The Na+/H+ and K+/H+ exchange pathways of Amphiuma tridactylum red blood cells (RBCs) are quiescent at normal resting cell volume yet are selectively activated in response to cell shrinkage and swelling, respectively. These alkali metal/H+ exchangers are activated by net kinase activity and deactivated by net phosphatase activity. We employed relaxation kinetic analyses to gain insight into the basis for coordinated control of these volume regulatory ion flux pathways. This approach enabled us to develop a model explaining how phosphorylation/dephosphorylation-dependent events control and coordinate the activity of the Na+/H+ and K+/H+ exchangers around the cell volume set point. We found that the transition between initial and final steady state for both activation and deactivation of the volume-induced Na+/H+ and K+/H+ exchange pathways in Amphiuma RBCs proceed as a single exponential function of time. The rate of Na+/H+ exchange activation increases with cell shrinkage, whereas the rate of Na+/H+ exchange deactivation increases as preshrunken cells are progressively swollen. Similarly, the rate of K+/H+ exchange activation increases with cell swelling, whereas the rate of K+/H+ exchange deactivation increases as preswollen cells are progressively shrunken. We propose a model in which the activities of the controlling kinases and phosphatases are volume sensitive and reciprocally regulated. Briefly, the activity of each kinase-phosphatase pair is reciprocally related, as a function of volume, and the volume sensitivities of kinases and phosphatases controlling K+/H+ exchange are reciprocally related to those controlling Na+/H+ exchange.
- ion transport
volume activated inorganic ion flux pathways that mediate cell volume regulation are well documented in a wide variety of cell types (for review, see Refs. 5, 15, 16, 21, 23, 30) although the biochemical processes controlling the activity of these pathways are poorly understood (8, 9, 12, 15, 16, 22, 33). Kinetic manifestations of the volume-sensitive biochemical control processes have been described by several investigators (2, 10, 17–19, 24, 25, 27). For example, transition times ranging from 5 to 25 min have been reported for activation of volume-sensitive ion flux pathways in red blood cells (RBCs) from an array of vertebrate species (10, 18, 20, 27, 34). Delays in activation of these pathways reflect the temporal characteristics of the volume-sensitive biochemical processes responsible for transporter regulation. Consequently, careful analysis of the time course of activation and deactivation of volume-sensitive pathways has proven to be a useful means of describing the rate-limiting processes underlying volume-dependent control of transport activity.
The first systematic quantitative analysis of the kinetics of activation and deactivation of a volume-sensitive transport system was performed in 1990 by Jennings and Al-Rohil (18), who proposed a two-state model for swelling-induced activation and shrinkage-induced deactivation of K+-Cl− cotransport in rabbit RBCs. This approach was prompted by their observation that both activation and deactivation of K+-Cl− cotransport proceed as a monotonic single exponential function of time, i.e., there is a single rate-limiting step for both activation and deactivation. According to the model, K+-Cl− cotransport activity is low in the resting state (in isosmotic medium) yet increases in proportion to the extent of cell swelling in hyposmotic media. The transition between resting and active states is characterized by unimolecular rate constants [forward (k12) and reverse (k21)] as shown in the model and summarized in Eq. 1 (below): (1) where U is the unidirectional uptake of K+ (86Rb+) at time t in millimoles ion per kilogram dry cell solid (dcs). Ji is the uptake rate (mmol ion·kg−1 dcs·min−1) in the initial steady state prior to cell volume perturbation, and Jf is the final steady-state uptake rate in response to cell volume perturbation. The relaxation time (τ) is the time required to make the transition from the initial to the final steady state and is equal to the reciprocal of the sum of the forward and reverse rate constants [(k12 + k21)−1]. Upon inspection of the model, it is clear that to make the transition from the resting state to active state, the ratio k12/k21 must increase. An increase in k12/k21 is achieved either by an increase in k12 or by a decrease in k21. These two scenarios have very different implications with regard to τ. If activation is solely due to an increase in k12, then τ will decrease as the stimulus increases and higher levels of activation are achieved. In contrast, if activation is solely the result of a decrease in k21, then increased activity will be accompanied by progressively longer relaxation times since the sum of k12 + k21 must decrease, and therefore τ [(k12 + k21)−1] must increase. Reciprocally, it follows that if activation proceeds as a result of an increase in k12 then deactivation will result from a decrease in k12 and τ will increase proportionate to the deactivating stimulus. Conversely, if activation is achieved by a decrease in k21, then deactivation must result from an increase in k21 and τ will decrease in proportion to the deactivating stimulus.
Employing the above model, Jennings and Al-Rohil (18) determined that τ for swelling-dependent activation of K+-Cl− cotransport was longer than that for K+-Cl− cotransport deactivation (when preswollen cells were acutely restored to normal volume by osmotic shrinkage). These results imply that subsequent to swelling (transport activation), the sum of k12 + k21 must be smaller than the sum of k12 + k21 upon shrinkage of preswollen cells (transport deactivation). Because cell swelling activates K+-Cl− cotransport, k12/k21 must increase yet the relatively slow rate of activation reflects a decrease in k12 + k21. These investigators reasoned that a simultaneous increase in k12/k21 and decrease in k12 + k21 occurs during activation only if k21 decreases with little or no change in k12. Furthermore, they found that protein phosphatase inhibitors slowed the rate of swelling-induced activation of K+-Cl− cotransport and concluded that k12 corresponds to net phosphatase activity (which decreases with swelling), whereas k21 corresponds to net kinase activity (unchanged in response to swelling). Consequently, swelling-induced activation of K+-Cl− cotransport in rabbit RBCs is the result of a single rate-limiting step: swelling-induced decrease in the activity of a kinase that deactivates K+-Cl− cotransport (18).
