|
|
|
|
|
|
|
||
Original Article |
Zero and Low K+ Conditions
b Departamento de Reconocimiento Molecular y Biologia Estructural, Instituto de Biotecnología, UNAM, Cuernavaca, Morelos 62250, México
Departamento de Fisiología, Facultad de Medicina, UNAM, Universitaria, México City 04510, México.52-73-172388
froylan{at}ibt.unam.mx
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
µNa. Finally, when XNa,i = 1, GK collapses by the binding of Nai+ to two sites, with apparent Kds of 2 and 14.3 mM.
Key Words: K+ affinity • Na+ block • conductance • selectivity • zero K+
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Almers and Armstrong 1980; Khodakhah et al. 1997). On the other hand, both DR channels of bullfrog sympathetic neurons (Block and Jones 1997) as well as mammalian DRs, of the Kv2.1 subfamily, remain active in 0 K+, allowing a substantial permeation of Na+, but becoming anomalously immune to the addition of TEA (Zhu, and Ikeda 1993; Ikeda and Korn 1995). Similarly, Kv1.5 channels also remain active in 0 K+, although they only allow a small, fast inactivating, and TEA-sensitive, flux of Na+ through them (Zhuren et al. 2000). In contrast, other K+ channels are apparently insensitive to K+ depletion. For example, Kv1.3 channels, do not allow a measurable permeation of Na+, and remain active after the exposure to 0 K+ (Immke et al. 1998; see also Jäger et al. 1998).
Recently, it was shown that in the absence of K+ on both sides of the membrane, the Shaker B K+ conductance (GK) collapses when the membrane is depolarized and then repolarized. Briefly, Gómez-Lagunas 1997, Gómez-Lagunas 1999 reported the following. First, GK collapses if the channels are gated, by the delivery of standard activating pulses, while they are immersed in Na+-containing, 0 K+, solutions. The extent of collapse depends on the number of pulses but not on their frequency. Second, in contrast, the halt in the conductance is completely prevented if the channels are kept closed (are not gated) while they are in 0 K+. Third, depolarized holding potentials (HP, above –50 mV) impede the drop of GK. Fourth, external K+ protects GK by binding to a site in the channels with rather low, millimolar, affinity (apparent Kd = 2.9 mM). Fifth, among divalent cations only Ba2+ is able to replace K+, protecting GK; and finally, very prolonged depolarizations (seconds to minutes) recover the, previously lost, K+ conductance.
The above observations were interpreted as meaning that closing without K+ sinks the channels into a stable nonconducting, noninactivated, closed state(s) (Gómez-Lagunas 1997). It seems that the channels need to close with a K+ ion(s) bound to them, otherwise they sink into a reluctant conformation where they remain unable to conduct K+, no matter how long Vm is kept at the HP or hyperpolarized potentials (Gómez-Lagunas 1997, Gómez-Lagunas 1999; Melishchuk et al. 1998). Looking at the selectivity of the site(s) involved in the drop of GK, preliminary observations have indicated that the presence of Na+ ions might significantly foster the collapse of GK in 0-K+ solutions, probably by promoting the displacement of K+ from the pertinent site(s) in the channels (Gómez-Lagunas 1997; Melishchuk et al. 1998). Thus, to further understand the mechanism underlying the fall of GK, in this work the role of Na+ ions in both the collapse of the Shaker K+ conductance in 0-K+ solutions, and in the behavior of the channels in low K+, is investigated.
Here, it is shown that Na+ ions foster the collapse of GK in the absence of K+. External K+ and Na+ compete, in a mutually exclusive manner, for binding to a externally located site (probably external to the selectivity filter), where GK is modulated (available versus reluctant). On the other hand, to see if the collapse of GK occurs as a discontinuous change in the properties of the channels in the limit of zero K+, experiments with internal solutions containing both K+ and Na+ ions were done. The results show that as the molar fraction of Nai+ (XNa,i) increases, the energy required to relieve Nai+ block changes in an extent not accounted for by the change in the driving force of Na+. This suggests that, as XNa,i increases, there is either an increased electrostatic repulsion between the blocking Na+ and a neighbor ion, or, more likely, a deformation (change of the energy profile) of the pore, that could be maximal in the limit XNa,i = 1. Under the latter conditions, GK collapses, and Nai+ interacts with the channels with a kinetics described by the noncooperative binding to two, kinetically distinguishable, sites.
