Invited Symposium: Molecular Physiology of Sodium-Calcium Exchange
Since the Na+/Ca2+ exchange system plays a major role in regulating cardiac contractility one would imagine that the activity of this transport system would itself be tightly regulated. This is certainly true for other mechanisms involved in Ca2+ homeostasis in cardiac cells: the L-type Ca2+ channel, the SERCA Ca2+ pump and the Ca2+ release channel are each regulated by phosphorylation-dependent processes, and perhaps by other mechanisms as well. Early 45Ca2+ flux studies with squid giant axons pointed to two factors which activated exchange activity: cytosolic Ca2+ and ATP [1-3] The advent of the giant patch technique in 1989 [4,5] confirmed this dual mode of regulation in cardiac myocytes, but introduced some complications.
One complication was that ATP was shown to activate exchange activity by counteracting a specific type of inactivation, called Na+-dependent inactivation [6,7]. This complex process involves the entry of the exchanger into an inactive state when it binds Na+ from the cytosolic membrane surface. It was observed experimentally as a time-dependent decline in the amplitude of the outward exchange current from a peak value to a lower steady-state value following the addition of Na+ to the cytosolic surface of the patch. In the presence of cytosolic ATP, the decline in current amplitude was greatly reduced. Elegant noise-analysis studies by Hilgemann and his colleagues  produced strong evidence for the flipping of the exchanger into and out of the inactivated state. The effects of ATP in the giant patches were subsequently shown to be entirely due to the synthesis of phosphatidyl-4,5-inositolbisphosphate (PIP2) in the membrane patch . No evidence was found in the patch studies for phosphorylation of the exchanger by a protein kinase.
The second complication involved the effects of cytosolic Ca2+. Ca2+ was itself shown to have a dual effect on exchange activity . First, Ca2+ binds directly to high affinity regulatory sites in the central, hydrophilic domain of the cardiac exchanger. The binding of Ca2+ to these sites is thought to be essential for all modes of exchanger operation, although there is a relaxation of this requirement under certain experimental conditions (e.g. alkaline pH). The second effect of Ca2+ is to counteract the type of inactivation discussed above, Na+-dependent inactivation. The direct activation of exchange activity by Ca2+ occurs over a concentration range of 0.3-0.6 ÁM in patches prepared from cardiac myocytes or from frog oocytes expressing the wild-type cardiac exchanger [10,11].
For the cardiac exchanger, these modes of regulation have been worked out primarily using the giant patch technique and so the question naturally arises as to whether these processes actually take place in intact cells. The regulatory activation of exchange activity by cytosolic Ca2+ has been clearly documented using exchange current measurements in cardiac myocytes.. The results, however, suggested that the affinity of the exchanger for regulatory Ca2+ in the intact cells was much higher than in the giant patch experiments: Two separate studies reported half-activation of outward exchange currents at Ca2+ concentrations of 20-50 nM [12,13]. Na+-dependent inactivation has also been demonstrated in intact myocytes , but it is uncertain whether this process could be important under physiological conditions where cytosolic Na+ concentrations are low and ATP concentrations are high.
Work in our laboratory has been directed toward investigating the modes of exchanger regulation in Chinese hamster ovary (CHO) cells that have been transfected to express the bovine cardiac exchanger. The results summarized below lead us to question whether either of the mechanisms described above is a physiologically relevant regulatory mechanism.
Na+/Ca2+ exchange activity is most often measured in the "reverse mode", i.e. as Ca2+ influx. The two methodological approaches to measuring exchange activity in transfected CHO cells involve assaying 45Ca2+ uptake or measuring changes in the cytosolic Ca2+ concentration in cells loaded with the Ca2+ indicating dye, fura-2. Either approach entails interpretational problems because once Ca2+ enters the cell, it can be accumulated by intracellular compartments, most notably the mitochondria. To minimize this difficulty, we used Ba2+ as a substitute for Ca2+ since Ba2+ did not appear to be sequestered by either the endoplasmic reticulum or the mitochondria . Using various preincubation protocols to modulate the cytosolic Ca2+ concentration, the initial rate of Ba2+ influx via the exchanger in transfected CHO cells was found to increase with increasing cytosolic [Ca2+] . The half-maximal [Ca2+] was determined to be 44 nM, a value similar to that observed with intact cardiac myocytes (see above) but an order of magnitude lower than found with excised giant patches. This dependence on [Ca2+] was not observed in cells expressing a mutant exchanger which was missing much of its hydrophilic, regulatory domain and was therefore insensitive to regulation by Ca2+.