Inspired by Jennings and Al-Rohil (18), two groups of investigators performed similar studies in RBCs from other mammalian species. Parker and coworkers (27) applied the model in Eq. 1 to the analyses of shrinkage-activated Na+/H+ exchange and swelling-activated K+-Cl− cotransport in dog RBCs. Their findings led them to conclude that net phosphorylation in response to cell shrinkage is responsible for activation of Na+/H+ exchange and deactivation of K+-Cl− cotransport. Conversely, net dephosphorylation events subsequent to cell swelling are responsible for activation of K+-Cl− cotransport and deactivation of Na+/H+ exchange. In agreement with the findings of Jennings and Al-Rohil, Parker and coworkers concluded that only the kinase is volume sensitive. A similar approach was employed by Dunham and coworkers (10) in the study of swelling-sensitive K+-Cl− cotransport in sheep RBCs. On the basis of their results, Dunham et al. concluded that swelling-induced activation of K+-Cl− cotransport in sheep RBCs is the result of inhibition of swelling-sensitive kinase activity in the presence of tonic opposing phosphatase activity.
Analogous to mammalian RBCs, Amphiuma RBCs possess volume-regulatory inorganic ion flux pathways (4). The shrinkage-activated Na+ influx pathway is a close structural homolog of the mammalian the type 1 Na+/H+ exchanger (NHE1), and the swelling-activated K+ efflux pathway is a K+/H+ exchange for which the molecular identity is not known. Tight reciprocal coordination of activity led us to speculate previously that K+/H+ exchange is due to loss of cation selectivity of the Na+/H+ exchanger; however, we have seen no convincing molecular evidence that the two pathways are in fact the same protein. Prior evidence suggests that the activation of both Na+/H+ and K+/H+ exchange are dependent on increased net phosphorylation (i.e., kinase activity). Briefly, exposure of Amphiuma RBCs to phorbol 12,13-myristate acetate (7) or the protein phosphatase inhibitor calyculin-A (26) results in simultaneous, robust stimulation of both Na+/H+ and K+/H+ exchange. However, if osmotic cell shrinkage or swelling are superimposed with exposure to calyculin-A, then preferential activation of either Na+/H+ or K+/H+ exchange occurs, respectively. This suggests that both ion flux pathways are activated by net phosphorylation, yet fine control of ion flux activity is achieved by volume-specific modulation of kinase activity. Furthermore, if kinase activity is responsible for activation of both Na+/H+ and K+/H+ exchange, then the volume-dependent events responsible for coordination of the Na+/H+ and K+/H+ exchangers around the volume set point in Amphiuma RBCs are markedly different from those described for Na+/H+ exchange and K+-Cl− cotransport in dog, rabbit, or sheep RBCs.
To understand the basis for coordinated volume-dependent control of Amphiuma RBC Na+/H+ and K+/H+ exchange, we employed the approach of Jennings and Al-Rohil (18). Our data are consistent with the interpretation that the activation and deactivation kinetics of the Na+/H+ and K+/H+ exchangers follow a monotonic steady-state transition and are described well by Eq. 1. Therefore we infer that these processes are controlled by a single rate-limiting step and that analysis based on a two state model is justified. Employing this approach, we present evidence that the relaxation times (τ) for the activation of both Na+/H+ and K+/H+ exchange pathways decrease with progressive activation (by cell shrinkage or swelling, respectively). In addition, for Na+/H+ exchange deactivation we find that τ decreases with subsequent incremental swelling of preshrunken cells, and similarly for K+/H+ exchange deactivation, τ decreases with subsequent incremental shrinkage of preswollen cells. In summary, the basis for coordinated regulation of the Na+/H+ and K+/H+ exchangers is that the kinases that activate Na+/H+ and K+/H+ exchange have reciprocal volume sensitivities as do the phosphatases responsible for their deactivation.
MATERIALS AND METHODS
As described previously (26), venous blood was drawn from adult Amphiuma tridactylum into 12-ml syringes containing 0.1 ml of heparin (10 kU/ml) as approved by IACUC Protocol 12754 (Shein Pharmaceutical, Florham Park, NJ) and separated from the plasma by gentle centrifugation (1,000 g) in a Clay Adams “Dynac” centrifuge (Clay Adams, Parsippany, NJ). Subsequently, RBCs were washed three times in 10–15 volumes of isosmotic Ringer (IR) solution matched to the animal’s plasma osmolarity (220–250 mosM). IR contained (in mM) 90–110 NaCl, 3–5 KCl, 1 MgCl2, 0.5 CaCl2, 30 HEPES, 5 glucose and pH adjusted to 7.65 (23°C). The cells were then resuspended at a 10% hematocrit (hct) and incubated for 90 min (preincubation period) prior to experimental treatment. Unless specified, alterations in osmolarity were achieved by varying NaCl. All experimental media contained 1 mM ouabain (Sigma Chemical, St. Louis, MO). To initiate an experiment, cells were centrifuged (1,000 g) and suspended in the specified experimental media (10% hct).
To activate and/or deactivate the K+/H+ or Na+/H+ exchangers, cells were exposed to a broad range of hyposmotic or hyperosmotic media. We used hyposmotic media of 0.8, 0.65, 0.6, and 0.55 times isosmotic (× IR) osmolarity, respectively (note: isosmotic osmolarity ranged from 220 to 250 mosM according to animal plasma osmolarity). Hyperosmotic media employed were 1.2, 1.4, and 1.6 × IR, respectively.
