialized for recovery of Na from the urine, or an erythyrocyte). Consideration of how a particular FXYD subtype is integrated into the specialized ion transport requirements of a particular tissue is exemplied by the role of FXYD4/CHIF in the medullary collecting duct (MCD) of the kidney. In times of whole body Na decit, the kidney needs to efciently salvage Na from the glomerular ltrate. A major mechanism involves the increased secretion of the mineralocorticoid hormone aldosterone that acts to increase expression of ENaC (epithelial Na channels) in the apical membranes of MCD epithelial cells. Although this greatly increases the rate of Na inux into the MCD cell, net Na recovery requires an equivalent increase in the rate of Na efux at the basolateral MCD plasma membrane. In this regard, aldosterone also increases expression of FXYD4 which then acts to increase the Na afnity of Na-K-ATPase and, thereby, the overall rate of transepithelial Na ux. It is likely that additional families of transport modulator proteins will be identied among the many orphan genes present in the various completed genomes.
SUMMARY
The past 10 years have witnessed major advances in identication and characterization of ion transport proteins at the molecular level. With the routine use of molecular biological approaches for the study of ion transport, we now understand the basic structural features and patterns of tissue expression for hundreds of different ion channels, ion exchangers, ion cotransporters, and ATP-driven ion pumps. Indeed, there are nearly complete ion transporter-omes for several of the favorite model organisms of transport physiologists. The challenge for future transport physiologists is to understand how the members of the various superfamilies of structurally related ion transport proteins are selectively used (or co-opted?) for tissue- or cell-specic ion homeostasis. This brief survey of new perspectives on ion transport indicates several take-home lessons that should guide such future studies. 1) Precise homeostasis of the major inorganic cations (Na, K , Ca2, H) and anions (Cl, phosphate, bicarbonate) is fundamental to all cells. 2) However, cell-specic expression of different membrane transport proteins and regulatory factors permits wide variations in the absolute rates of transmembrane ux of these ions. 3) These cell-specic differences in ionic ux are exploited for highly contextual and tissue-specic differences in function, such as solute ow (e.g., transepithelial movements of metabolites) or information transfer. 4) These tissue-specic differences in ionic ux are regulated at multiple levels: via increased/decreased expression of membrane transport protein genes, via changes in the steadystate trafcking of membrane transport protein to and from the plasma membrane, via direct posttranslational modication (e.g., phosphorylation) of the membrane transport proteins, via direct association with tissue-specic adapter or modulator proteins, and via the local lipid composition of the membrane bilayer. Undoubtedly, such studies will uncover additional, as-yetunappreciated nuances for the regulation of the ion transport proteins. If ndings from the past few years are an indicator, one can be certain there will be plenty of new new perspectives for the next Refresher Course on Cell Homeostasis.
REFERENCES 1. Accardi A and Miller C. Secondary active transport mediated by a prokaryotic homologues of ClC Cl- channels. Nature 427: 803807, 2004. 2. Ahnert-Hilger G, Chhatwal GS, Hesller HJ, and Habermann E. Changes in erythrocyte permeability due to palytozin as compared to amphotericin B. Biochim Biophys Acta 688: 486494, 1981. 3. Artigas P and Gadsby DC. Na/K pump ligands modulate gating of palytoxin-induced ion channels. Proc Natl Aacd Sci USA 100: 501505, 2003. 4. Artigas P and Gadsby DC. Large diameter of palytoxin-induced Na/K pump channels and modulation of palytoxin interaction by Na/K pump ligands. J Gen Physiol 123: 357376, 2004. 5. Caterina MV and Julius D. The vanilloid receptor: a molecular gateway to the pain pathway. Annu Rev Neurosci 24: 487517, 2001. 6. Chuang HH, Prescott ED, Kong H, Shields S, Jordl SE, Basbaum AI, Chao MV, and Julius D. Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns(4,5)P2-mediated inhibition. Nature 411: 957962, 2001. 7. Cornelius F and Mahmmoud YA. Functional modulation of the sodium pump: The regulatory proteins Fixit. News Physiol Sci 18: 119124, 2003.
28 DECEMBER 2004
Report
154 NEW PERSPECTIVES ON ION TRANSPORT PROTEINS roid hormone and second messengers by acting at multiple sites in opossum kidney cells. Mol Endocrinol 17: 23552364, 2003. Mahon MJ, Donowitz M, Yun CC, and Segre GV. Na/H exchanger regulatory factor 2 directs parathyroid hormone 1 receptor signaling. Nature 417: 858861, 2002. Mahon MJ and Segre GV. Stimulation by parathyroid hormone of a NHERF-1-assembled complex consisting of the parathryroid hormone 1 receptor, phospholipase C, and actin increases intracellular calcium in opossum kidney cells. J Biol Chem 279: 2355023558, 2004. Moore RE and Scheuer PJ. Palytoxin: a new marine toxin from coelenterate. Science 172: 495498, 1971. Murumatsu I, Uemura D, Fujiwara M, and Narahashi T. Characteristics of palytoxin-induced depolarization in