A preliminary account of this work has been presented in abstract form (Gómez-Lagunas 2000).
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Gómez-Lagunas 1997). Experiments were performed 48 h after infection.
Electrophysiology
Macroscopic currents were recorded under whole-cell patch clamp, with an Axopatch-1D (Axon Instruments). The currents were filtered at 5 KHz with the built in filter of the amplifier, and sampled at 100 µs/point, with a TL1 interface (Axon Instruments). Electrodes were pulled from borosilicate glass (KIMAX 51) to a final resistance of 1–2 M
; 
80% of the series resistance was electronically compensated. Activating pulses (unless otherwise indicated = +20 mV/30 ms) in 0-K+ solutions, were delivered at 1 Hz. This procedure will be referred to as pulsing.
Solutions
Solutions will be named by their main cation and represented as external/internal (e.g., Ko/NMGi). Their composition is listed in Table . All other solutions in which the concentration of the test cation exceeded 5 mM, were made by the appropriate mixing of the listed solutions, keeping the osmolarity constant. Total exchange of the external solution was achieved in at most 15 s. Pulsing in 0 K+ was performed under continuous perfusion, beginning 1 min after the start of the perfusion.
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
|
Fig. 1 C compares the reduction of GK after pulsing in a variety of solutions. Notice that, whereas in Nao/Nai (last bar) GK is basically erased (99 ± 1% reduction, n = 4), in NMGo/NMGi (second bar) GK drops only 20 ± 2% (n = 16). Thus, pulsing in 0 Na+, NMG-containing, solutions collapses GK, but far less than in the presence of Na+ ions (in fact, in Nao/Nai, only 10–15 pulses are needed to completely eliminate GK, see Gómez-Lagunas 1997, Gómez-Lagunas 1999). Moreover, after pulsing in NMG most of the channels (
80%) remain in a state from which they readily collapse upon the addition of Na+ (not shown). The third bar in Fig. 1 C, shows that external choline, another impermeant and nonblocking cation, is as inert as NMG (GK drop = 21 ± 6%, n = 5; see DISCUSSION).
In summary, the combined condition, presence of Na+ and lack of K+, makes GK more liable to collapse. Finally, notice that pulsing with Na+ ions present (added) in only the external solution (in Nao/NMGi, first bar) drops GK in about the same extent (90 ± 4%, n = 9) as it does with Na+ on both sides of the membrane (99 ± 1%, last bar). So, even when under physiological conditions, external Na+ neither permeates nor blocks Kv channels (but see Block and Jones 1996), it effectively fosters the collapse of GK in 0-K+ solutions.
External Na+ Interaction with Shaker Channels
The interaction of external Na+ with Shaker was further studied by looking at the extent of GK collapse produced by pulsing in solutions of variable [Nao+], with NMGi as the internal solution (in (NMGo + [Nao])/NMGi, see MATERIALS AND METHODS). Fig. 2 A shows that as [Nao+] increases, GK drops following a Hill saturation curve (line through the points, labeled 0 K+), with a Kd = 3.3 mM, and maximal extent of collapse = 1.05 (Hill number n = 0.97). When the same measurements are done in the presence of either 0.08 or 0.3 mM Ko+, it is seen that the apparent Kd for Na+ increases to either 4.7 or 5.3 mM, respectively, without a significant change in the maximal extent of collapse (0.98 or 0.91 with 0.08 or 0.3 mm K+, respectively). This indicates that K+ protection (reduction of GK drop) is more efficient at the lower [Nao+], and thus shows that K+ inhibits in a competitive manner the binding of Na+, to the site where GK is modulated. The inset shows the plot in an expanded [Nao+] scale.
|
), as (Segel 1993): Kd(Ko+) Kapp(Ko+)/(1 + ([I]/Ki)) = 80 µM.
However, it is pertinent to mention that considering the possibility of ion permeation through the channels, deviations from the equilibrium conditions, on which the saturation Michaelis-Menten or Hill equations are based (Segel 1993), might be expected, particularly at the higher [K+], so both the K+ and Na+ Kds are likely to be overestimations of their true equilibrium values.