In some experiments it was necessary to activate the Na+/H+ or K+/H+ exchangers yet prevent net ion flux and volume changes mediated by these pathways. Accordingly, cells were exposed to “thermodynamically nulled” media with [Na+] and [K+] (brackets denote concentration) chosen to maintain the net force acting either or both exchangers at zero (thermodynamic equilibrium) (see preceding article, Ref. 26). This was accomplished by altering media [Na+] and/or [K+] so that the difference between the chemical potential for Na+ (and/or K+) and that for H+ equals zero (3, 5). At thermodynamic equilibrium, cell volume perturbation will activate volume-sensitive Na+/H+ or K+/H+ exchange, yet neither exchanger can mediate net ion flux since at equilibrium influx must equal efflux. Consequently, in nulled medium the exchangers are volume activated, yet they cannot contribute to volume recovery.
Net ion and water flux measurements.
Net changes in cell Na+, K+, Cl−, and water content were measured by transferring 400 μl of cell suspension to preweighed polyethylene tubes (Stockwell Scientific, Monterrey Park, CA), at predetermined times following initiation of an experiment. Net transport was terminated by centrifugation at 12,000 g for 4 min. The supernatant was then separated from the cell pellet and both were stored for analysis. The tubes containing the cell pellet were cleaned by aspiration to remove any remaining supernatant and weighed to determine wet cell weight. Cell pellets were lysed in 250 μl of 40 mM ZnSO4 and 5 mM MgSO4 (Mg2+ is a cofactor for endogenous nuclease to prevent DNA hemoglobin gel formation upon cell disruption, and Zn2+ was used to precipitate proteins) followed by mechanical agitation to ensure complete lysis. The lysate was centrifuged for 4 min at 12,000 g to separate the insoluble cell matter from the clear supernatant. Supernatants were analyzed for Na+ and K+ concentration by flame photometry (Instrumentation Laboratories, model 443, Boston, MA) and for Cl− by potentiometric titration with silver ions (Buchler Chloridometer, Searle Diagnostics, Fort Lee, NJ). The insoluble cell matter was dried at 70°C for 24 h and used to determine cell water content (wet weight − dry weight) and to normalize cell ion and water content to kilograms dcs. Cell ion and H2O content were corrected for extracellular trapped volume by an empirically determined factor (1, 6).
Unidirectional sodium (22Na+) and potassium (86Rb+) influx measurements.
Unidirectional Na+ or K+ influx was determined by suspending cells (10% hematocrit) in experimental media containing 22Na+ or 86Rb+ (5–10 μCi/ml; NEN Life Sciences Products, Boston, MA). At predetermined time intervals, 100 μl aliquots were transferred to 1.5-ml polyethylene tubes containing 900 μl of isotope-free flux medium (stop medium) floated on top of 400 μl of dibutyl phthalate (Sigma Chemical, St. Louis, MO). Cells were separated from the isotope solution by centrifugation through dibutyl phthalate. The supernatant was removed along with residual oil by inverting and cutting the tube just above the cell pellet (to minimize extracellular trapped isotope). The radioactivity associated with the pellet was quantified with a gamma counter (Packard Instruments, Downers Grove, IL) for 22Na+ or a beta counter (Packard Instruments) for 86Rb+. Parallel samples were transferred to 400-μl polyethylene tubes for determination of dry cell weight and medium specific activity. Unidirectional Na+ or K+ uptake (with 86Rb+ as a tracer for K+) was calculated on the basis of medium specific activity and expressed as millimoles of ion per kilogram dcs. Initial rates of Na+ or K+ uptake were determined by linear regression (r2 > 0.95) and expressed as millimoles ion per kilogram dcs per minute. During a specified sampling interval, medium specific activity was minimally three orders of magnitude greater than that of the intracellular compartment to minimize back flux.
Regression and statistical analyses of data.
To estimate τ, Na+ (22Na+) or K+ (86Rb+) uptake data obtained during Na+/H+ or K+/H+ exchange activation or deactivation were fit to Eq. 1 using Delta Graph (SPSS) to perform a two-parameter best fit by using an interactive algorithm based on the Levenberg-Marquardt method (31). This method yielded estimates of Jf and τ, based on Na+ or K+ uptake data and experimentally determined values of Ji. The data were fit by the same algorithm by using GraphPad Prism version 4.0c for Macintosh (GraphPad Software, San Diego, CA; www.graphpad.com), with a predetermined value for Jf and a one-parameter fit to generate mean ± SE values for τ in table data. Differences between mean values for τ were considered statistically significant when P < 0.05 using one-way ANOVA with a Tukey-Kramer posttest (for multiple comparisons) or Student's t-test (for pairs of values) (GraphPad Prism version 4.0c for Macintosh; GraphPad Software, San Diego, CA; www.graphpad.com).
Na+/H+ and K+/H+ exchange activity as a function of cell volume.
To illustrate the coordination of volume regulatory ion flux activity around the cell volume set point, ouabain-insensitive Na+ (22Na+) and K+ (86Rb+) uptake rates were measured as a function of media osmolarity and hence cell volume in Amphiuma RBCs (Fig. 1). The data demonstrate that in isotonic media (240 mosM) Na+/H+ exchange and K+/H+ exchange fluxes are small yet finite and that flux rates increase selectively and dramatically in increasingly anisosmotic media. In hyposmotic media, K+ uptake (via K+/H+ exchange) is progressively increased as an inverse function of osmolarity (proportional to cell volume) yet the rate of Na+ uptake (via Na+/H+ exchange) remains unchanged. When cells are suspended in media of increasing hyperosmolarity, the rate of Na+ uptake increases with increasing osmolarity (inversely proportional to cell volume), yet the rate of K+ uptake remains constant or decreases. Hence, volume-dependent Na+/H+ exchange and K+/H+ exchange in Amphiuma RBCs are coordinately controlled around the cell volume set point.
Shrinkage-induced Na+/H+ exchange activation.
