AJH
1992;
5:733-739
NMR Measurement of Intracellular Potassium in the Perfused Normotensive and Spontaneously Hypertensive Rat Aorta by a Multinuclear Subtraction Procedure Joseph C Veniero and Raj K. Gupta
Intracellular free potassium ion concentrations have been measured in the isolated perfused aortae from Wistar-Kyoto and spontaneously hypertensive rats by a novel multinuclear NMR procedure, which allows the estimation and subtraction of the extracellular potassium signal from the total K resonance. When tested on a sample of erythro cytes, the method yielded an intracellular potas sium concentration of 149 mmol/L cell water, which is similar to the previously reported average of 140 mmol/L cell water, indicating full (100%) NMR "visibility" of red cell K . In contrast, the K resonance of the perfused rat aorta preparation indicated very low (—15%) NMR visibility of intra cellular potassium in the aortic tissue, similar to 3 9
3 9
3 9
+
+
+
T
that previously reported for perfused hearts. More importantly, aortae from spontaneously hyperten sive rats yielded a corrected (for low NMR visibil ity) intracellular potassium concentration of only 90 ± 10 (mean ± SE; η = 6) mmol/L cell water, which was significantly lower (40%; Ρ = .02) than that in aortae from normotensive rats. The NMR data thus reveal a significant depletion of intracel lular potassium in hypertensive arterial smooth muscle cells. Am J Hypertens 1992;5:733-739
here is a large body of evidence demonstrating the existence, in essential hypertension, of an abnormality in the plasma membrane of many cell types, including erythrocytes, " vascular smooth muscle cells, and kidney cells. The normal 1
2
6
5
7
KEY WORDS: Nuclear magnetic resonance, intracel lular potassium, vascular smooth muscle, spontane ously hypertensive rat, aorta.
physiology and regulation of various cations appear al tered in hypertension. Previous work has demon strated significant increases in the intracellular concen trations of free calcium and sodium ions, and suggested a decrease in the intracellular free magnesium ion con centration in the spontaneously hypertensive rat aorta. However, little is known about the derangement of in tracellular potassium in hypertension, despite increas ing evidence for the involvement of potassium ions in the pathophysiology of essential hypertension. The significance of dietary potassium and its potential use as a therapeutic agent is discussed in the early litera ture. ' More recently, total body potassium was found to be significantly decreased in patients with essential hypertension. Limiting dietary potassium raises the blood pressure of patients with essential hypertension; 1
6
Received February 21, 1992. Accepted July 2, 1992. From the Departments of Physiology and Biophysics (JCV, RKG), and Biochemistry (RKG), Albert Einstein College of Medicine, Bronx, New York. This research was supported by Public Health Service National Institutes of Health research grant DK32030. The Einstein NMR Re search Facility is supported in part by the National Cancer Institute Core Grant CA13330. Address correspondence and reprint requests to Dr. Raj K. Gupta, Albert Einstein College of Medicine, 1300 Morris Park Avenue, De partment of Physiology and Biophysics, Bronx, NY 10461.
© 1992 by the American Journal of Hypertension, Inc.