If the site where Nao+ (and Ko+) binds was located within the electric field of the membrane, it would be expected that pulsing from a hyperpolarized HP should increase the drop of GK, by favoring the Nao+ occupancy of the site.
In contrast to the above prediction, Fig. 2 C shows that, with a nonsaturating [Nao+] = 2.5 mM, the drop of GK is slightly, although significantly, reduced (P < 0.05), instead of increased, by pulsing from a more negative HP (from 49 ± 4% [n = 7] at HP –80 mV; to 37 ± 1% [n = 6] at HP = –120 mV). On the other hand, Fig. 2 D shows that within the range of 0 to +60 mV, the pulse potential (Vp) does not play a significant role on the extent of collapse. The dependence on holding potential of Nao+ action is qualitatively equivalent to that of Bao2+ protection (Gómez-Lagunas 1999). The latter decreases by pulsing from hyperpolarized HPs, and, like Nao+ action, it is not dependent on Vp. The following nonexcluding possibilities had been proposed as an explanation of this voltage dependence (Gómez-Lagunas 1999): (1) hyperpolarized HPs might promote ion (Nao+ or Ba2+) permeation through the channels, thus reducing the strength of their corresponding effects; or (2) there could be more than one site where the state of GK is modulated, a negative HP might favor the occupancy of innermost and less effective site, in terms of the corresponding effect of the ions (see Fig. 6). Whatever the case, the effect of voltage on Nao+ and Bao2+ actions, along with the previous finding that the potency with which ions other than K+ protect GK, is not related to the permeability sequence of the channels (Gómez-Lagunas 1997, Gómez-Lagunas 1999) suggests that the external site, whose state of occupancy determines the state of GK (available or reluctant), may not be part of the selectivity filter, but instead that it might be externally located to it (see DISCUSSION).
|
; French and Wells 1977; Marty 1983; Yellen 1984; Neyton and Pelleschi 1991; Starkus et al. 2000). To see if the collapse of GK occurs as a discontinuous change only observed in the limit of 0 K+ (molar fraction of Na+, XNa = 1), or if there is a gradual change in the properties of the channels as XNa increases, the behavior of the channels as the molar fraction of Nai+ (XNa,i) approached 1 was studied, in internal solutions where XNa,i + XK,i = 1.
For a reference, Fig. 3 A shows a typical I-V relationship obtained under standard conditions (Nao/Ki). Above 0 mV, where the probability of opening is 1, IK increases linearly with the voltage. With the XNa,i = 0, this is always observed. An I-V obtained with a low XNa,i = 0.21 is shown in Fig. 3 B. First IK increases with the voltage, but at about +40 mV, the current starts to deviate from the linearity and after that a region of negative conductance develops, as Nai+ blocks the channels. The departure from linearity, at positive voltages (
+40 mV), is more clearly seen by comparing the experimental points with the straight line in the plot. The latter is the least-square fit of the points between 0 and +30 mV, and therefore gives an estimate of the expected Ik if XNa,i was zero. Thus, the difference between the line and the points in the graph is an estimate of the extent of block at each voltage, the latter is plotted in Fig. 3 C, which presents the mean ± SEM of four cells. The line is the fit of the points with a Woodhull equation (Fig. 3 legend), with electrical distance 
= 0.7, and Kd(0 mV) = 538.3 mM (Woodhull 1973). A representative I-V (n = 6 cells) obtained with a higher XNa,i = 0.42 is shown in Fig. 3 D. See that, qualitatively, it looks like that with XNa,i = 0.21. With the Nao solution, and XNa,i up to 0.42, Nai+ block becomes continuously stronger as the voltage is made more positive, for voltages up to +160 mV.
|
0.42, Fig. 4 A shows that with XNa,i = 0.71, the I-V relationship acquires an N shape (French and Wells 1977; Starkus et al. 2000) within the voltage range Vp +160 mV. IK first increases, but starts to deviate from the linearity as Nai+ blocks the channels, and thereafter a region of negative conductance develops, until a "critical" voltage (Vc) is reached (in this case Vc = +70 mV) where Nai+ block is relieved and the current increases again with the voltage (pointed out by the arrow), giving the I-V its N shape. The traces in the right panel of Fig. 4 A illustrate the N shape of the I-V relationship. See that the peak currents are in the following order: I (+110 mV) > I (+30 mV) > I (+70 mV). Finally, Fig. 4 B shows an I-V obtained with XNa,i = 0.83. Notice that although its overall shape is like that observed with XNa,i = 0.71, Vc is smaller (Vc = +50 mV). The traces in the right panel illustrate the N shape of the I-V relationship.