To determine the kinetics of Na+/H+ exchange activation in response to cell shrinkage in hyperosmotic media, ouabain-insensitive unidirectional 22Na+ uptake rates were measured in NaCl media at 1.2, 1.4, or 1.6 × IR as a function of time (Fig. 2). The data in Fig. 2 illustrate that Na+/H+ exchange activity increases and the time required for activation decreases, as cells are progressively shrunken. Regression analysis of the 22Na+ uptake data with Eq. 1 reveals that the τ values for shrinkage-induced activation of Amphiuma RBC Na+/H+ exchange in media 1.2, 1.4, or 1.6 × IR were 64, 36, and 22 min, respectively (see Table 1, top). These results demonstrate that the rate at which Amphiuma RBC Na+/H+ exchange activity approaches the new steady state (k12 + k21) increases as cell volume decreases.
The flux data in Fig. 2 are well fit to Eq. 1, yet it is noteworthy that cell volume recovers by nearly 30% during the 30-min flux measurement period, due to Na+/H+ exchange-mediated uptake of Na+ and osmotically obliged water. To the extent that volume recovery occurs during the flux measurement interval, the Na+/H+ exchanger should begin to deactivate, which could lead to underestimates of τ. To address this issue, we measured 22Na+ uptake and calculated τ in hyperosmotic thermodynamically “nulled” media where [Na+] was adjusted to set the thermodynamic driving force for the Na+/H+ exchange at zero; i.e., thermodynamic equilibrium (see materials and methods). In nulled media, it is energetically impossible for Na+/H+ exchange to mediate net Na+ transport and therefore changes in cell water content (cell volume recovery). Hence, by measuring unidirectional 22Na uptake (Fig. 3) in nulled media, it is possible to determine the temporal behavior of Na+/H+ exchange activity in response to cell shrinkage, in the absence of volume recovery and any resultant deactivation of Na+/H+ exchange. The calculated τ values for Na+/H+ exchange activation in nulled 1.2, 1.4, or 1.6 × IR media were 76, 47, and 22 min, respectively, very similar to values obtained from the data in Fig. 2. Therefore, the partial volume recovery that occurs during the 22Na+ flux measurement period does not significantly affect calculations of τ. Accordingly, all subsequent data were obtained in normal (not nulled) hyperosmotic media. The results of studies of shrinkage-induced activation of Na+/H+ exchange (Fig. 2) are summarized in the top half of Table 1.
Swelling-induced deactivation of Na+/H+ exchange.
Our studies of Na+/H+ exchange activation in response to cell shrinkage reveal that, as cells are progressively shrunken in hyperosmotic media, the rate of activation, and hence the sum of the forward and reverse rate constants (k12 + k21) governing activation to the new steady state, increases. Since k12 is the forward rate constant (representing kinase activity), an increase in the sum of k12 + k21 during shrinkage activation must be in part due to an increase in k12, which is therefore volume sensitive. However, from these results it is not possible to reach conclusions about the volume sensitivity of the reverse rate constant k21. To assess the volume sensitivity of k21, we determined τ for Na+/H+ exchange deactivation by acutely swelling preshrunken cells and monitoring the time course for Na+/H+ exchange deactivation (Fig. 4). The data in Fig. 4 were obtained from cells preshrunken in thermodynamically nulled hyperosmotic (1.6 × IR) medium for 30 min, followed by swelling to or above control volume by transfer to isosmotic or hyposmotic medium, respectively, at time zero. The data illustrate that reswelling previously shrunken cells deactivates Na+/H+ exchange (Na+ uptake) and that the rate of deactivation increases with progressive increases in cell volume. Accordingly, when preshrunken cells were swollen to normal control volume by suspension in isosmotic medium, the relaxation time for Na+/H+ exchange deactivation was 4.4 min, yet when swollen to volumes greater than control in hyposmotic medium, τ for deactivation was 2.6 min (Table 1). These results indicate that the reverse rate constant k21 increases as shrunken cells are progressively swollen to and beyond normal control volume, consistent with the interpretation that Na+/H+ exchange is deactivated by a swelling-activated phosphatase. Hence the phosphatase responsible for Na+/H+ exchange deactivation is volume sensitive.
Swelling-induced K+/H+ exchange activation.
To complement the studies of Na+/H+ exchange activation we investigated the kinetics of swelling-induced K+/H+ exchange activation. These studies were designed to provide insight into the basis for K+/H+ exchange activation and ultimately, through comparison to the Na+/H+ exchange pathway, permit us to address the basis for coordinated control of shrinkage-induced Na+/H+ exchange and swelling-induced K+/H+ exchange. The τ values for swelling-induced K+/H+ exchange (unidirectional 86Rb+ uptake) activation in moderately hyposmotic media (0.8 and 0.65 × IR) were calculated as 30 and 13 min, respectively (Fig. 5), whereas τ for cells swollen in extreme hyposmotic media (0.55 × IR) was too rapid to measure (< 1 min) with the techniques employed in this study. The results of the analyses of K+/H+ exchange activation are summarized in Table 2, top and are qualitatively similar to those for Na+/H+ exchange activation. Briefly, the rate of K+/H+ exchange activation increases with increasing cell volume, and therefore relaxation times decrease as cells are progressively swollen. As with activation of Na+/H+ exchange, K+/H+ exchange activation is characterized by increases in k12/k21 and (k12 + k21). This pattern is only possible if the rate of the activating kinase (k12) increases with increasing cell volume.
Again, it might be argued that volume recovery during the flux measurement period could affect the calculated relaxation times for K+/H+ exchange activation. However, the decrease in cell volume during the measurement period was only 3–4% in the most extreme hyposmotic medium (0.55 × IR). In more modest hyposmotic media (0.8 and 0.65 × IR), volume-regulatory decreases in water content were 2.0 ± 0.6 and 2.3 ± 1.3%, respectively (means ± SE, n = 3). Therefore it is unlikely that the relaxation times for K+/H+ exchange activation are affected by the minimal volume recovery that occurs during the flux measurement period. This assumption seems even more reasonable if one considers that a 30% volume recovery during regulatory volume increase has no significant effect on the apparent relaxation times for Na+/H+ exchange activation (see Shrinkage-induced Na+/H+ exchange activation above).