squid giant axons. J Pharmacol Exp Ther 231: 488491, 1984. Noury C, Grant SGN, and Borg JP. PDZ domain proteins: plug and play! Science-STKE 179-RE7: 112, 2003. Ogawa H and Toyoshima C. Homology modeling of the cation binding sites of the Na,K-ATPase. Proc Natl Aacd Sci USA 99: 1597715982, 2002. ONeil RG and Brown RC. The vanilloid receptor family of calciumpermeable channels: molecular integrators of microenvironmental stimuli. News Physiol Sci 18: 226231, 2003. Prescott ED and Julius D. A modular PIP2 binding site as a determinant of capsaicin receptor sensitivity. Science 300: 12841288, 2003. Raghuram V, Mak DD, and Foskett JK. Regulation of cystic brosis transmembrane conductance regulator single-channel gating by bivalent PDZ-domain-mediated interaction. Proc Natl Acad Sci USA 98: 1300 1305, 2001. Scheiner-Bobis G and Schneider H. Palytoxin-induced channel formation within the Na,K-ATPase does not require a catalytically active enzyme. Eur J Biochem 248: 717723. Shenolikar S, Voltz JW, Minkoff CM, Wade JB, and Weinman EJ. Targeted disruption of the mouse NHERF-1 gene promotes internalization of proximal tubule sodium-phosphate cotransporter type IIa and renal phosphate wasting. Proc Natl Aacd Sci USA 99: 1147011475, 2002. Shenolikar S and Weinman EJ. NHERF: targeting and trafcking membrane proteins. Am J Physiol Renal Physiol 280: F389F395, 2001. Toyoshima C, Nakasake M, and Nomura H, and Ogawa H. Crystal structure of the calcium pump of sarcoplasmic reticulum at 2.6 A resolution. Nature 405: 647655, 2000. Toyoshima C and Nomura H. Stuctural changes in the calcium pump accompanying dissociation of calcium. Nature 418: 605611, 2002. Voltz JW, Weinman EJ, and Shenolikar S. Expanding the role of NHERF, a PDZ-domain containing protein adapter to growth regulation. Oncogene 20: 63096314, 2001. Wang S, Yue H, Derin RB, Guggino WB, and Li M. Accessory protein facilitated CFTR-CFTR interaction, a molecular mechanism to potentiate the chloride channel activity. Cell 103: 169179, 2000. Weinman EJ, Steplock D, Wade JB, and Shenolikar S. Ezrin binding domain-decient NHERF attenuates cAMP-mediated inhibition of Na/H exchange in OK cells. Am J Physiol Renal Physiol 281: F374 F380, 2001.
8. Crambert G and Geering K. FXYD proteins: New tissue-specic regulators of the ubiquitous Na,K-ATPase. Science-STKE 166-RE1: 19, 2003. 9. Cunningham R, Stelock D, Wang F, Huang H, Xiaofei E, Shenolikar S, and Weinman EJ. Defective parathyroid hormone regulation of NHE3 activity and phosphate adaptation in cultured NHERF1-/- renal proximal tubule cells. J Biol Chem 279: 3781537821, 2004. 10. Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT, MacKinnon R. The structure of the potassium channel: molecular basis of K conduction and selectivity. Science 280: 6977, 1998. 11. Dranseld DT, Bradford AJ, Smith J, Martin M, Roy C, Mangeat PH, and Goldenring JR. Ezrin is a cyclic AMP-dependent protein kinase anchoring protein. EMBO J 16: 3543, 1997. 12. Dutzler R, Campbell EB, Cadene M, Chait BT, and MacKinnon R. X-ray structure of a ClC chloride channel at 3.0 Ao reveals the molecular basis of anion selectivity. Nature 415: 287294, 2002. 13. Estevez R, Schroeder BC, Accardi A, Jentsch TJ, and Pusch M. Conservation of chloride channel structure revealed by an inhibitor binding site in ClC-1. Neuron 38: 4759, 2003. 14. Gadsby DC. Spot the difference. Nature 427: 795796, 2004. 15. Guennoun S and Horisberger JD. Cysteine scanning mutagenesis study of the 6th transmembrane segment of the Na, K-ATPase subunit: a cysteine-scanning mutagenesis study. FEBS Lett 513: 277288, 2000. 16. Guennoun S and Horisberger JD. Structure of the 5th transmembrane segment of the Na, K-ATPase subunit: a cysteine-scanning mutagenesis study. FEBS Lett 482: 144148, 2000. 17. Habermann E and Chhatwal GS. Ouabain inhibits the increase due to palytoxin of cation permeability of erythrocytes. Naunyn Schmiedebergs Arch Pharmacol 319: 101107, 1982. 18. Hall RA, Premont RT, Chow CW, Blitzer JT, Pitcher JA, Claing A, Stoffel RH, Barak LS, Shenolikar S, Weinman EJ, Grinstein S, and Lefkowitz RJ. The beta2-adrenergic receptor interacts with the Na/Hexchanger regulatory factor to control Na/H exchange. Nature 392: 626630, 1998. 19. Hernando N, Wagner CA, Gisler SM, Biber J, and Murer H. PDZ proteins and proximal ion transport. Curr Opin Nephrol Hypertens 13: 569574, 2004. 20. Hilgemann DW. From pump to a pore: how palytoxin opens the gates. Proc Natl Aacd Sci USA 100: 386388, 2003. 21. Hilgemann DW, Feng S, and Nasuhoglu C. The complex and intriguing lives of PIP2 with ion channels and transporters. Science-STKE 111-RE19: 18, 2001. 22. Horisberger JD, Kharoubi-Hess S, Guennoun S, and Michielin O. The fourth transmembrane segment of the Na, K-ATPase subunita systematic mutagenesis study. J Biol Chem 279: 2954229550, 2004. 23. Kockskamper J, Ahmmed GU, Zima AV, Sheehan KA, Glitsch HG, and Blatter LA. Palytoxin disrupst cardiac excitation-contraction coupling through interactions with P-type ion pumps. Am J Physiol Cell Physiol 287: C527C538, 2004. 24. Mahon MJ, Cole JA, Lederer ED, and Segre GV. Na/H exchanger regulatory factor 1 mediates inhibition of phosphate transport by parathy-
25. 26.
27. 28. 29. 30. 31. 32. 33.
34. 35.
36. 37. 38. 39. 40. 41.
28 DECEMBER 2004

