Regulation of Na/Mg antiport in rat erythrocytes

In rat erythrocytes, the regulation of Na⁺/Mg²⁺ antiport by protein kinases (PKs), protein phosphatases (PPs), intracellular Mg²⁺, ATP and Cl⁻ was investigated. In untreated erythrocytes, Na⁺/Mg²⁺ antiport was slightly inhibited by the PK inhibitor staurosporine, slightly stimulated by the PP inhibitor calyculin A and strongly stimulated by vanadate. PMA stimulated Na⁺/Mg²⁺ antiport. This effect was completely inhibited by staurosporine and partially inhibited by the PKC inhibitors Ro-31-8425 and BIM I. Participation of other PKs such as PKA, the MAPK cascade, PTK, CK I, CK II, CAM II-K, PI 3-K, and MLCK was excluded by use of inhibitors. Na⁺/Mg²⁺ antiport in rat erythrocytes can thus be stimulated by PKCα.In non-Mg²⁺-loaded erythrocytes, ATP depletion reduced Mg²⁺ efflux and PMA stimulation in NaCl medium. A drastic activation of Na⁺/Mg²⁺ antiport was induced by Mg²⁺ loading which was not further stimulated by PMA. Staurosporine, Ro-31-8425, BIM I and calyculin A did not inhibit Na⁺/Mg²⁺ antiport of Mg²⁺-loaded cells. Obviously, at high [Mg²⁺]_i Na⁺/Mg²⁺ antiport is maximally stimulated. PKC_α or PPs are not involved in stimulation by intracellular Mg²⁺. ATP depletion of Mg²⁺-loaded erythrocytes reduced Mg²⁺ efflux and the affinity of Mg²⁺ binding sites of the Na⁺/Mg²⁺ antiporter to Mg²⁺. In non-Mg²⁺-loaded erythrocytes Na⁺/Mg²⁺ antiport essentially depends on Cl⁻. Mg²⁺-loaded erythrocytes were less sensitive to the activation of Na⁺/Mg²⁺ antiport by [Cl⁻]_i.     

Species-specific Mn2+/Mg2+ antiport from Mg2(+)-loaded erythrocytes.

Mg2(+)-loaded rat erythrocytes performed Mn2+/Mg2+ antiport, which was nonspecifically stimulated by anions and cations. Mn2+/Mg2+ antiport was shown to operate via the Na+/Mg2+ antiporter because extracellular Na+ and Mn2+ inhibited the intracellular uptake of each other's ions competitively. Furthermore, Mn2+/Mg2+ antiport and Na+/Mg2+ antiport were identically inhibited by various amiloride derivatives. Na+/Mg2+ antiport of chicken and human erythrocytes cannot perform Mn2+/Mg2+ antiport although chicken erythrocytes took up more Mn2+ than rat erythrocytes.     

Characterization of Na(+)-dependent Mg2+ efflux from Mg2(+)-loaded rat erythrocytes

Na(+)-dependent Mg2+ efflux from Mg2(+)-loaded rat erythrocytes was determined from the increase of extracellular Mg2+ concentration or decrease of intracellular Mg2+ content, as measured by means of atomic absorption spectrophotometry. Mg2+ efflux was specifically combined with the uptake of Na+ at a stoichiometric ratio of 2Na+:1Mg2+, indicating electroneutral Na+/Mg2+ antiport. Na+/Mg2+ antiport depended on intracellular ATP and was inhibited by amiloride and quinidine, but was insensitive to strophanthin. Net Mg2+ efflux was only occurring at increased concentration of intracellular Mg2+ ([Mg2+]i), and stopped when the physiological Mg2+ content was reached. Intracellular Mg2+ acted cooperatively with a Hill coefficient of 2.4, which may indicate gating of Na+/Mg2+ antiport at increased [Mg2+]i. At increased intracellular Na+ concentration, Na+ competed with intracellular Mg2+ for Mg2+ efflux and Na+ could leave the rat erythrocyte via this transport system. Na+/Mg2+ antiport was working asymmetrically with respect to extra- and intracellular Na+ and Mg2+, and did not perform net Mg2+ uptake.     