|
). From the value of Vc, the minimal energy required to unblock the channels (BNa) can be estimated as follows (Latorre and Miller 1983): BNa = |–FVc|, where F is the Faraday constant.
Fig. 5 A compares BNa at the two XNa,i where a corresponding Vc could be reached within the range Vp +160 mV. In going from a XNa,i of 0.71 to 0.83, BNa changes from 8.4 ± 0.3 (n = 5) to 5.5 ± 0.5 kJ/mol (n = 5). On the other hand, the electrochemical gradient of Na+ (µNa), evaluated at Vc, µNa = F(Vc – VNa), where VNa is the Nernst potential of Na+, changes from 5.5 (XNa,i = 0.71) to 4.0 kJ/mol (XNa,i = 0.83). So, whereas BNa decreases 2.9 kJ/mol, µNa decreases only 1.5 kJ/mol, this indicates that the change in Vc, with XNa,i, cannot be entirely accounted for by the change in the driving force of Na+.
|
µNa is also seen when Fig. 3 and Fig. 4 are compared. For example, with XNa,i = 0.42 the I-V is not N-shaped, so Vc +160 mV, then when XNa,i increases to for example 0.83, BNa changes from at least 15.5 (XNa,i = 0.42, underestimating Vc as +160 mV) to 5.5 ± 0.5 kJ/mol (XNa,i = 0.83) that is 10 kJ/mol, whereas µNa changes from at least 12.9 to 4.0 kJ/mol, that is 8.9 kJ/mol.
By looking at Nai+ block of the squid K channel, N-shaped I-Vs were first observed by French and Wells (French and Wells 1977), whom after observing that block was overcome by very positive potentials, proposed that the relative permeability of Na+ to that of K+ (PNa/PK) increased at those voltages (more than +160 mV). This proposal has received support from a recent study of the change of the PNa/PK ratio of Shaker channels, during the onset of slow inactivation (Starkus et al. 2000). In the simple case where a single, rate limiting, barrier B limits ion permeation through a channel, the relative permeability of a test ion (Na+) to that of a reference ion (K+) can be estimated with the relation (Reuter and Stevens 1980; Latorre and Miller 1983): PNa/PK = exp[(BNa – BK)/RT], where R, T have their usual meanings.
Following this simplifying approach, and taking BK = 0, and BNa as in Fig. 5 A, it is seen that when XNa,i increases from 0.71 to 0.83 (1.2 times), PNa/PK increases about four times (Fig. 5 B). The results in Fig. 3Fig. 4Fig. 5 show that as XNa,i increases there is an actual reduction (i.e., a reduction not entirely accounted for by the change in µNa) of the energy required to relieve Nai+ block, yielding a substantial change of the PNa/PK ratio.
The above effect could be explained as the result of an increase in the electrostatic repulsion between the blocking Na+ and a neighbor ion, as XNa increases. However, it is not easy to see why this repulsion should be bigger as the molar fraction of the permeant K+ ion decreases. Alternatively, it could be that as XNa,i increases, there is a deformation of the pore, measured as a reduction of the energy required to allow Na+ permeation. If that were the case, it could be that this deformation were maximal in the limit where XNa,i = 1. Pulsing under the latter conditions (Nao/Nai) collapses GK (Gómez-Lagunas 1997). Finally, as aforementioned (Fig. 2), the voltage dependence of external Na+ and Ba2+ actions, as well as that of internal Ba2+ (Gómez-Lagunas 1999), might be explained by the involvement of two sites in the drop of GK; thus, to explore this point, the role of internal Na+ on the fall of GK, when XK = 0, was studied.
