Biochhnica et Biophysica Acta, 1094 (1991) 51-54 © 1991 Elsevier Science Publishers B.V. 0167-4889/91/$03.50 ADONIS 0167488991002136
51
BBAMCR 12981
Relationship between total magnesium concentration and free intracellular magnesium in sheep red blood cells Hiroshi Fujise
1.,
Phillip Cruz 2 Nicholas V. Reo 2 and Peter K. Lauf 1
; Department of Physiology and Biophysics, 2 Department of Biochemistry, and Kettering-Scott Magnetic Resonance Laboratory, Wright State Unil'ersity School of Medicine and Kettering Medical Center, Dayton, OH (U.S.A.) (Received 2 October 1990) (Revised manuscript received 7 May 1991)
Key words: Magnesium ion, intracellular; NMR; (Sheep red blood cell)
The cellular free magnesium concentration of ionophore A23187 permeabilized high potassium sheep erythrocytes was measured by 3~p nuclear magnetic resonance spectroscopy, and the total cellular magnesium concentration was determined by atomic absorption spectroscopy. The free versus total cellular magnesium concentrations yield a linear relationship on a log-log scale in the concentration range from 0.3 to 1.92 mmol Mg/liter cells. Thus, free intracellular magnesium concentrations can be calculated from atomic absorption data. The method permits the estimation of physiologically or experimentally induced variations of intracellular free magnesium concentrations between 7 and 405 /zM magpes~am in cell water. This range encompasses the free magnesium concentration of 335 + 6 0 / z M in cell water de:ermined for untreated erythroeytes.
Introduction
Tissue and s e r u m M g 2+ levels are important determinants of several physiological and pharmacological responses [1-4] and have been related to several diseases [5-10]. Intracellular M g 2+ is well known as a cofactor for many enzymes and membrane transporters [1,3,4]. For example, the Na, K-pump of the red cell membrane requires Mg 2+ as an activator [11,12] and M g 2+ m a y regulate K/Cl cotransport [13]. The ionized magnesium levels are clearly more important than the total cellular magnesium content when investigating the effect of this species on cellular functions [6,8,9], and are experimentally more difficult to measure. Previously, the concentration of ionized magnesium in red cells was estimated using adenylate kinase equilibrium in intact cells and red cell lysates, since this enzyme requires M g 2+ as an activator [14]. An alterna-
* Current address: Department of Pathology, School of Veterinary Medicine, Azabu University, Fuchinobe, Sagamihara, Kanagawa 229, Japan. Correspondence: P.K. Lauf, Department of Physiology and Biophysics, Wright State University School of Medicine, Dayton, OH 45435, U.S.A.
tive approach is provided by equilibrating the medium with the cellular compartment by treating the cells with the ionophore A23187, EDTA, and Mg 2+ [15]. More recently, phosphorus-31 nuclear magnetic resonance (NMR) spectroscopy was used to determine the free (ionized) cellular magnesium concentration in red cells [16]. Using this method, several investigators have shown a correlation between free magnesium levels and certain physiological red cell disorders [8,9,17]. In the present study, the relationship between the total magnesium content, [Mg2+] T, as determined by atomic absorption spectroscopy, and the free cellular m a g n e s i u m , [Mg2+] F, as measured by aIp-NMR was investigated in sheep red cells permeabilized for Mg 2+ with the divalent cation ionophore A23187.
Materials and Methods
Blood was drawn by jugular venipuncture from mixed-bred HK sheep using heparin as anticoagulant. Cells were washed in an isosmotic solution (pH 7.4 and 0 ° C ) of the following composition [11]: 150 n~t'~ Nmethyl_D.glucamine-NO3 (NMDG-NO3); 10 mM tris(hydroxymethyl)aminomethane-3-(N-morpholino)propanesulfonic acid (Tris-Mops). This preparation is referred to as untreated cells.