Mg2+ efflux is accomplished by an amiloride-sensitive Na+/Mg2+ antiport

Mg2+ efflux from Mg2+-preloaded chicken erythrocytes is caused by an electroneutral Na+/Mg2+ antiport. It depends specifically on extracellular Na+, according to Michaelis-Menten kinetics (Km = 25 mM), and is reversibly noncompetitively inhibited by amiloride (Ki = 0.59 mM). In contrast to Na+/H+ antiport, Li+, Ca2+ and N-ethylmaleimide do not interfere with Na+/Mg2+ antiport. The Na+/Mg2+ antiport is driven by the intracellular/extracellular Mg2+ gradient.     

Characterization of Mg2+ efflux from human, rat and chicken erythrocytes

Net Mg2+ efflux from Mg2+-loaded, human, rat and chicken erythrocytes was measured in sucrose, NaCl and choline Cl medium. Thus, Na+-dependent (NaCl minus choline Cl) and Na+-independent Mg2+ efflux (in sucrose) were determined. Na+-dependent Mg2+ efflux amounted to 0.16, 8.9 and 1.57 mmol/l cells x 30 min, Na+-independent Mg2+ efflux amounted to 0.89, 1.55 and 0.37 mmol/l cells x 30 min for human, rat and chicken erythrocytes. Na+-dependent Mg2+ efflux was inhibited by quinidine. Na+-independent Mg2+ efflux was inhibited by SITS and Cl-. A small fraction of Na+-independent Mg2+ efflux (in choline Cl) was resistant to SITS and Cl-. Ca2+ loading increased Mg2+ efflux similar to K+ efflux (Gardos effect). This effect was differently expressed in human and chicken erythrocytes.     

Na+-independent Mg2+ efflux from Mg2+-loaded human erythrocytes

Net Mg2+ efflux from Mg2+-loaded human erythrocytes was maximal after reincubation in sucrose. Net Mg2+ efflux was not inhibited by furosemide or bumetanide and, therefore, was not performed by the (Na,K,Cl)- or (K,Cl)-cotransport system. A component of net Mg2+ efflux was inhibited by extracellular NaC1, KCl, LiCl, choline Cl and SITS, in analogy to the inhibition of net Cl- and SITS. Therefore, it was concluded that net Mg2+ efflux is dependent on net Cl- efflux for charge compensation. Cl- -dependent net Mg2+ efflux was inhibited by amiloride. Only 10% of the maximal net Mg2+ efflux may depend on extracellular Na+.     

Activation of Na+/Mg2+ antiport in thymocytes by cAMP.

Mg2+ efflux from Mg(2+)-loaded rat thymocytes was stimulated by 0.1 mM dibutyryl cAMP (db cAMP). The activation of Mg2+ efflux by db cAMP was more expressed at lower Mg(2+)-loading. cAMP induced only a very small increase in the concentration of intracellular free Mg2+ which cannot explain the activation of Na+/Mg2+ antiport. From these results it was concluded that cAMP increases the affinity of the Na+/Mg2+ antiporter for intracellular Mg2+, probably by phosphorylation.     

Characterization of Na+/Mg2+ antiport by simultaneous 28Mg2+ influx

During net Mg2+ efflux from Mg2+-preloaded chicken erythrocytes, which occurs via Na+/Mg2+ antiport, 28Mg2+ is taken up intracellularly. Km of 28Mg2+ influx amounted to 1 mM. In Na+-free medium Vmax of 28Mg2+ influx was increased and Km was reduced to 0.2 mM. 28Mg2+ influx was noncompetitively inhibited by amiloride as was found for Na+/Mg2+ antiport. The results indicate that, extracellularly, Mg2+ can compete with Na+ for common binding sites of the Na+/Mg2+ antiporter, resulting in 28Mg2+-24Mg2+ exchange. The rate of Mg2+ exchange depends on extracellular Na+ and on the rate of net Mg2+ efflux.     

Na+/Ca2+ antiport in cultured arterial smooth muscle cells. Inhibition by magnesium and other divalent cations