Activation of Na+/Ca2+ exchange activity by similarly low values of cytosolic [Ca2+] was also observed in studies of Ca2+ efflux in these cells (manuscript submitted). We used the Ca2+ ionophore ionomycin to release Ca2+ from internal stores in the presence or absence of extracellular Na+. In fura-2 loaded cells, the amplitude of the ionomycin-induced [Ca2+] transient was reduced in the presence of extracellular Na+, reflecting the acceleration of Ca2+ efflux by Na+/Ca2+ exchange. These effects of Na+ were not observed in parental cells which lacked Na+/Ca2+ exchange activity. The effects of Na+ on Ca2+ efflux were observed at cytosolic [Ca2+] values of 75 nM or less, confirming that the exchanger was active at these low values. The results were successfully simulated by a kinetic model which depicted the exchanger as a high Vmax, low Km (5 ÁM) transporter, with half-maximal regulatory activation by cytosolic Ca2+ at a concentration of 47 nM.
Thus, whether operating in a Ca2+ influx or Ca2+ efflux mode, the exchanger would appear to be nearly fully activated at "resting" cytosolic [Ca2+] values of 100 nM. This, of course, makes the physiological function of regulatory Ca2+ activation difficult to understand, since the exchanger would be "turned on" at all cytosolic [Ca2+] values likely to be experienced by the cell.
Hilgemann and Ball's study  of the effects of PIP2 on Na+/Ca2+ exchange and K-ATP channel activity ushered in a new appreciation of the modulatory effects of phosphoinositides on membrane transporters. It was suggested that a potential binding site for PIP2 might be the XIP region, a stretch of 20 amino acids containing 8 positively charged residues located immediately after the exchanger's 5th transmembrane segment. A peptide corresponding to the sequence of this region was found to inhibit Na+/Ca2+ exchange activity, hence the name XIP or eXchange Inhibitory Peptide . This observation led to the concept that the XIP segment might have an autoregulatory function and inactivate the exchanger through an intramolecular interaction with an as yet unidentified docking site. In giant patch studies, certain mutations within this region were found to alter the time course and extent of Na+-dependent inactivation, as well as the sensitivity to activation by regulatory Ca2+. Interaction of PIP2 with the XIP region would be expected to disrupt an autoinhibitory configuration and provide an easily understood mechanism for activation of exchange activity. Two questions arise in consideration of this speculative scenario: (1) Do cellular PIP2 levels vary sufficiently to alter exchange activity under physiological conditions? (2) Is the XIP region involved in physiological regulation of exchange activity?
With respect to the first question, cellular PIP2 levels are regulated by an exceedingly complex interplay of pathways for the degradation and resynthesis of this important phosphoinositide . When CHO cells are plated on a plastic surface coated with either fibronectin or polylysine, the cells on fibronectin flatten out and spread over the substratum within a few hours while the cells on polylysine remain round. These differing morphologies are associated with differences in the cellular PIP2 content: The cells on fibronectin contain nearly twice as much PIP2 as those on polylysine (unpublished observations). Similar effects of the extracellular matrix on PIP2 levels have been reported for other cells [19,20]. Despite these differences, the rate of Na+/Ca2+ exchange activity, measured as 45Ca2+ uptake, was identical under the two conditions. Our inability to detect changes in exchange activity upon a 50% reduction in PIP2 levels suggests that, at least in CHO cells, the exchanger is not highly sensitive to variations in the amount of this phosphoinositide. Of course, it is possible that we would begin to see inhibition of exchange activity if we were able to reduce PIP2 levels by more than 50%. Even so, the physiological significance of such an effect would be questionable: PIP2 levels are highly protected in cells and transient reduction of PIP2, e.g. through G-protein mediated phospholipase C activation, leads to a rapid resynthesis of PIP2. Thus, reductions in total PIP2 of more than 50% would rarely be observed under physiological conditions.