52 In order to alter the cellular Mg 2+ concentration, the cells were treated with the ionophore A23187 in the presence or absence of ethylenediaminetetraacetic acid (EDTA) or added Mg 2+ as described previously [11]. After adjusting the cellular Mg 2÷ concentrations, the cells were suspended (50% v/v) in 10 mM TrisMops and 150 mM NMDG-NO 3 at pH 7.4 and 37 o C. This preparation is referred to as treated cells. The cell suspension was transferred into a 10 mm NMR tube. A capillary tube. containing hexachlorocyclotriphosphazene (HCTP) was also placed into the NMR tube to serve as an external atp chemical shift reference [18], Phosphorus-31 NMR spectra were recorded at a centerband frequency of 145.8 MHz on a Bruker AM 360 NMR spectrometer without sample spinning or field lock. The temperature of the probe was maintained at 37"C. Two thousand transients were recorded for each spectrum using 60 ° pulses at 0.48 second intervals (18 rain per analysis). The intracellular free magnesium concentration was calculated usir.g the method of Gupta [9] as follows. The chemical shifts of the a-phosphate (P,) and /3phosphate (P~) resonances of ATP depend on the extent of ATP complex formation with Mg 2+ ion. The separation between the P~ and PB resonances (8,~/3)was measured for ATP alone ( 8 ATP , , ~ ) and Mg-saturated ATP /~Ms'ATP~ ,,,~,Z ,. The fraction of free ATP to total ATP concentration in the cell preparation, ~ ffi [ATP]F/[ATP]T, can be determined from these values for 8 ~ at the extreme concentration limits and the separation between the P,, and Pz resonances for the given cell preparation (8c~u) according to the following equation: t~ ~ 1¢C¢11
.¢ M g , A T P ~ / / ' .~ ATP
J~Mg'ATP~
Results and Discussion
Fig. 1 shows the 31P-NMR spectra of sheep red ceils upon varying magnesium concentration as described in Materials and Methods. As reported previously [9], increasing [Mg2+] F causes a downfield shift in the ATP resonances and the separation between t h e P,~ and Pt3 resonances, (8~z), decreases. The Pa resonance is narrow at both low magnesium concentration and when ATP is saturated with magnesium, but broadens at intermediate magnesium concentrations. This broadening effect is due to chemical exchange between free ATP and Mg. ATP complex, which is in the intermediate exchange kinetics regime at the experimental resonance frequency [20]. The high spectral resolution obtained in these experiments enabled an accurate mea-
~ ~
T P-
(I)
The free Mg 2+ was then calculated from the value of 4} and the dissociation constant, K d, of the [Mg. ATP] complex: [Mg~* ]~
"- Kd(~,-'
-
l)
(2)
The value used for K d (3.57.10 -5 M at pH 7.4) was that given by Gupta et al. [19] at 37 *C and pH 7.2, and corrected for pH change by the method of Book et al.
[ZT].
Total cellular magnesium, [MgZ+]T, was measured on a Perkin-Elmer Model 5000 Atomic Absorption Spectrophotometer according to the method of Flatman and Lew [15], This concentration is expressed in mmoi/I cells based on the hemoglobin concentration [11]. Experimental measurements (NMR and atomic absorption) were conducted on untreated cells prepared from five independently obtained blood samples. Four of these blood samples were used to prepare treated cells for experimental analyses.
• -~a
' -~o
' -~2
' -~,
" -~6
' -~.
PPH
' -~o
' -;2
' -~4
" -~6
' -2,. '
Fig. 1. 3zp-NMR spectra at 8.5 tesla of ATP resonances from HK sheep red blood cells at 37 ° C. These spectra illustrate the change in chemical shift with varying amounts of free magnesium, increasing from zero magnesium (bottom spectrum) to saturated magnesium (top spectrum). Spectra were processed using 8K total data points and a 30 Hz exponential filter. Chemical shifts are relative to the HCTP resonance which was set to zero ppm.