8
9
10
8
11
0895-7061/'92/'$5.00
734
VENIERO AND GUPTA
AJH-OCTOBER
and in an analysis of published clinical trials, Cappuccio and MacGregor have shown that potassium supple ments result in a significant reduction in both systolic and diastolic blood pressure. This effect is more pro nounced in hypertensive patients than in normotensive subjects and it is increased with a longer duration of supplementation. However, with extended potassium supplementation, adrenal secretion of aldosterone should increase, so that potassium induced natriuresis is inhibited and sodium losses are reduced. Thus, further decreases in blood pressure are inhibited by the normal homeostatic response to increased potassium intake. In spontaneously hypertensive rats, dietery supplemen tation with a 1 % (w/v) potassium chloride solution sig nificantly diminished the rise in blood pressure that normally occurs in this strain with age. Also, the accel eration of the rise in blood pressure precipitated by a sodium load in these rats is attenuated by a high oral potassium intake. In addition, Ganguli and Tobian have shown that a "Nad-resistant" strain of SHR on a high K diet loses its NaCl-resistance and has a higher mortality rate when placed on a low potassium diet. Studies have also shown that high potassium diets re duce arteriolar hypertensive hypertrophy, protect against strokes, and reduce the extent of kidney disease in hypertensive r a t s . The measurement of the intracellular free potassium ion concentration ([K ]i) by NMR has been less popular than that of N a , C a , and M g ions because of its very low sensitivity and the large deviation of its reso nance frequency from those of commonly measured nuclei. Instead, investigators have often used C s or R b NMR to measure potassium fluxes because these ions are reported to follow potassium changes in v i t r o . " These ions, however, must be preloaded into the cell system being investigated and, therefore, are inappropriate for studies in which a measurement of the steady state [K ]i is desired. While the direct NMR mea surements of [Κ \ are currently performed using anionic paramagnetic shift reagents that usually are able to ade quately separate intracellular and extracellular K reso nances, these reagents may be toxic and therefore inap propriate for some systems. Attempts have also been made to directly observe [K ]i using double quantumfiltered K NMR. However, the sensitivity of double quantum-filtered K NMR is an order of magnitude lower than that of single quantum K N M R , which limits its applicability. To circumvent the above deficiencies, we have devel oped a multinuclear NMR procedure to measure [K ] by directly and noninvasively estimating and subtracting the contribution of the extracellular ions. In this paper, we describe this subtraction method and report compar ative measurements of [K ]i in the aortae of normoten sive and spontaneously hypertensive rats using this method. 12
13
14
15
+
16
17/18
+
+
2 +
2 +
1 3 3
87
19
22
1992-VOL
5, NO. 10
MATERIALS AND M E T H O D S For each run, the aortae from either two male WistarKyoto (WKY) rats weighing 200 to 400 g or two age matched male spontaneously hypertensive (SHR) rats of similar weight were used. Aortae were perfused as previously described. Briefly, after an injection of hepa rin (1500 U/kg, intraperitoneally) to reduce clotting, the aorta was removed under pentobarbital anesthesia (100 mg/kg, intraperitoneally), trimmed of visible fat and other extravascular tissue; two aortae were simulta neously perfused via a Y-connector. The aortae were tied off at the bottom so that the perfusate had to exit through the intercostal arteries. The preparations were suspended in moist air, rather than submerged in perfu sion medium. The perfusion medium was a Kreb's buffer consisting of (mmol/L) NaCl, 120; KCl, 4.8; M g S 0 , 1 . 2 ; C a C l , 1 . 3 ; N a H C 0 , 2 4 ; glucose, 10; pyru vate, 10, and 1 % H 0 , maintained at 37°C, and bub bled with a 9 5 % 0 / 5 % C 0 gas mixture. The aprtae were placed in a 10 mm (outside diameter) NMR tube and the perfusate flow was set at 6 mL/min. A diagram of the perfusion setup is shown in Figure 1. All NMR spectra were run on a Varian VXR-500 spectrometer equipped with a low band probe which allowed the recording of K and C1. Using an inductor stick in parallel with the tuned circuit made it possible to record the H resonance on the same probe's observe coil. This permitted all measurements to be done without a probe change and assured that all measurements were made on the same section of the aorta. For each K spectrum 30,000 free induction decays (FID) were accumulated at the resonance frequency of 23.3 MHz using 90° pulses, a receiver deadtime of 30 //sec and a recycle time of 0.205 sec with a spectral width of 10,000 Hz. All C1 spectra were acquired using 90° pulses at a frequency of 49.0 MHz. A recycle time of 0.245 sec and a spectral 6
4
2
3
2
2
2
39
2
35
2
39
35
+
+
39
23
+
39
39
39
24,25
+
{
+
FIGURE 1. Diagram of the aorta perfusion apparatus.