Shrinkage-induced K+/H+ exchange deactivation.
To gain insight into the kinetics of K+/H+ exchange deactivation and more specifically the volume-sensitivity of k21, we measured K+ (86Rb+) flux in preswollen cells subsequently shrunken to or below normal control volume. The data in Fig. 6 illustrate that shrinking preswollen cells to normal volume results in K+/H+ exchanger deactivation with τ = 6 min. However, when preswollen cells were placed in hyperosmotic media and shrunken below normal control volume, τ was reduced to 2.7 min. The results of three representative experiments (Table 2, bottom) demonstrate that τ for K+/H+ exchange deactivation decreases in proportion to decreasing cell volume. Therefore k12 + k21 increases, whereas k12/k21 decreases, indicating that k21 increases in proportion to cell shrinkage during K+/H+ exchange deactivation.
The aim of this study was to gain insight to the basis for the coordinated control of the volume-induced Na+/H+ and K+/H+ exchangers in Amphiuma RBCs. To this end, we measured the time course of activation and deactivation of shrinkage-induced Na+/H+ exchange and swelling-induced K+/H+ exchange in Amphiuma RBCs and fit the data to Eq. 1 to calculate relaxation times [τ = (k12 + k21)−1]. The data demonstrate that the rate of Na+/H+ exchange activation increases as a graded function of cell shrinkage and that the rate of Na+/H+ exchange deactivation is proportional to the volume to which preshrunken cells are swollen. Likewise, the rate of K+/H+ exchange activation increases with cell swelling, whereas the rate of K+/H+ exchange deactivation increases in proportion to the degree to which preswollen cells are shrunken. The behavior described above is only possible if the kinases and the phosphatases that control the flux pathways are all volume sensitive.
Shrinkage-induced activation of Na+/H+ exchange: volume dependence of the forward rate constant, k12.
According to the two-state model applied in this study, the relative magnitudes of k12 and k21 reflect the fraction of the Na+/H+ exchanger population in the resting or active state. If in isosmotic media k12 and k21 were nearly equal in magnitude, then half of the Na+/H+ exchanger population would be in the active state and half in the resting state. If this were the case, then the highest attainable Na+ uptake rate following activation in hyperosmotic media would be a doubling of that in isosmotic media. Yet, as shown in Fig. 1, in Amphiuma RBCs the Na+ uptake rate in hyperosmotic media is as much as two orders of magnitude greater than that in isosmotic medium. Thus prior to determining τ, we reasoned that in isosmotic medium the rate constant for Na+/H+ exchange deactivation (k21) (phosphatase activity) must be much larger than that for Na+/H+ exchange activation (k12) (kinase activity) (k21 ≫ k12).
The data in Table 1 and Figs. 2 and 3 establish that τ for shrinkage-induced activation of Na+/H+ exchange decreases as cell volume decreases. Therefore, the rate of approach to the new steady state (k12 + k21) increases in proportion to the degree of cell shrinkage (Table 1). On the other hand, since Na+/H+ exchange activity increases with decreasing cell volume (Figs. 1⇑–3), the ratio k12/k21 must increase during shrinkage-induced activation. In order for both k12 + k21 and k12/k21 to increase with shrinkage, the forward rate constant (k12), which represents net kinase activity, must increase as cell volume decreases. Taken together, the data suggest that, in isosmotic media, k21 is much larger than k12 (k21 ≫ k12), yet in hyperosmotic media k12 increases as cell volume decreases, resulting in graded activation of Na+/H+ exchange and a decrease in the relaxation time for Na+/H+ exchange activation that is inversely proportional to cell volume (proportional to medium osmolarity). To determine the volume-sensitivity of the reverse rate constant, we studied swelling-induced Na+/H+ exchange deactivation.
Swelling-dependent deactivation of Na+/H+ exchange: volume dependence of the reverse rate constant, k21.
We (and others) have shown that activation of Na+/H+ exchange is dependent on kinase activity (7, 11–13, 26, 27). If Na+/H+ exchange activation is governed by net kinase activity, denoted by the forward rate constant k12, then the deactivation rate constant k21 must represent net phosphatase activity. To assess the volume sensitivity of k21, preshrunken cells were swollen to or above normal control volume and τ for swelling-dependent Na+/H+ exchange deactivation (Table 1, bottom, and Fig. 4) were determined. Our data illustrate that τ for Na+/H+ exchange deactivation decreases progressively as cells are swollen to or above isosmotic, control volume: the rate (k12 + k21) of approach to the new steady-state increases with cell swelling (Table 1). Since k12 increases as cell volume decreases (see shrinkage-induced Na+/H+ exchange activation: volume dependence of the forward rate constant, k12, above), it follows that k12 must decrease as preshrunken cells are swollen. Given that Na+/H+ exchange deactivation requires a decrease in k12/k21 and that k12 decreases with deactivation, yet k12 + k21 increases as preshrunken cells are reswollen (τ decreases), we conclude that k21 increases with swelling and that the increase in k21 is greater than the absolute decrease in k12. Therefore, upon reswelling of preshrunken cells, the rate constant representing net phosphatase activity, k21, increases more rapidly than k12 (net kinase activity) decreases and consequently k21 is also volume sensitive. If deactivation were solely the result of a volume-sensitive decrease in k12 then progressive reswelling of shrunken cells should result in longer relaxation times since in this scenario k12 + k21 must decrease. Because this is not the case, k12 + k21 must increase even though k12 decreases with swelling. We conclude that k21 increases as a function of cell swelling.