53 sure of 6~# even in the presence of slight broadening of the P~ resonance. Previous reports showed that deoxyhemoglobin (but n o t oxyhemoglobin) binds ATP and reduces 6 ~ [21]. In this present study, the high magnetic field (8.5 tesla) provides good sensitivity and, thus, the short time required for data acquisition (18 rain) ensures aerobic conditions throughout the analysis. This was confirmed by obtaining NMR spectra while stirring with an air driven screw [22] and bubbling 95% 0 2 / 5 % CO 2 through the cell suspension. The results of these experiments showed n o difference in the separation of the P~ and P~ peaks in comparison with spectra obtained without stirring and 0 2 gas bubbling; however, a slight increase (approx. 10%) in NMR sensitivity was obse~ed for spectra in which the sample was stirred and supplied with 0 2. This sensitivity difference may be attributed to a partial sedimentation of the cells in the absence of stirring. Since stirring and gas bubbling can sometimes cause the cells to hemolyze, all subsequent NMR experiments were performed without these procedures. The cellular free magnesium ion concentration as determined by 31P-NMR in untreated (void of ionophore A23187, EDTA, and added Mg 2+) sheep red cells was 335 + 60/.tM (n = 5). In comparison, the total magnesium concentration as measured by atomic
2. E
absorption spectrophotometry in the untreated cells was 1.8 + 0.08 mmol/l cells (n = 5). The [Mg2÷] F in HK sheep red cells is similar to the value obtained in human red cells (214 + 7 #M) as measured by 31P-NMR [9,15,26] and to that obtained for LK sheep red cells by another technique [27]. Total magnesium, [Mg2+]x, was varied from 0.04 to 3.6 mmol/! cells. A log-log plot of [Mg2+] F versus [Mg2+] T yields a straight line over the range in total magnesium concentration from 0.3 to 1.92 mmol/l of cells (corresponding to a range in free Mg from 7 to 405 /~M) as shown in Fig. 2. These data have the correlation equation: iog[Mg2+ ]F = 2.11-log[Mg 2+ ]T+ 1.91 (r = 0.984)
Beyond these limits in [Mg2+]T, deviations from linearity are observed. Inaccuracy in the NMR-measured free magnesium at concentrations far from the dissociation constant (Kd) may account for these deviations. There has been some controversy in the literature regarding the value of K d [23-25]; however, the linear relationship between the NMR data and atomic absorption data is independent of the value of K d. Indeed, a linear relationship exists between the [Mg2+]T and either of the two variables in Eqn. 2 (i.e., [Mg2+] F or (~b- ' - 1)). The K d is just a proportionality constant relating [Mg2+] F to (~b-I - 1) as indicated in Eqn. 2. Thus, Eqn. 3 can be recast in the form: i o g ( ~ - I _ 1) = 2.11 "log[Mg 2+ ]T+ 6.36
.""
6
iog[Mg 2+ ]F = 2.1 I. Iog[Mg 2 + ]T + log K d + 6.36 v
1.4
1.¢
-0'.4 Log
'
-0'.2
Total
' Mg
010 (mmol/L
'
012
(4)
by substitution from Eqn. 2 using 3.57.10 -5 M for the value of K a. A more general form of this equation can be written as:
• o
'
(3)
'
0.4
cell)
Fig. 2. Log-log plot of total magnesium measured by atomic absorption spectroscopy verus free magnesium as calculated from NMR data and Eqns. 1 and 2. These data represent total magnesium concentrations varying from 0.3 to 1.92 mmol/I cells. The result of a linear regression analysis is depicted by the solid line and yields Eqn. 3 in the text.
(5)
With this equation, ary value of K d can be used in conjunction with measured values of [Mg2+] T to yield the free cellular magnesium concentration. Conversely, if one can independently measure both total and free magnesium without knowledge of the K d, then this relationship represents a means to obtain a value for the dissociation constant of the [Mg. ATP] complex. Based upon Eqn. 5, it is possible to use [Mg2+]T as measured by atomic absorption spectrophotometry to obtain the free magnesium concentration. This should facilitate examination of the effect of changes in magnesium concentration on red cell functions such as the Na, K-pump activity and the binary and ternary cotransport mechanisms. The range of concentration over which the above relationship is valid encompasses the physiological and pathological (hypomagnesemia) total magnesium concentrations. The authors caution, however, that this correlation between free and total Mg 2.