Cultured smooth muscle cells from rat aorta were loaded with Na+, and Na+/Ca2+ antiport was assayed by measuring the initial rates of 45Ca2+ influx and 22Na+ efflux, which were inhibitable by 2',4'-dimethylbenzamil. The replacement of extracellular Na+ with other monovalent ions (K+, Li+, choline, or N-methyl-D-glucamine) was essential for obtaining significant antiport activity. Mg2+ competitively inhibited 45Ca2+ influx via the antiporter (Ki = 93 +/- 7 microM). External Ca2+ or Sr2+ stimulated 22Na+ efflux as would be expected for antiport activity. Mg2+ did not stimulate 22Na+ efflux, which indicates that Mg2+ is probably not transported by the antiporter under the conditions of these experiments. Mg2+ inhibited Ca2+-stimulated 22Na+ efflux as expected from the 45Ca2+ influx data. The replacement of external N-methyl-D-glucamine with K+, but not other monovalent ions (choline, Li+), decreased the potency of Mg2+ as an inhibitor of Na+/Ca2+ antiport 6.7-fold. Other divalent cations (Co2+, Mn2+, Cd2+, Ba2+) also inhibited Na+/Ca2+ antiport activity, and high external potassium decreased the potency of each by 4.3-8.6-fold. The order of effectiveness of the divalent cations as inhibitors of Na+/Ca2+ antiport (Cd2+ greater than Mn2+ greater than Co2+ greater than Ba2+ greater than Mg2+) correlated with the closeness of the crystal ionic radius to that of Ca2+.     

A one-to-one Mg2+:Mn2+ exchange in rat erythrocytes

Mg2+ efflux in rat erythrocytes was stimulated by increases in external Na+ concentration following a Michaelian-like function with an apparent dissociation constant (KNa) of 11 +/- 3 mM (mean +/- S.D. of three experiments) and a variable maximal rate ranging from 150 to 1200 mumol (liter (1) cells X h)-1. Na+-stimulated Mg2+ efflux was inhibited by quinidine and by ATP depletion. In the absence of external Na+, Mg2+ efflux was stimulated by increases in external Mn2+ concentration following a Michaelian-like function with an apparent dissociation constant (KMn) of 35 +/- 15 microM (mean +/- S.D. of four experiments) and a variable maximal rate ranging from 350 to 1400 mumol (1 cells X h)-1. Mn2+-stimulated Mg2+ efflux was inhibited by quinidine, by ATP depletion, and by increasing the external Na+ concentration. Quinidine-sensitive (or ATP-dependent) Mg2+ efflux exhibited very similar values when compared with quinidine-sensitive (or ATP-dependent) Mn2+ influx. Mn2+ efflux in rat erythrocytes (loaded with total internal Mn2+ contents of 230-450 mumol/l cells) was stimulated by increases in external Na+ concentration and inhibited by quinidine. In the absence of external Na+, Mn2+ efflux was stimulated by increases in external Mg2+ concentration following a Michaelian-like function with an apparent dissociation constant (KMg) of about 35 +/- 5 microM (mean +/- range of two experiments) and a maximal rate of about 60-100 mumol (1 cells X h)-1. In conclusion, the Na+-stimulated Mg2+ carrier of rat erythrocytes may catalyze a one-to-one and reversible Mn2+:Mg2+ exchange in the absence of external Na+.     

Activation of sodium transport in rat erythrocytes by inhibition of protein phosphatases 1 and 2A

Four structurally different protein phosphatases (PPs) inhibitors - fluoride, calyculin A, okadaic acid and cantharidin--were tested for their ability to modulate unidirectional Na(+) influx in rat red blood cells. Erythrocytes were incubated at 37 degrees C in isotonic and hypertonic media containing 1 mM ouabain and (22)Na in the absence or presence of PP inhibitors. Exposure of the cells to 20 mM fluoride or 50 nM calyculin A for 1 h under isosmotic conditions caused a significant stimulation of Na(+) influx, whereas addition of 200 microM cantharidin or 100 nM okadaic acid had no effect. After 2 h of treatment, however, all these PPs blockers significantly enhanced Na(+) transport in rat erythrocytes. Selective inhibitors of PP-1 and PP-2A types, calyculin A, cantharidin and okadaic acid, produced similar ( approximately 1.2-1.4-fold) stimulatory effects on Na(+) influx in the cells. Activation of Na(+) influx was unchanged with increasing calyculin A concentration from 50 to 200 nM. No additive stimulation of Na(+) influx was observed when the cells were treated with combination of 20 mM fluoride and 50 nM calyculin A. Na(+) influx induced by PPs blockers was inhibited by 1 mM amiloride and 200 muM bumetanide approximately in the equal extent, indicating the involvement of Na(+)/H(+) exchange and Na-K-2Cl cotransport in sodium transport through rat erythrocytes membrane. Activation of Na(+) transport in the cells induced by calyculin A and fluoride was associated with increase of intracellular Na(+) content. Shrinkage of the rat erythrocytes resulted in 2-fold activation of Na(+) influx. All tested PPs inhibitors additionally activated the Na(+) influx by 70-100% above basal shrinkage-induced level. Amiloride and bumetanide have diminished both the shrinkage-induced and PPs-inhibitors-induced Na(+) influxes. Thus, our observations clearly indicate that activities of Na(+)/H(+) exchanger and Na-K-2Cl cotransporter in rat erythrocytes are regulated by protein phosphatases and stimulated when protein dephosphorylation is inhibited.     