Fig. 6 A shows that pulsing with Na+ present (added) in only the internal solution (NMGo/Nai), drops GK in about the same extent (93 ± 5%, n = 6) as it does with Na+ present on both sides of the membrane (last bar, 99 ± 1%, n = 4). For comparison, the figure also shows the collapse in NMGo/NMGi, as in Fig. 1. The above observation suggests that Na+ could possibly act on the same site regardless of the side of the membrane from where it comes. To further explore this point the collapse of GK as a function of [Nai+] (in NMGo/(NMGi-[Nai])) was determined. Fig. 6 B shows that as [Nai+] increases GK drops following a Hill curve (line through the points) with n = 1.16 (which indicates that there is, at least, one site where Nai+ binds) plus an offset, that accounts for the collapse in NMGo/NMGi (0.20, Fig. 6 A). However, Fig. 6 C shows that the double-reciprocal plot of the points in B is best fitted by two straight lines, which indicates that Nai+ binds to two sites: one with a maximal collapse (m.c) of 60% and Kd = 2 mM, and a second, lower affinity site, with m.c = 36% and Kd = 14.3 mM. The presence of two sites is also clearly distinguished in the inset, which presents the Eadie-Scatchard plot of the points in B (Segel 1993) (see DISCUSSION). The voltage dependence of Nai+ action is presented in Fig. 6 D, which shows that with a nonsaturating [Nai+] = 10 mM, increasing Vp from +20 to +70 mV significantly increases (P < 0.05) the drop of GK: as if promoting the Na+ occupancy of the most externally located site, fostered the fall of GK. On the other hand, the HP during pulsing (–80 vs. –120 mV), does not exert a significant effect on the extent of collapse (not shown).
It is pertinent to mention that by contrast to the qualitative similarity of the voltage dependence of the Nao+ and Bao2+ effects, the voltage dependence of Nai+ and Bai2+ actions are different. The latter decreases markedly with hyperpolarized HPs, and it is not dependent on Vp (Gómez-Lagunas 1999). Further work is needed to understand this difference.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, Gómez-Lagunas 1999; Melishchuk et al. 1998). Prolonged depolarizations are needed to recover GK (Gómez-Lagunas 1997). The drop of GK seems to involve major changes in the conformation of the channels including not only the pore (Gómez-Lagunas 1997), but also the voltage sensor (Melishchuk et al. 1998). So it could be that recovery is so slow due to the stability of the collapsed conformation, and to the number of rearrangements needed to reset the channels in their normal conducting state. Here, it was shown that the combined condition 0 K+ and presence of Na+ makes GK more liable to collapse, as Na+ ions speed (in number of pulses) the drop of GK.
Nonetheless, Na+ is not necessary for GK to collapse, as it also falls in NMG or choline solutions. Considering that Na+ binds with millimolar affinity from both sides of the membrane, it seems unlikely that the drop of GK in the 0-Na+ solutions, could be produced by contaminant Na+ ions. More likely, the binding of Na+, instead of K+, or just the absence of K+ in the pertinent site(s) when the channels close, elicits the drop of GK, although with significantly different speeds that might be accounted for by a smaller dwelling time of K+, in the pertinent site(s), in the presence of Na+ ions than in their absence, in agreement with the observed mutually exclusive binding of K+ and Na+ ions to the externally located site.
Extracellular Na+ and the Drop of GK
Under physiological conditions, Nao+ neither permeates nor blocks Kv channels (but see Block and Jones 1996), however, it effectively fosters the drop of GK, by binding to a site with an apparent Kd = 3.3 mM. External Na+ and K+ compete, in a mutually exclusive manner, for binding to this site, and from this observation a high, micromolar, affinity for Ko+ was estimated, Kd = 80 µM. The presence of high affinity binding sites for K+ in the selectivity filter of the channels, from where, under physiological conditions, Na+ is excluded because of its lower affinity, has been shown to underlie the structural basis of the mechanism of selectivity of K channels (e.g., Korn and Ikeda 1995; Doyle et al. 1998; Ogielska and Aldrich 1998).
Could the site where Ko+ and Nao+ bind modulating GK be located within the selectivity filter? The protection exerted by TEAo suggests that the site is externally located (Gómez-Lagunas 1997), as the selectivity filter is (Doyle et al. 1998). Nonetheless, the voltage dependence of Nao+, Bao2+, and Ko+ actions (Gómez-Lagunas 1999; Fig. 2), along with the observation that the potency with which monovalent cations protect is not related to the selectivity of the channels (Gómez-Lagunas 1997), together suggest that this site could be external to the selectivity filter.