54 may only be valid for the particular system studied, namely sheep erythrocytes in conjunction with ionophore A23187. The applicability of this procedure has not been tested under various pathological conditions or in red ceils from other species. It is possible that a different correlation equation may exist for erythrocytes from other species, since the buffering mechanisms associated with free magnesium may be species dependent. Further studies are necessary to define the extent of applicability of our observations. References I Murphy, R.A,, Bohr, D.F. and Newman, D.L. (1969) Am. J. Physiol. 217. 666-673. 2 Meissner, G. and Henderson, J.S, (1987) J. Biol. Chem. 262, 3065-3073. 3 Skou, J. (1975) Q, Rev. Biophys. 7, 401-434. 4 Wacker, W,E.C, (1987) Magnesium 6, 61-64. 5 Altura, B,M, and AItura, B.T. (1981) Fed. Proc. Exp, Am. Soc. Exp, Biol. 40, 2672-2679. 6 Corkey, B.E., Duszynski, J., Rich, T.L., Matschins~, B. and Williamson, J,R, (1986) J. Biol. Chem. 261, 2567-2574. 7 Turlapaty, P.D.M.V. and AItura, B.M. (1980) Science 208, 198200. 8 Matsuura, T., Kohno, M., Kanayama, Y., Yasunari, K., Murakawa, K., Takeda, T., Ishimori, K., Morshima, I. and Yonezawa, T. (1987) Biochem. Biophys. Res. Commun. 143, 1012-1017. 9 Resnick, L.M., Gupta, R.K. and Larage, J.H. (1984) Proc. Natl. Acad. Sci. USA 81, 6511-6515.
10 Anderson, T.W., Neri, L.C., Schreiber, G., Talbot, F.D.F. and Zdrejewski, A. (1975) Can. Med. Assoc. J. 113, 199-203. 11 Fujise, H. and Lauf, P.K. (1988) .l. Physiol. 405, 605-614. 12 Flatman, P.W. and Lew, V.L. (1981) J. Physiol. 315, 421-446. 13 Lauf, P.K. (1988) Am. J. Physiol. 255 (Cell Physiol. 24), C331C339. 14 Rose, I.A. (1968) Proc. Natl. Acad. Sci. USA 61, 1079-1086. 15 Flatman, P.W. and Lew, V.L. (1980) J. Physiol. 305, 13-30. !6 Gupta, R.K., Benovic, J.L. and Rose, Z.B. (1978) J. Biol. Chem. 253, 6172-6176. 17 Bock, J.L., Wenz, B. and Gupta, R.K. (1985) Blood 65, 1526-1530. 18 Gard, J.K. and Ackerman, J.J.H. (1983) J. Magn. Reson. 51, 125-127. 19 Gupta, R.K. and Benovic, J.L. (1978) Biochem. Biophys. Res. Commun. 84, 130-137. 20 Misawa, K., Lee, T.M. and Ogawa, S. (1982) Biochim. Biophys. Acta 718, 227-229. 21 Gupta, R.K., Benovic, J.L. and Rose, Z.B. (1978) J. Biol. Chem. 253, 6165-6171. 22 Murphy, E., Levy, L., Berkowitz, L.R., Orringer, E.P., Gabel, S.A. and London, R.E. (1986) Am. J. Physiol. 251, C496-C504. 23 Gupta, R.K. and Moore, R.D. (1980) J. Biol. Chem. 255, 39873993. 24 Wa, S.T., Pieper, G.M., Salhany, J.M. and Eliot, R,S. (1981) Biochemistry 20, 7399-7403. 25 Garfinkel, L. and Garfinkel, D. (1984) Biochemistry 23, 35473552. 26 Gupta, R.K. and Gupta, P. (1987) NMR Spectroscopy of Cells and Organisms (Gupta, R.K., ed.), Vol. 2, Chap. 8, p. 39. CRC Press, Boca Raton, FL. 27 Delpire, E. and Lauf, P.K. (1991) J. Physiol. (Lond.), in press.