AJH-OCTOBER
1992-VOL
NMR MEASUREMENT OF POTASSIUM IN HYPERTENSIVE AORTA
5, NO. 10
width of 5200 Hz were used to accumulate 1000 FID. The N a spectra were recorded using 90° pulses with a spectral.width of 10,000 Hz, and a recycle time of 0.205 sec; a total of 100 FID were recorded for each spectrum. The H spectra were recorded at 76.7 MHz using 90° pulses with a spectral width of 5000 Hz, an acquisition time of 0.5 sec, and a recycle time of 1.0 sec; for each spectrum 100 FID were accumulated. In order to estimate [K \ from the observed K reso nance, it is necessary to subtract the contribution of extracellular K . The extracellular contribution to the K signal of the perfused aorta preparation can be esti mated if the fractional extracellular volume of the prepa ration is known. This fractional volume was calculated from the ratio of the (extracellular) C1 resonance of the aorta to the C1 resonance of a perfusate sample which fills the sensitive volume of the coil. The ratio of the extracellular to perfusate C1 resonance represents the extracellular water as a fraction of the active volume. The intracellular C1~ concentration in tissues is too small and the C1 resonance too broad to be detectable under our experimental conditions. This assumption was checked by independently determining extracellu lar N a space using N a NMR, and comparing the N a and C1 measurements. Since [Na \ is small and the intracellular volume is small compared to the extra cellular volume in the perfused aorta preparation, the intracellular N a represents less than 1 % of the tissue N a resonance in this preparation, and thus the ob served N a resonance gave a good estimate of extracel lular space. The fractional extracellular spaces calcu lated using N a and C1 NMR were essentially identical, and directly allowed calculation of the extra cellular K signal from the measured perfusate medium K signal. Figure 2 shows the extracellular K contribu tion (upper trace) to the aorta K signal. This was calcu lated by multiplying the perfusate K signal by a factor which represents the extracellular chloride or sodium space as a fraction of the total active volume. This extra cellular contribution was subtracted from the K reso nance of the perfused aorta preparation (lower trace) to obtain the true intracellular K signal.
735
39
23
2
3 9
+
+
39
3 9
391^
35
35
35
35
35
2 3
+
23
23
35
+
6
+
23
FIGURE 2. The K resonance of an isolated perfused rat aorta (lower trace, labeled K + K ) along with the extracellular contribution (upper trace, labeled K ). The extracellular signal was derived by multiplying the resonance of the perfusate sample with the factor representing the extracellular volume as a fraction of the sensitive volume, as described in the text. 39
39
39
in
0Ut
39
0Ut
23
23
35
+
39
+
resonances in these spectra can then be used to calculate [ K ] i using the following equation, which can be derived in a straightforward manner. +
3 9
39
3 9
A
K
- ( a 5 x § ) A Αο'Χ^Γ C1
Aput
3 9
In order to calculate [K \ on a cell water basis, the ratio of the H resonances of the aorta preparation and the perfusate, which represents the total water in the aorta preparation as a fraction of the sensitive volume of the coil, was measured. Figure 3 shows representative spec tra of various nuclei; the spectra in panel A show K signals of the aorta preparation (left trace) and perfusate medium (right trace), panel Β contains spectra showing N a signals of the perfused aorta preparation (left trace) and perfusion medium (right trace), the spectra in panel C show H signals of the aorta preparation (left trace) and the perfusate medium (right trace), and the spectra in panel D show extracellular (left trace) and perfusate C1 resonances (right trace). The integrated areas of the
X
+
2
3 9
23
2
35
A ?
A 1
D
*aorta Ao
^ Γ
χ
[
κ
+
] ο
(i)
AqCI
where A denotes the area of a resonance peak, the su perscript denotes the nucleus, the subscript identifies the sample ("aorta" refers to the perfused aorta prepara tion; " o " refers to measurements when the NMR tube is completely filled with perfusate with no aortae present; " o u t " refers to the extracellular volume of the aorta preparation); and [ K ] is the [K ] in the perfusate. As discussed above, the C1 resonance (A^ ) measures ex tracellular volume because the intracellular C1~ is un detectable. The basis for the calculation of the concen tration is essentially the same as that presented by +
35
0
+
t
35
736
VENIERO AND GUPTA
i 9
K + i n
3 9
AJH-OCTOBER
Na + Na i n
2 3
0 1
U
ι η
5, NO. 10
K 2 3
Ή
1992-VOL
J5
+\
,5
J
Κ.
ci
ol
J
ci
0
v.