Characterization of Mg(2+) efflux from rat erythrocytes non-loaded with Mg(2+).

Non-Mg(2+)-loaded rat erythrocytes with a physiological level of Mg(2+)(i) exhibited Mg(2+) efflux when incubated in nominally Mg(2+)-free media. Two types of Mg(2+) efflux were shown: (1) An Na(+)-dependent Mg(2+) efflux in NaCl and Na gluconate medium, which was inhibited by amiloride and quinidine, as was Na(2+)/Mg(2+) antiport in Mg(2+)-loaded rat erythrocytes; and (2) an Na(+)-independent Mg(2+) efflux in sucrose medium and choline Cl medium, which may be differentiated into SITS-sensitive Mg(2+) efflux at low Cl(-)(o) (in sucrose) and into SITS-insensitive Mg(2+) efflux at high Cl(-)(o) (in 150 mmol/l choline Cl).     

Volume-dependent regulation of ion carriers in human and rat erythrocytes: role of cytoskeleton and protein phosphorylation

This study examines the effect of heat-induced cytoskeleton transitions and phosphoprotein phosphatase inhibitors on the activity of shrinkage-induced Na+, K+, 2Cl- cotransport and Na+/H+ exchange in rat erythrocytes and swelling-induced K+, Cl- cotransport in human and rat blood cells. Preincubation of human and rat erythrocytes at 49 degrees C drastically activated K+, Cl- cotransport and completely (rat) or partly (human) abolished its volume-dependent regulation. The same procedure did not affect basal activity of Na+, K+, 2Cl- cotransport but completely abolished its activation by shrinkage thus suggesting the involvement of a thermosensitive element of cytoskeleton network in the volume-dependent regulation of cotransporters. Both the shrinkage- and electrochemical proton gradient-induced Na+/H+ exchange was inhibited by the heat treatment to the same extent (50-70%), thus indicating the different signaling pathways involved in the activation of Na+, K+, 2Cl- cotransport and Na+/H+ exchange by cell shrinkage. This suggestion is in accordance with data on the different kinetics of volume-dependent activation and inactivation of these carriers as well as on their sensitivity to medium osmolality. Both swelling- and heat-induced increments of K+, Cl- cotransport activity were diminished by inhibitors of phosphoprotein phosphatases (okadaic acid and calyculin). In rat erythrocytes these compounds potentiate shrinkage-induced Na+/H+ exchange. On the contrary, neither basal nor shrinkage-induced Na+, K+, 2Cl- cotransport was affected by these compounds. Our results indicate a key role of cytoskeleton network in volume-dependent activation of K+, Cl- and Na+, K+, 2Cl- cotransport and the involvement of protein phosphorylation-dephosphorylation cycle in regulation of the activity of K+, Cl- cotransport and Na+/H+ exchange.     

Osmotic and phorbol ester-induced activation of Na+/H+ exchange: possible role of protein phosphorylation in lymphocyte volume regulation

The Na+/H+ antiport is stimulated by 12-O-tetradecanoylphorbol-13, acetate (TPA) and other phorbol esters in rat thymic lymphocytes. Mediation by protein kinase C is suggested by three findings: (a) 1-oleoyl-2-acetylglycerol also activated the antiport; (b) trifluoperazine, an inhibitor of protein kinase C, blocked the stimulation of Na+/H+ exchange; and (c) activation of countertransport was accompanied by increased phosphorylation of specific membrane proteins. The Na+/H+ antiport is also activated by osmotic cell shrinking. The time course, extent, and reversibility of the osmotically induced and phorbol ester-induced responses are similar. Moreover, the responses are not additive and they are equally susceptible to inhibition by trifluoperazine, N-ethylmaleimide, and ATP depletion. The extensive analogies between the TPA and osmotically induced effects suggested a common underlying mechanism, possibly activation of a protein kinase. It is conceivable that osmotic shrinkage initiates the following sequence of events: stimulation of protein kinase(s) followed by activation of the Na+/H+ antiport, resulting in cytoplasmic alkalinization. The Na+ taken up through the antiport, together with the HCO3- and Cl- accumulated in the cells as a result of the cytoplasmic alkalinization, would be followed by osmotically obliged water. This series of events could underlie the phenomenon of regulatory volume increase.     