51
BBAMCR 12981
Relationship between total magnesium concentration and free intracellular magnesium in sheep red blood cells Hiroshi Fujise
1.,
Phillip Cruz 2 Nicholas V. Reo 2 and Peter K. Lauf 1
; Department of Physiology and Biophysics, 2 Department of Biochemistry, and Kettering-Scott Magnetic Resonance Laboratory, Wright State Unil'ersity School of Medicine and Kettering Medical Center, Dayton, OH (U.S.A.) (Received 2 October 1990) (Revised manuscript received 7 May 1991)
Key words: Magnesium ion, intracellular; NMR; (Sheep red blood cell)
The cellular free magnesium concentration of ionophore A23187 permeabilized high potassium sheep erythrocytes was measured by 3~p nuclear magnetic resonance spectroscopy, and the total cellular magnesium concentration was determined by atomic absorption spectroscopy. The free versus total cellular magnesium concentrations yield a linear relationship on a log-log scale in the concentration range from 0.3 to 1.92 mmol Mg/liter cells. Thus, free intracellular magnesium concentrations can be calculated from atomic absorption data. The method permits the estimation of physiologically or experimentally induced variations of intracellular free magnesium concentrations between 7 and 405 /zM magpes~am in cell water. This range encompasses the free magnesium concentration of 335 + 6 0 / z M in cell water de:ermined for untreated erythroeytes.
Introduction
Tissue and s e r u m M g 2+ levels are important determinants of several physiological and pharmacological responses [1-4] and have been related to several diseases [5-10]. Intracellular M g 2+ is well known as a cofactor for many enzymes and membrane transporters [1,3,4]. For example, the Na, K-pump of the red cell membrane requires Mg 2+ as an activator [11,12] and M g 2+ m a y regulate K/Cl cotransport [13]. The ionized magnesium levels are clearly more important than the total cellular magnesium content when investigating the effect of this species on cellular functions [6,8,9], and are experimentally more difficult to measure. Previously, the concentration of ionized magnesium in red cells was estimated using adenylate kinase equilibrium in intact cells and red cell lysates, since this enzyme requires M g 2+ as an activator [14]. An alterna-
* Current address: Department of Pathology, School of Veterinary Medicine, Azabu University, Fuchinobe, Sagamihara, Kanagawa 229, Japan. Correspondence: P.K. Lauf, Department of Physiology and Biophysics, Wright State University School of Medicine, Dayton, OH 45435, U.S.A.
tive approach is provided by equilibrating the medium with the cellular compartment by treating the cells with the ionophore A23187, EDTA, and Mg 2+ [15]. More recently, phosphorus-31 nuclear magnetic resonance (NMR) spectroscopy was used to determine the free (ionized) cellular magnesium concentration in red cells [16]. Using this method, several investigators have shown a correlation between free magnesium levels and certain physiological red cell disorders [8,9,17]. In the present study, the relationship between the total magnesium content, [Mg2+] T, as determined by atomic absorption spectroscopy, and the free cellular m a g n e s i u m , [Mg2+] F, as measured by aIp-NMR was investigated in sheep red cells permeabilized for Mg 2+ with the divalent cation ionophore A23187.
Materials and Methods
Blood was drawn by jugular venipuncture from mixed-bred HK sheep using heparin as anticoagulant. Cells were washed in an isosmotic solution (pH 7.4 and 0 ° C ) of the following composition [11]: 150 n~t'~ Nmethyl_D.glucamine-NO3 (NMDG-NO3); 10 mM tris(hydroxymethyl)aminomethane-3-(N-morpholino)propanesulfonic acid (Tris-Mops). This preparation is referred to as untreated cells.