Resting Potential

Enviado por

Dados do documento

Descrição original:

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Resting Potential

Enviado por

Direitos autorais:

Formatos disponíveis

Adv Physiol Educ 28: 139142, 2004; doi:10.1152/advan.00029.2004.

Generation of resting membrane potential

Downloaded from ajpadvan.physiology.org on November 16, 2007

1043-4046/04 $5.00 Copyright 2004 The American Physiological Society

Table 1. Cytoplasmic and extracellular K, Na, Cl concentrations

1107 1109 1108

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Adv Physiol Educ 28: 143154, 2004; doi:10.1152/advan.00046.2004.

Ion homeostasis, channels, and transporters: an update on cellular mechanisms

ION TRANSPORT PROTEINS

Downloaded from ajpadvan.physiology.org on November 16, 2007

1043-4046/04 $5.00 Copyright 2004 The American Physiological Society

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

Downloaded from ajpadvan.physiology.org on November 16, 2007

FXYD1 (PLM) FXYD2 FXYD4 (CHIF) FXYD7

Heart and muscle Kidney Kidney and colon Brain

Downloaded from ajpadvan.physiology.org on November 16, 2007

Advances in Physiology Education VOL

27. 28. 29. 30. 31. 32. 33.

Downloaded from ajpadvan.physiology.org on November 16, 2007

36. 37. 38. 39. 40. 41.

Advances in Physiology Education VOL

Você também pode gostar