We approached the second question by expressing a mutant exchanger in which all the positively charged residues (R and K) in the exchanger's XIP region (219-RRLLFYKYVYKRYRAGKQR) were changed to alanines. The ability of the XIP peptide to act as an exchange inhibitor was shown to be critically dependent upon the presence of the positively charged residues , and a mutation in one of these residues (K229Q) was shown to eliminate Na+-dependent inactivation . These changes might therefore be expected to eliminate any interaction between the XIP region and either PIP2 or an intramolecular docking site. The behavior of CHO cells expressing this mutant was remarkably similar to that of cells expressing the wild-type exchanger. In particular, the initial rate of Ba2+ influx was found to increase with cytosolic [Ca2+] over the same concentration range as for the wild-type exchanger. Other properties of exchange activity in these cells are currently being investigated, and it is possible that subsequent studies will reveal more extensive alterations in exchanger function in this mutant. Nevertheless, these preliminary results suggest that modulatory effects on exchange activity involving the XIP region are either very subtle, or are not manifest in the intracellular environment provided by CHO cells.
Questions and Speculative Answers
Is Na+/Ca2+ exchange regulated? It is hard to imagine that it would not be, considering its general importance in Ca2+ homeostasis in many cells and its specific importance in cardiac physiology. However, the results discussed above raise serious questions about our understanding of this issue. Our results, and the electrophysiologic data obtained with intact cardiac myocytes, suggest that the exchanger is substantially activated at "resting" levels of cytosolic [Ca2+], generally considered to be about 100 nM. Thus, there would appear to be very little room for regulatory modulation of exchange activity over the range of cytosolic [Ca2+] values where the exchanger is most active, i.e. following Ca2+ release from the sarcoplasmic reticulum, when cytosolic [Ca2+] increases into the ÁM range. Perhaps regulatory modulation of exchange activity is physiologically more efficient during diastole than during systole. The brief duration of the [Ca2+] transient following Ca2+ release from the sarcoplasmic reticulum may not provide sufficient time for efficient regulatory activation of the exchanger, which occurs with a time constant of 0.2 - 1 s [10,23]. Changes in the diastolic [Ca2+] might therefore preset the exchanger at an appropriate degree of activation to handle the higher levels of cytosolic [Ca2+] attained during the subsequent systolic interval.
What accounts for the difference in the exchanger's sensitivity to [Ca2+] between intact cells and excised patches? In intact cells, the local concentration of Ca2+ beneath the plasma membrane appears to be much greater than the bulk cytosol, as recently disclosed with plasma membrane targeted aequorins [16,24,25]. The disparity between submembrane and bulk Ca2+ concentrations might be substantially smaller under the experimental conditions of the excised patch studies, with the result that higher Ca2+ concentrations would appear to be required to activate exchange activity.
Do PIP2 and Na+-dependent inactivation regulate exchange activity in intact cells? This has always been an issue with this mode of regulation since cytosolic Na+ levels are generally quite low in cardiac cells (< 12 mM) and ATP levels are high, conditions that would appear to minimize Na+-dependent inactivation. Perhaps Na+-dependent inactivation merely serves a protective role, turning off exchange activity when ATP levels fall and cytosolic Na+ levels rise, as in ischemia. More speculative considerations, based on an increased Na+ concentration in a submembrane "fuzzy space" , could provide scenarios whereby Na+-dependent inactivation plays a more prominent role in exchanger regulation. As mentioned above, total cellular PIP2 levels are highly protected in intact cells, and large changes in total PIP2 seem unlikely. However, local changes in the vicinity of the exchanger could conceivably occur and modulate function, as recently demonstrated for K-ATP channels [27,28],
None of these speculative considerations is particularly satisfying. Compartmentalization and local changes of [Ca2+] and PIP2 are, quite literally, a "fact of life" in the intracellular environment, but without experimental access to such local effects, it is far easier to formulate hypotheses than to test them. Is Na+/Ca2+ exchange regulated? Probably. Do we understand the physiological mechanisms involved? Probably not.
| Discussion Board | Previous Page | Your Symposium |
|Reeves, J; Condrescu, M; Fang, Y; (1998). Is Sodium-Calcium Exchange Regulated?. Presented at INABIS '98 - 5th Internet World Congress on Biomedical Sciences at McMaster University, Canada, Dec 7-16th. Invited Symposium. Available at URL http://www.mcmaster.ca/inabis98/lytton/reeves0755/index.html|
|© 1998 Author(s) Hold Copyright|