52 In order to alter the cellular Mg 2+ concentration, the cells were treated with the ionophore A23187 in the presence or absence of ethylenediaminetetraacetic acid (EDTA) or added Mg 2+ as described previously [11]. After adjusting the cellular Mg 2÷ concentrations, the cells were suspended (50% v/v) in 10 mM TrisMops and 150 mM NMDG-NO 3 at pH 7.4 and 37 o C. This preparation is referred to as treated cells. The cell suspension was transferred into a 10 mm NMR tube. A capillary tube. containing hexachlorocyclotriphosphazene (HCTP) was also placed into the NMR tube to serve as an external atp chemical shift reference [18], Phosphorus-31 NMR spectra were recorded at a centerband frequency of 145.8 MHz on a Bruker AM 360 NMR spectrometer without sample spinning or field lock. The temperature of the probe was maintained at 37"C. Two thousand transients were recorded for each spectrum using 60 ° pulses at 0.48 second intervals (18 rain per analysis). The intracellular free magnesium concentration was calculated usir.g the method of Gupta [9] as follows. The chemical shifts of the a-phosphate (P,) and /3phosphate (P~) resonances of ATP depend on the extent of ATP complex formation with Mg 2+ ion. The separation between the P~ and PB resonances (8,~/3)was measured for ATP alone ( 8 ATP , , ~ ) and Mg-saturated ATP /~Ms'ATP~ ,,,~,Z ,. The fraction of free ATP to total ATP concentration in the cell preparation, ~ ffi [ATP]F/[ATP]T, can be determined from these values for 8 ~ at the extreme concentration limits and the separation between the P,, and Pz resonances for the given cell preparation (8c~u) according to the following equation: t~ ~ 1¢C¢11
.¢ M g , A T P ~ / / ' .~ ATP
J~Mg'ATP~
Results and Discussion
Fig. 1 shows the 31P-NMR spectra of sheep red ceils upon varying magnesium concentration as described in Materials and Methods. As reported previously [9], increasing [Mg2+] F causes a downfield shift in the ATP resonances and the separation between t h e P,~ and Pt3 resonances, (8~z), decreases. The Pa resonance is narrow at both low magnesium concentration and when ATP is saturated with magnesium, but broadens at intermediate magnesium concentrations. This broadening effect is due to chemical exchange between free ATP and Mg. ATP complex, which is in the intermediate exchange kinetics regime at the experimental resonance frequency [20]. The high spectral resolution obtained in these experiments enabled an accurate mea-
~ ~
T P-
(I)
The free Mg 2+ was then calculated from the value of 4} and the dissociation constant, K d, of the [Mg. ATP] complex: [Mg~* ]~
"- Kd(~,-'
-
l)
(2)
The value used for K d (3.57.10 -5 M at pH 7.4) was that given by Gupta et al. [19] at 37 *C and pH 7.2, and corrected for pH change by the method of Book et al.
[ZT].
Total cellular magnesium, [MgZ+]T, was measured on a Perkin-Elmer Model 5000 Atomic Absorption Spectrophotometer according to the method of Flatman and Lew [15], This concentration is expressed in mmoi/I cells based on the hemoglobin concentration [11]. Experimental measurements (NMR and atomic absorption) were conducted on untreated cells prepared from five independently obtained blood samples. Four of these blood samples were used to prepare treated cells for experimental analyses.
• -~a
' -~o
' -~2
' -~,
" -~6
' -~.
PPH
' -~o
' -;2
' -~4
" -~6
' -2,. '
Fig. 1. 3zp-NMR spectra at 8.5 tesla of ATP resonances from HK sheep red blood cells at 37 ° C. These spectra illustrate the change in chemical shift with varying amounts of free magnesium, increasing from zero magnesium (bottom spectrum) to saturated magnesium (top spectrum). Spectra were processed using 8K total data points and a 30 Hz exponential filter. Chemical shifts are relative to the HCTP resonance which was set to zero ppm.