Role of phospholipase C and protein kinase C in <Emphasis Type="Italic">Aspergillus nidulans</Emphasis> during growth on pectin or glucose: Effects on germination and duplication cycle

The effects of PLC and Pkc inhibitors on Aspergillus nidulans depend on the carbon source. PLC inhibitors Spm and C48/80 delayed the first nuclear division in cultures growing on glucose, but stimulated it in media supplemented with pectin. Less intense were these effects on the mutant transformed with PLC-A gene rupture (AP27). Neomycin also delayed the germination in cultures growing on glucose or pectin; however, on glucose, the nuclear division was inhibited whereas in pectin it was stimulated. These effects were minor in AP27. The effects of Ro-31-8425 and BIM (both Pkc inhibitors) were also opposite for cultures growing on glucose or pectin. On glucose cultures of both strains BIM delayed germination and the first nuclear division, whereas on pectin both parameters were stimulated. Opposite effects were also detected when the cultures were growing on glucose or pectin in the presence of Ro-31-8425.     

22Na+ fluxes in thymic lymphocytes. I. Na+/Na+ and Na+/H+ exchange through an amiloride-insensitive pathway

The Na+ transport pathways of normal rat thymocytes were investigated. Na+ conductance was found to be lower than K+ conductance, which is consistent with reported values of membrane potential. In contrast, the isotopically measured Na+ permeability was greater than 10-fold higher than that of K+, which indicates that most of the flux is electroneutral. Cotransport with Cl- (or K+ and Cl-) and countertransport with Ca2+ were ruled out by ion substitution experiments and use of inhibitors. Countertransport for Na+ or H+ through the amiloride-sensitive antiport accounts for only 15-20% of the resting influx. In the presence of amiloride, 22Na+ uptake was increased in Na+-loaded cells, which suggests the existence of Na+/Na+ countertransport. Cytoplasmic pH determinations using fluorescent probes indicated that under certain conditions this amiloride-resistant system will also exchange Na+ for H+, as evidenced by an internal Na+- dependent acidification is proportional to internal [Na+] but inversely related to extracellular [Na+]. Moreover, 22Na+ uptake is inhibited by increasing external [H+]. The results support the existence of a substantial amiloride-insensitive, electroneutral cation exchange system capable of transporting Na+ and H+.     

Regulation of Na+/H+ and Cl-/HCO3- antiports in Vero cells

The effect of serum, phorbol-12-myristate-13-acetate (TPA), and forskolin on the activity Na+/H+ antiport and the Na(+)-coupled and Na(+)-independent Cl-/HCO3- antiport was studied in Vero cells by measuring 22Na+ and 36Cl- fluxes and changes in cytosolic pH (pHi). The Na(+)-independent Cl-/HCO3- antiport, which acts as an acidifying mechanism, is strongly pH-sensitive. In serum-starved cells it is activated at alkaline cytosolic pH, with a half-maximal activity at pHi approximately 7.20. Incubation with serum increased the activity of the Na(+)-independent Cl-/HCO3- antiport at pHi values from 6.8 to 7.2. Thus serum appeared to alter the pHi sensitivity of this antiporter such that the threshold value for activation of the antiport was shifted to a more acidic value. Na+/H+ antiport was somewhat stimulated initially by addition of serum, but further incubation with serum (greater than 45 min) decreased its activity. The activity of the Na(+)-coupled Cl-/HCO3- antiport, which is the major alkalinizing antiport in Vero cells, was not altered by short-term incubation with serum (less than 10 min) but decreased after prolonged incubation (greater than 45 min). Our findings with TPA and forskolin indicate that the effect of serum is partly mediated by the protein kinase C pathway, whereas the cyclic adenosine monophosphate pathway does not appear to play an important role. The net effect of serum on the pHi-regulating antiports was a slight decrease in intracellular pH.     