Comparison of the sum of the forward and reverse rate constants (k12 + k21) from cells transferred from hyperosmotic to isosmotic media with that from cells transferred from isosmotic to hyperosmotic media (Table 1) provides insights into the volume-sensitivity of the forward (k12) and reverse (k21) rate constants. The data at the bottom half of Table 1 illustrate that k12 + k21 for cells transferred from hyperosmotic to isosmotic medium (i.e., control cell volume) is larger than k12 + k21 for cells transferred from isosmotic to hyperosmotic media. Since, as illustrated in Fig. 1, Na+/H+ exchange activity is low in isosmotic medium, k21 must be dominant. Therefore, the reduction in k12 + k21 as cells are transferred from isosmotic to hyperosmotic media reflects a substantial reduction in k21 in the transition from isosmotic to hyperosmotic media. This interpretation is further supported by the fact that decreases in τ with progressive shrinkage must reflect shrinkage-dependent increases in k12. The only way that the sum k12 + k21 can increase with progressive shrinkage yet remain smaller than k12 + k21 in isosmotic media is if k21 decreases precipitously whereas k12 increases with shrinkage. Finally, since τ decreases (increasing k12 + k21) as preshrunken cells are progressively swollen (hyperosmotic → isosmotic or hyposmotic media), even though k12 decreases as cell volume is increased, we must conclude that k21 increases as preshrunken cells are swollen to and beyond control volume. Because k21 corresponds to the rate constant associated with net phosphatase activity, we conclude that the phosphatase activity responsible for deactivating Na+/H+ exchange is volume sensitive and increases in proportion to cell volume.
In light of the above, we conclude that both the kinase-dependent rate-limiting step (k12) in Na+/H+ exchange activation and the phosphatase-dependent rate-limiting step (k21) responsible for Na+/H+ exchange deactivation are volume dependent. Moreover, the data support a scheme in which Na+/H+ exchange activity is controlled by a shrinkage-activated (swelling-inactivated) kinase and a swelling-activated (shrinkage-inhibited) phosphatase. Clearly then, the activity levels of the controlling kinase and phosphatase are reciprocally related as a function of cell volume. This arrangement provides the basis for understanding the empirically observed fine control of shrinkage-induced (swelling-inhibited) Na+/H+ exchange in Amphiuma RBCs.
Swelling-dependent activation of K+/H+ exchange: volume dependence of the forward rate constant, k12.
Following the same rationale used for the analysis of Na+/H+ exchange, if in isosmotic media k12 and k21 were equal, then the maximum possible K+/H+ exchange (86Rb+ uptake) rate in hyposmotic media would be double that in isosmotic media. Since the K+ uptake rate in hyposmotic media is as much as three orders of magnitude greater than in isosmotic medium (Fig. 1), the rate constant responsible for K+/H+ exchange deactivation (k21) must be much larger than that governing K+/H+ exchange activation (k12).
In accordance with the two-state model, activation of K+/H+ exchange requires that the ratio k12/k21 increase relative to that in the resting state. The magnitude of k12/k21 can increase upon swelling if k12 increases or k21 decreases. However, if increases in k12/k21 with progressive swelling are due to decreased k21 then k12 + k21 will decrease and τ will increase. Because in our studies τ decreases as cells are progressively swollen, we conclude that k12 increases with progressive swelling and is therefore volume sensitive (Fig. 5 and Table 2).
Shrinkage-dependent deactivation of K+/H+ exchange: volume dependence of the reverse rate constant, k21.
Our studies of K+/H+ exchange activation establish the volume sensitivity of k12 yet provide little information regarding the volume sensitivity of k21. To assess the volume sensitivity of k21, we determined τ for shrinkage-induced deactivation of K+/H+ exchange in previously swollen cells (Fig. 6, Table 2). Given the results of these studies, we determined that τ decreased as the volume of preswollen cells progressively decreased to or below normal resting volume. Clearly then, k12 + k21 increases as preswollen cells are shrunken to or below normal volume (Table 2). Furthermore, since k12 increases with swelling, it follows that k12 must decrease with cell shrinkage. This being the case, the only way for k12 + k21 to increase as preswollen cells are shrunken is if k21 increases. We conclude that k21, the rate constant associated with net phosphatase activity, increases during shrinkage-induced deactivation of the K+/H+ exchanger and is therefore volume sensitive.
The above conclusion is analogous to that with regard to the control of Na+/H+ exchange activity and is based on examination of the magnitudes of k12 + k21 for cells transferred from hyposmotic to isosmotic and from isosmotic to hyposmotic medium. Briefly, Table 2, bottom illustrates that k12 + k21 is larger for K+/H+ exchange deactivation as cells are transferred from hyposmotic to isosmotic media than for K+/H+ exchange activation as cells are transferred from isosmotic to mildly hyposmotic media (0.8 or 0.65 × IR). The observation that k12 + k21 is larger in isosmotic medium, where k21 dominates and K+/H+ exchange is virtually inactive, leads us to the conclusion that k21 decreases dramatically during the transition from the inactive (isosmotic medium) to the active state (hyposmotic media).
The kinetic data for volume-dependent control of the Amphiuma K+/H+ exchange are similar and analogous to those discussed regarding the control of Na+/H+ exchange. Briefly, volume-dependent K+/H+ exchange activity is controlled by a swelling-activated (shrinkage-deactivated) kinase and a swelling-deactivated (shrinkage-activated) phosphatase. Thus swelling-dependent activation of K+/H+ exchange is achieved through swelling-induced increases in kinase activity and simultaneous decreases in the activity of the opposing phosphatase. In contrast, shrinkage-dependent deactivation of K+/H+ exchange is achieved through shrinkage-dependent decreases in kinase activity and shrinkage-induced increases in phosphatase activity. The activities of the controlling kinase-phosphatase pair are both volume sensitive and reciprocally related through changes in cell volume.