FIGURE 3. Panel A shows K NMR spectra of an isolated perfused rat aorta preparation (left trace) and a perfusate sample run under the same conditions (right trace). Panel Β contains Na spectra of the same aorta preparation (left trace) and the perfusate sample (right trace). Panel C contains H spectra of the aorta preparation (left trace) and the perfusate sample (right trace). Panel D contains Cl spectra of the aorta preparation (left trace) and the perfusate sample (right trace). 39
23
2
35
Gupta and Gupta for the determination of [Na \ in cell suspensions, and by Jelicks and Gupta in an isolated perfused organ. 26
+
26
6
yielded a value of 21.4 ± 2.7 mmol/L cell water (n = 7) for the normotensive aorta (from rats with mean systolic blood pressure of 137 ± 6 mm Hg). Assuming that the K resonance represents only the central magnetic tran sition (V2 —> — Vi) for the K nucleus (I = % ) , which would lead to 4 0 % visibility for the aortic K , our mea surements would represent an intracellular free potas sium ion concentration of about 54 mmol/L cell water in the aortic cells from normal rats. These values are still far below those expected for electrochemical equilibrium, and suggest that additional mechanisms, such as the presence of quadrupole splitting, must be operative to reduce the fractional NMR visibility of intracellular K in the aortic tissue. The presence of nonvanishing quadrupolar splitting requires the existence of long range order in electric field gradients, which could occur in the interior of muscle cells because of high concentrations (—50 mEq/L) of polyanions. In the presence of a large quadrupolar splitting, even moderately soft radio fre quency pulses will only excite the central (V2 —* — Vi) transition, and, under these conditions, the K nuclear spin behaves like a "fictitious" I = Vi nucleus. However, this fictitious spin responds to radio frequency excita tion with an effective gyromagnetic ratio that is larger 39
39
+
RESULTS The multinuclear subtraction method was tested on a sample of freshly drawn and washed erythrocytes that were lightly packed. Integrating the resulting K , C1, and H spectra and applying Equation 1 yielded a [K ] of 149 mmol/L cell water. This value is very close to the literature value of approximately 140 mmol/L cell w a t e r and demonstrates that in a system in which intracellular potassium is fully NMR v i s i b l e " this method can give an accurate value for [K ] . Figure 4 shows the K resonances of isolated per fused hypertensive (left trace) and normotensive rat aortae (right trace). The K resonance of the hyperten sive aorta preparation is significantly smaller than that of the normotensive aorta preparation. The ratio of ex tracellular and intracellular water volumes of the per fused aorta preparations are given in Table 1. Un corrected [K ]| measurements using the above method 39
35
2
+
27
28
+
3 9
39
+
{
30
{
31
+
39
AJH-OCTOBER
1992-VOL.