53 sure of 6~# even in the presence of slight broadening of the P~ resonance. Previous reports showed that deoxyhemoglobin (but n o t oxyhemoglobin) binds ATP and reduces 6 ~ [21]. In this present study, the high magnetic field (8.5 tesla) provides good sensitivity and, thus, the short time required for data acquisition (18 rain) ensures aerobic conditions throughout the analysis. This was confirmed by obtaining NMR spectra while stirring with an air driven screw [22] and bubbling 95% 0 2 / 5 % CO 2 through the cell suspension. The results of these experiments showed n o difference in the separation of the P~ and P~ peaks in comparison with spectra obtained without stirring and 0 2 gas bubbling; however, a slight increase (approx. 10%) in NMR sensitivity was obse~ed for spectra in which the sample was stirred and supplied with 0 2. This sensitivity difference may be attributed to a partial sedimentation of the cells in the absence of stirring. Since stirring and gas bubbling can sometimes cause the cells to hemolyze, all subsequent NMR experiments were performed without these procedures. The cellular free magnesium ion concentration as determined by 31P-NMR in untreated (void of ionophore A23187, EDTA, and added Mg 2+) sheep red cells was 335 + 60/.tM (n = 5). In comparison, the total magnesium concentration as measured by atomic
2. E
absorption spectrophotometry in the untreated cells was 1.8 + 0.08 mmol/l cells (n = 5). The [Mg2÷] F in HK sheep red cells is similar to the value obtained in human red cells (214 + 7 #M) as measured by 31P-NMR [9,15,26] and to that obtained for LK sheep red cells by another technique [27]. Total magnesium, [Mg2+]x, was varied from 0.04 to 3.6 mmol/! cells. A log-log plot of [Mg2+] F versus [Mg2+] T yields a straight line over the range in total magnesium concentration from 0.3 to 1.92 mmol/l of cells (corresponding to a range in free Mg from 7 to 405 /~M) as shown in Fig. 2. These data have the correlation equation: iog[Mg2+ ]F = 2.11-log[Mg 2+ ]T+ 1.91 (r = 0.984)
Beyond these limits in [Mg2+]T, deviations from linearity are observed. Inaccuracy in the NMR-measured free magnesium at concentrations far from the dissociation constant (Kd) may account for these deviations. There has been some controversy in the literature regarding the value of K d [23-25]; however, the linear relationship between the NMR data and atomic absorption data is independent of the value of K d. Indeed, a linear relationship exists between the [Mg2+]T and either of the two variables in Eqn. 2 (i.e., [Mg2+] F or (~b- ' - 1)). The K d is just a proportionality constant relating [Mg2+] F to (~b-I - 1) as indicated in Eqn. 2. Thus, Eqn. 3 can be recast in the form: i o g ( ~ - I _ 1) = 2.11 "log[Mg 2+ ]T+ 6.36
.""
6
iog[Mg 2+ ]F = 2.1 I. Iog[Mg 2 + ]T + log K d + 6.36 v
1.4
1.¢
-0'.4 Log
'
-0'.2
Total
' Mg
010 (mmol/L
'
012
(4)
by substitution from Eqn. 2 using 3.57.10 -5 M for the value of K a. A more general form of this equation can be written as:
• o
'
(3)
'
0.4
cell)
Fig. 2. Log-log plot of total magnesium measured by atomic absorption spectroscopy verus free magnesium as calculated from NMR data and Eqns. 1 and 2. These data represent total magnesium concentrations varying from 0.3 to 1.92 mmol/I cells. The result of a linear regression analysis is depicted by the solid line and yields Eqn. 3 in the text.
(5)
With this equation, ary value of K d can be used in conjunction with measured values of [Mg2+] T to yield the free cellular magnesium concentration. Conversely, if one can independently measure both total and free magnesium without knowledge of the K d, then this relationship represents a means to obtain a value for the dissociation constant of the [Mg. ATP] complex. Based upon Eqn. 5, it is possible to use [Mg2+]T as measured by atomic absorption spectrophotometry to obtain the free magnesium concentration. This should facilitate examination of the effect of changes in magnesium concentration on red cell functions such as the Na, K-pump activity and the binary and ternary cotransport mechanisms. The range of concentration over which the above relationship is valid encompasses the physiological and pathological (hypomagnesemia) total magnesium concentrations. The authors caution, however, that this correlation between free and total Mg 2.