Model for regulation of the Amphiuma RBC Na+/H+ and K+/H+ exchangers in isosmotic media and anisosmotic media.
The data presented in this study support the conclusion that volume-dependent regulation of the Amphiuma RBC Na+/H+ and K+/H+ exchangers are the result of the activities of volume-sensitive kinases and phosphatases. Specifically, Na+/H+ exchange activity is determined by the relative magnitudes of a shrinkage-activated kinase (k12sh) and a swelling-activated phosphatase (k21sw; Fig. 7). On the other hand, K+/H+ exchange activity is determined by the relative magnitudes of a swelling-activated kinase (k12sw) and a shrinkage-activated phosphatase (k21sh; Fig. 7). Thus in Amphiuma RBCs the kinases and phosphatases regulating the activities of the Na+/H+ and K+/H+ exchangers are volume sensitive. At control volume in isosmotic medium, the activities of both the Na+/H+ and K+/H+ exchangers are minimal (Fig. 1) because the activities of the deactivating reverse rate constants (k21) are much greater than those of the activating forward rate constants (k12). Specifically, in isosmotic media, k21sw ≫ k12sh and therefore the Na+/H+ exchanger is in the resting or inactive state (Fig. 7). Following the same reasoning, the low basal activity of K+/H+ exchange in isosmotic media is consistent with the interpretation that k21sh ≫ k12sw (Fig. 7). In summary, high levels of phosphatase activity in isosmotic media are responsible for maintaining the low Na+/H+ and K+/H+ exchange activities observed in isosmotic media (Fig. 1).
The data obtained from volume-perturbed cells lead us to conclude that coordinated regulation of the Na+/H+ and K+/H+ exchangers around the volume set point in Amphiuma RBCs is the result of volume-sensitive kinases and phosphatases that are reciprocally regulated. In brief, as cells are shrunken, the shrinkage-activated (swelling-deactivated) kinase (k12sh) activity responsible for Na+/H+ exchange activation increases and its opposing shrinkage-inactivated (swelling-activated) phosphatase (k21sw) decreases (Fig. 7). Conversely, as cells are swollen the swelling-activated (shrinkage-deactivated) kinase (k12sw) activity responsible for K+/H+ exchange activation increases and the activity of the opposing shrinkage-activated (swelling-inactivated) phosphatase (k21sh) decreases (Fig. 7). Therefore, the activity of the shrinkage-activated phosphatase that deactivates the K+/H+ exchanger is greatest in shrunken cells whereas the activity of the swelling-activated phosphatase that deactivates the Na+/H+ exchanger is greatest in swollen cells. In this scheme, both kinases and phosphatases are volume sensitive and the opposing kinase-phosphatase pairs vary reciprocally with volume, permitting selective volume-dependent activation of Na+/H+ or K+/H+ exchange. This ensures coordinated, selective activation of Na+/H+ or K+/H+ exchange, since the kinases that regulate the exchange pathways are reciprocally related as a function of cell volume, as are their opposing phosphatases.
The kinetic model for coordinated control of the Na+/H+ and K+/H+ exchangers explains the effects of calyculin-A in isosmotic and anisosmotic media.
The data presented in this study are consistent with, and in fact explain, our prior observations regarding activation of Na+/H+ and K+/H+ exchange by the protein phosphatase inhibitor CLA (26). Briefly, exposure of Amphiuma RBCs to CLA in isosmotic medium results in simultaneous activation of Na+/H+ and K+/H+ exchange, yet superimposition of CLA exposure and cell shrinkage or swelling results in selective activation of Na+/H+ or K+/H+ exchange, respectively. Since CLA inhibition of phosphatase activity in isosmotic medium results in activation of both Na+/H+ and K+/H+ exchangers, then we conclude that in isosmotic medium the basal activities of shrinkage- and swelling-induced kinases that activate Na+/H+ and K+/H+ exchangers, respectively, are finite (Fig. 7). Furthermore, according to the model, cell shrinkage decreases the activity of the swelling-activated kinase (k12sw), responsible for activating K+/H+ exchange, relative to the activity in isosmotic medium (Fig. 7). Consequently, exposure of shrunken cells to CLA results in stimulation of Na+/H+ exchange with a lesser stimulation of K+/H+ exchange relative to cells in isosmotic medium exposed to CLA, because in shrunken cells the swelling-activated (shrinkage-inhibited) kinase (k12sw) that activates K+/H+ exchange is minimized. As depicted in Fig. 7, cell swelling increases the activity of the kinase that activates K+/H+ exchange and decreases the activity of the kinase responsible for Na+/H+ exchange activation. Consistent with the model, exposure of osmotically swollen cells to CLA stimulates K+/H+ exchange, with only modest stimulation of Na+/H+ exchange because the shrinkage-activated (swelling-inhibited) kinase (k12sh) that activates Na+/H+ exchange is minimal in swollen cells (compare Fig. 7, A and C). Therefore, the models presented in Fig. 7 account for the effects of CLA on Na+/H+ and K+/H+ exchange activity by Amphiuma RBCs in isotonic and anisotonic media in that they predict that superimposition of CLA exposure and osmotic cell shrinkage or swelling should result in preferential activation of Na+/H+ or K+/H+ exchange, respectively.
Contrast of Amphiuma RBCs to other systems.