However, considering that in multi-ion pores, the voltage dependence of ligand binding depends on ion occupancy (Neyton and Miller 1988a,Neyton and Miller 1988b; Spassova and Lu 1999; Thompson and Begenisich 2001), and that selectivity of Kv channels depends on the [K+] (Korn and Ikeda 1995; Ogielska and Aldrich 1998; Fig. 5), the selectivity filter cannot yet be ruled out as the externally located place where GK is modulated (Melishchuk et al. 1998). Although, the drop of GK with [Nao+] follows a Hill equation with n = 0.97 and a linear double reciprocal plot, the reduced effectiveness of Nao+ at hyperpolarized HPs, suggested the involvement of more than one site in the collapse of GK. This possibility was reinforced by the kinetics of the drop of GK with [Nai+] (see below).
Internal Na+ Interaction with Shaker
For voltages up to +160 mV, I-V relationships obtained with the Nao solution, and with a variable XNa,i, have a shape that depends on XNa,i. With XNa,i 0.71, the I-Vs are N-shaped. In contrast to the behavior of Shaker channels, French and Wells 1977 did not find a significant change of Vc with the internal Na+ in the squid K channel. It would be interesting to determine if the differential dependence of the I-V relationships on XNa,i of these channels, keeps any relation to their different behavior in 0-K+ solutions (Almers and Armstrong 1980; Khodakhah et al. 1997; Gómez-Lagunas 1997, Gómez-Lagunas 1999).
Recently, by looking at the onset of the slow inactivation of Shaker channels lacking the NH2 terminus domain, Starkus et al. 2000 afforded evidence showing that once Vc is reached and the conductance becomes again positive, Na+ actually flows throughout the channels, carrying current from the internal to the external solution. This supports the proposal of French and Wells 1977 that selectivity is voltage-dependent. The results presented here show that the Shaker selectivity also varies with XNa,i. It is pertinent to point out that an analogous statement has been made for the squid Na+ channel (Cahalan and Begenisich 1976).
It has been shown that mammalian Kv2.1 channels as well as DR channels of bullfrog neurons although remain stable in 0 K+, conducting Na+, undergo a conformational change of the pore region, which is observed as a reduction of the capability of TEA to block the channels (Korn and Ikeda 1995; Block and Jones 1997). Likewise, it seems that as XNa,i increases there is a deformation of the Shaker pore, which is observed as a reduction of the energy barrier that limits Nai+ efflux. That deformation could be maximal when XK = 0, but further studies are needed to determine its relation with the collapse of GK here studied.
When XK,i = 0, pulsing with Na+ ions added to the internal solution alone drops GK in about the same extent as it does with Na+ on both sides of the membrane, suggesting that Nai+ might be able to reach the externally located site where GK is modulated, in agreement with the observation that Vc decreases as XNa,i approaches one, and with reports showing Na+ currents through Kv channels in 0 K+ (Zhu and Ikeda 1993; Korn and Ikeda 1995; Block and Jones 1997; Starkus et al., 1997, 2000; Melishchuk et al. 1998; Ogielska and Aldrich 1998; Zhuren et al. 2000). On the other hand, the [Nai+] dependence of GK drop indicates that there are actually two binding sites, of which the one with the higher affinity (Kd = 2 mM) could be the externally located site (Kd = 3 mM). It might be that presence of two sites could form the structural basis of the XNa,i dependence of Vc. Whatever the case, it is pertinent to mention that the possible involvement of two sites in the collapse of GK was previously suggested by the characteristics of the protection afforded by internal Ba2+ (Gómez-Lagunas 1999).
|
ACKNOWLEDGMENTS |
|---|
This work was supported by Dirección General de Asuntos del Personal Academico (IN-216900) and Consejo Nacional de Ciencia y Tecnología grants 26525N and Z-005.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
) or 0.3 mM Ko+ (
). The lines are the fit of the points with a Hill equation with parameters (•): n = 0.97, Kd = 3.3 mM, and maximal collapse (m.c) = 1.05; and (