Effect of protein kinase C activation on Na+-H+ exchange in erythrocytes of frog Rana temporaria

The treatment of frog erythrocytes incubated in standard nitrate medium with 100 nM phorbol ester (PMA) induced a sharp increase in the 22Na uptake by the cells and intracellular Na(+) concentration. The PMA-induced enhancement in 22Na uptake was stimulated by the addition of 0.1 mM ouabain to the incubation medium and completely blocked by 1 mM amiloride. The time course of 22Na uptake by frog red cells in the presence of PMA showed a lag phase ( approximately 5 min), after which was linear within 5-15 min. The calculated Na(+) influx in erythrocytes treated with PMA was 49.4+/-3.7 mmol l(-1) cells h(-1) as compared with 1.2+/-0.25 mmol l(-1) h(-1) for control cells. 5-(N-ethyl-N-isopropyl)-amiloride, selective blocker of NHE1, caused a dose-dependent inhibition of the PMA-induced Na(+) influx with IC(50) of 0.27 microM. The PMA-induced Na(+) influx was almost completely inhibited by 0.1 microM staurosporine, protein kinase C blocker. Pretreatment of frog red blood cells for 5, 10 or 15 min with 10 mM NaF, non-selective inhibitor of protein phosphatase, led to a progressive stimulation of the PMA effect on Na(+) influx. Both amiloride and NaF did not affect the basal Na(+) influx in frog erythrocytes. The data indicate that the Na(+)-H(+) exchanger in the frog erythrocytes is quiescent under basal conditions and can be markedly stimulated by PMA.     

Role of a Na+-dependent Cl-/HCO3- exchange in regulation of intracellular pH in fibroblasts

We previously reported that, in a HCO3(-)-free medium, cytoplasmic pH (pHi) of hamster fibroblasts (CCL39) is primarily regulated by an amiloride-sensitive Na+/H+ antiport (L'Allemain, G., Paris, S., and Pouysségur, J. (1984) J. Biol. Chem. 259, 5809-5815). Here we demonstrate the existence of an additional pHi-regulating mechanism in CCL39 cells, namely a Na+-dependent HCO3-/Cl- exchange. Evidence for this system is based on 36Cl- influx studies and on pHi measurements in PS120, a CCL39-derived mutant lacking the Na+/H+ antiport activity. 36Cl- influx rate is a saturable function of external [Cl-] (apparent Km approximately equal to 7 mM), is competitively inhibited by external HCO3- (KI approximately equal to 3 mM), and by stilbene derivatives (KI approximately equal to 20 microM for 4-acetamido-4'-isothiocyanostilbene-2,2'-disulfonic acid). Measurements of pHi recovery after an acute acid load indicate that PS120 cells possess an acid-extruding mechanism dependent on external HCO3-, which is inhibited by stilbene derivatives and requires external Na+. Since 22Na+ influx is stimulated upon addition of HCO3- to acid-loaded cells and this effect is completely abolished by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid, we conclude that Na+ is co-transported with HCO3-, in exchange for intracellular Cl-. In a HCO3(-)-containing medium, this pHi-regulating mechanism appears to have two essential physiological functions for the Na+/H+ antiport-deficient mutant: protection of the cells against excessive cytoplasmic acidification and establishment of a steady-state pHi permissive for growth, at neutral or slightly acidic pHo values (6.6-7.2).     

Further studies on alterations in magnesium binding during cold storage of erythrocytes

Free intracellular Mg2+ concentration ([Mg2+]i) was measured in cold-stored human erythrocytes by the method of null-point titration with ionophore A23187. [Mg2+]i was 311 +/- 41 microM (mean +/- S.D.) for cells stored 0-10 days, increasing to 458 +/- 64 microM for cells stored 22-48 days. The values for stored cells were higher than those previously determined by a 31P-NMR method (Bock et al. (1985) Blood 65, 1526-1530); however, the null-point method requires extensive washing of the cells, which we have found to increase NMR-measured [Mg2+]i. The null-point values still represent a small fraction of total cell Mg2+, and confirm that binding of Mg2+ to ligands other than ATP and 2,3-bisphosphoglycerate must increase during storage. As an initial test of whether this may imply suboptimal availability of Mg2+ for cell preservation, we used A23187 to prepare erythrocytes with altered Mg2+ content, then removed ionophore and stored the cells in plasma-free medium for up to 2 weeks. Higher Mg2+ content had a very significant positive correlation (P less than 0.0001) with higher cell ATP concentrations. Storage did not significantly affect basal or Na+-stimulated efflux of Mg2+ from Mg2+-loaded red cells.