54 may only be valid for the particular system studied, namely sheep erythrocytes in conjunction with ionophore A23187. The applicability of this procedure has not been tested under various pathological conditions or in red ceils from other species. It is possible that a different correlation equation may exist for erythrocytes from other species, since the buffering mechanisms associated with free magnesium may be species dependent. Further studies are necessary to define the extent of applicability of our observations. References I Murphy, R.A,, Bohr, D.F. and Newman, D.L. (1969) Am. J. Physiol. 217. 666-673. 2 Meissner, G. and Henderson, J.S, (1987) J. Biol. Chem. 262, 3065-3073. 3 Skou, J. (1975) Q, Rev. Biophys. 7, 401-434. 4 Wacker, W,E.C, (1987) Magnesium 6, 61-64. 5 Altura, B,M, and AItura, B.T. (1981) Fed. Proc. Exp, Am. Soc. Exp, Biol. 40, 2672-2679. 6 Corkey, B.E., Duszynski, J., Rich, T.L., Matschins~, B. and Williamson, J,R, (1986) J. Biol. Chem. 261, 2567-2574. 7 Turlapaty, P.D.M.V. and AItura, B.M. (1980) Science 208, 198200. 8 Matsuura, T., Kohno, M., Kanayama, Y., Yasunari, K., Murakawa, K., Takeda, T., Ishimori, K., Morshima, I. and Yonezawa, T. (1987) Biochem. Biophys. Res. Commun. 143, 1012-1017. 9 Resnick, L.M., Gupta, R.K. and Larage, J.H. (1984) Proc. Natl. Acad. Sci. USA 81, 6511-6515.
10 Anderson, T.W., Neri, L.C., Schreiber, G., Talbot, F.D.F. and Zdrejewski, A. (1975) Can. Med. Assoc. J. 113, 199-203. 11 Fujise, H. and Lauf, P.K. (1988) .l. Physiol. 405, 605-614. 12 Flatman, P.W. and Lew, V.L. (1981) J. Physiol. 315, 421-446. 13 Lauf, P.K. (1988) Am. J. Physiol. 255 (Cell Physiol. 24), C331C339. 14 Rose, I.A. (1968) Proc. Natl. Acad. Sci. USA 61, 1079-1086. 15 Flatman, P.W. and Lew, V.L. (1980) J. Physiol. 305, 13-30. !6 Gupta, R.K., Benovic, J.L. and Rose, Z.B. (1978) J. Biol. Chem. 253, 6172-6176. 17 Bock, J.L., Wenz, B. and Gupta, R.K. (1985) Blood 65, 1526-1530. 18 Gard, J.K. and Ackerman, J.J.H. (1983) J. Magn. Reson. 51, 125-127. 19 Gupta, R.K. and Benovic, J.L. (1978) Biochem. Biophys. Res. Commun. 84, 130-137. 20 Misawa, K., Lee, T.M. and Ogawa, S. (1982) Biochim. Biophys. Acta 718, 227-229. 21 Gupta, R.K., Benovic, J.L. and Rose, Z.B. (1978) J. Biol. Chem. 253, 6165-6171. 22 Murphy, E., Levy, L., Berkowitz, L.R., Orringer, E.P., Gabel, S.A. and London, R.E. (1986) Am. J. Physiol. 251, C496-C504. 23 Gupta, R.K. and Moore, R.D. (1980) J. Biol. Chem. 255, 39873993. 24 Wa, S.T., Pieper, G.M., Salhany, J.M. and Eliot, R,S. (1981) Biochemistry 20, 7399-7403. 25 Garfinkel, L. and Garfinkel, D. (1984) Biochemistry 23, 35473552. 26 Gupta, R.K. and Gupta, P. (1987) NMR Spectroscopy of Cells and Organisms (Gupta, R.K., ed.), Vol. 2, Chap. 8, p. 39. CRC Press, Boca Raton, FL. 27 Delpire, E. and Lauf, P.K. (1991) J. Physiol. (Lond.), in press.