Although it is well documented that volume regulatory ion flux pathways are reciprocally regulated around the volume set point (4, 14, 25, 28, 29, 32), the basis for this regulation has remained largely obscure. Parker and coworkers observed that the time necessary for activation or deactivation of the Na+/H+ exchanger and K+-Cl− cotransporter are reciprocally related in dog RBCs (29). In dog RBCs, cell swelling slowly activates the K+-Cl− cotransporter and slowly deactivates the Na+/H+ exchanger, whereas cell shrinkage rapidly activates the Na+/H+ exchanger and rapidly deactivates the K+-Cl− cotransporter. This time course reciprocity suggests that a common regulatory element is responsible for the coordinated control of Na+/H+ exchange and K+-Cl− cotransport activity. These investigators employed relaxation kinetics and found: 1) the regulatory mechanism controlling the activity of the Na+/H+ exchanger and K+-Cl− cotransporter in dog RBCs is a single kinase-phosphatase pair, 2) the kinase (k21) activates the Na+/H+ exchanger and deactivates the K+-Cl− cotransporter and 3) the phosphatase (k12) activates the K+-Cl− cotransporter and deactivates the Na+/H+ exchanger. In addition, within this kinase-phosphatase pair, only the kinase activity (k21) is volume sensitive. This model is compelling because it accounts for the activity of swelling and shrinkage-sensitive ion flux pathways. Moreover, because of its mutually exclusive nature it explains the basis for coordinated, selective activation of the shrinkage-induced Na+/H+ exchange and swelling-sensitive K+-Cl− cotransport around the cell volume set point in dog RBCs.
In contrast to dog RBCs, our results from Amphiuma RBCs are consistent with volume-induced increases in k12 + k21 during both activation and deactivation of the Na+/H+ and K+/H+ exchangers. Moreover, as illustrated in Tables 1 and 2, both the forward (k12) and reverse (k21) rate constants are volume sensitive. Therefore, unlike dog RBCs where a single kinase-phosphatase pair controls the activity of the swelling-sensitive K+-Cl− cotransport and shrinkage-sensitive Na+/H+ exchange and only the kinase is volume sensitive, the data obtained from Amphiuma RBCs suggest that the activity of the Na+/H+ and K+/H+ exchangers are controlled by two volume-sensitive kinase-phosphatase pairs and that the activity of both kinases and phosphatases are volume sensitive (Fig. 7). Having K+/H+ exchange and Na+/H+ exchange under the control of different kinase-phosphatase pairs permits a greater range of dynamic activity changes for both transporters (as experimentally evident in Fig. 1) than is possible under the scheme proposed for the dog RBC.
Tight reciprocal coordination of the activity of Na+/H+ and K+/H+ exchangers by their controlling kinase-phosphatase pairs suggests that the four volume-dependent kinetic events (k12 and k21, for shrinkage or swelling) (Fig. 7) are initiated by a single cell volume sensor. In principle, the activity of the volume-sensitive kinases and phosphatases could correspond to four independent volume sensors. Our data do not support or refute either model. Collective observations of Amphiuma RBC volume regulation are consistent with yet do not indicate macromolecular crowding, raft partitioning, actin cytoskeleton, or membrane mechanical deformation as primary volume sensors during regulatory volume increase or regulatory volume decrease. The relaxation times for activation and deactivation of Na+/H+ and K+/H+ exchange are not characteristic of the primary volume sensor; rather they correspond to the rate-limiting events in signal transduction downstream of volume sensing, reflecting the magnitudes of volume-sensitive kinases and phosphatases. In the present study, relaxation time constants for transporter activation and deactivation were very fast (0 to 30 min) (Tables 1–2), supporting a role for rapid posttranslational control of Na+/H+ and K+/H+ exchange activity. Preliminary cell surface proteolysis experiments with Amphiuma RBCs showed no increase in surface expression of NHE1 during cell shrinkage, suggesting that membrane protein trafficking is not involved in shrinkage activation of NHE1 (Z. Zhuang, R. R. Rigor, and P. M. Cala unpublished data). Therefore, control of Na+/H+ exchange activity is likely due to phosphorylation of NHE1 or an associated regulatory protein. Although the molecular identity of the K+/H+ exchanger is not known, the short (0 to 10 min) relaxation times for activation or deactivation of K+/H+ exchange suggest that volume-dependent control of K+/H+ exchange activity is also dependent on direct phosphorylation or dephosphorylation of a membrane transport protein or of an associated regulatory protein and does not involve protein trafficking to the membrane. Finally, despite convincing evidence of a role for volume-sensitive kinases and phosphatases in control of Na+/H+ and K+/H+ exchange activity, the molecular identities of these kinases and phosphatases are yet to be determined.
In summary, regulation of Na+/H+ and K+/H+ exchange around the volume set point in Amphiuma RBCs occurs as a result of the action of two kinase-phosphatase pairs (one pair for each transport pathway) in which both kinase and phosphatase activities are volume sensitive (Fig. 7, B and C). When cells are shrunken, there is an increase in the activity of a shrinkage-activated kinase that activates Na+/H+ exchange and a decrease in the activity of a swelling-activated phosphatase that deactivates Na+/H+ exchange. In contrast, cell swelling increases the activity of a swelling-activated kinase that stimulates K+/H+ exchange and decreases the activity of shrinkage-activated phosphatase that deactivates K+/H+ exchange; thus K+/H+ exchange is activated by cell swelling.
This work was supported by National Heart, Lung, and Blood Institute research grant R01 HL 21179 (to P.M. Cala). A. Ortiz-Acevedo was supported by a National Research Service Award (5 F31 GM18985-02) from the National Institutes of Health.
No conflicts of interest are declared by the author(s).
We thank the late Daniel C. Tosteson for being a source of inspiration and encouragement to those of us engaged in the study membrane transport processes.
Present address of A. Ortiz-Acevedo: Department of Natural Sciences, University of Puerto Rico, Utuado, Puerto Rico 00641.
Present address of H. M. Maldonado: Departamento de Farmacología, Universidad Central del Caribe, Bayamón, Puerto Rico 00960.
Present address of R. R. Rigor: Division of Research, Department of Surgery, School of Medicine, University of California at Davis, Sacramento, CA 95758.
- Copyright © 2010 the American Physiological Society