5, NO. 10
N M R M E A S U R E M E N T OF POTASSIUM IN HYPERTENSIVE A O R T A
737
WKY K + K, 39
ln
FIGURE 4. Comparison of the K resonances of the isolated perfused rat aorta preparation from spontaneously hypertensive (left trace, labeled SHR) and normotensive (right trace, labeled WKY) rats. 39
than that of the K nucleus as a whole. Thus, the NMR signal resulting from the use of radio frequency excitation pulses that were optimized on standard aqueous solutions may be significantly smaller than expected from the K nucleus. This mechanism may explain the very low NMR visibility of intracellular K observed here. Low visibility of intracellular K , similar to that seen here in perfused aorta, has previously been reported for perfused h e a r t s . Assuming the cytosolic [K ] in the normal aorta to be the same as in the red blood cell (149 mmol/L cell water), the fractional NMR visibility of the aortic K resonance would be about 15%. In the aorta from spontaneously hypertensive rat (mean systolic blood pressure of 170 ± 9 mm Hg), the uncorrected [K \ according to Equation 1 was only 12.8 ± 1.4 mmol/L cell water (n = 6) (Table 1). These data demonstrate significant depletion of intracellular free potassium (P = .02) in the hypertensive tissue. The 39
39
+
+
32,33
+
3 9
+
+
invisibility factor, which is based on the subtle aspects of the physics of NMR of quadrupolar nuclei discussed above, would, however, be expected to be the same for normotensive and hypertensive aortae. This would yield, for intracellular [K ] in the hypertensive aorta, a corrected (for low NMR visibility) value of 90 ± 10 mmol/L cell water (Table 1). Irrespective of the mechanism of the NMR invisibility of K , our results clearly indicate significant depletion of intracellular potassium in a vascular smooth muscle tissue in hypertension; and this depletion can be measured noninvasively by the multinuclear subtraction procedure developed here. +
+
DISCUSSION The decrease in the intracellular potassium ion concentration in hypertension is consistent with the previously reported increases in the intracellular free sodium and free calcium ion concentrations in this hypertensive tissue. One possible mechanism for the decreased intra6
TABLE 1. INTRACELLULAR POTASSIUM IN NORMOTENSIVE AND HYPERTENSIVE RAT AORTAE
Blood pressure (mm Hg) Intracellular [K ] (mmol/L cell water) Extracellular/intracellular water +
All values presented are mean ± SEM (n = 6 for SHR; n-7
for WKY).
* Significantly different from WKY control (P < .02; unpaired t test). | Not significantly different from WKY control.
SHR
WKY
170 ± 9 * 90 ± 10* 3.73 ± 0.43f
137 ± 6 149 ± 19 4.42 ± 0.76
738
AJH-OCTOBER
VENIERO AND GUPTA
cellular potassium concentration is a decreased N a - K ATPase activity in essential hypertension, although there is recent evidence to the contrary. Another expla nation would be an increased loss of intracellular potas sium ions via calcium-activated potassium channels due to the higher concentration of intracellular free C a in the hypertensive tissue. An increased passive leak of potassium ions in hypertensive vascular smooth muscle could also account for the observed decrease in [K+Jj. Since the concentration of intracellular K is thought to reflect the magnitude of the membrane po tential, our data would indicate significant depolariza tion of the plasma membrane in hypertension. This could occur as a consequence of either decreased activity of the N a / K p u m p or an increase in N a permeability. Both explanations are consistent with the increase in the intracellular sodium concentration previously reported by Jelicks and Gupta. Syme et al have reported a threefold increase over that of normotensive controls in the efflux rate of Rb , a substitute for K , in the skeletal muscle of SHRs in vivo. They propose that this may be in response to a reported increase in the activity of the N a - H antiporter in SHR. Since, in cellular volume regulation, activity of the N a - H antiporter increases intracellular volume and potassium release decreases cellular vol ume, an increase in potassium efflux would be a normal homeostatic response to the increase in cell volume caused by the increase in the activity of the N a - H antiporter. The opening of potassium channels has also been associated with a relaxation of vascular tone. Potas sium channel openers are being studied as pharmacolog ical vasodilators. It is possible that an increased efflux of intracellular potassium in hypertension may be an at tempt, by the body, to regulate blood pressure in the same fashion. The low [K \ in hypertension observed in this study may be a result of this attempt at blood pres sure regulation. +
in hypertension, diabetes, and obesity: a nuclear mag netic resonance spectroscopic study. Hypertension 1991;17:951-957.
+
34
35
2 +
6
6. Jelicks LA, Gupta RK: NMR measurement of cytosolic free calcium, free magnesium, and intracellular sodium in the aorta of the normal and spontaneously hyperten sive rat. J Biol Chem 1990;265:1394-1400. 7.
Dowd TL, Gupta RK: Multinuclear NMR studies of in tracellular cations in perfused hypertensive rat kidney. J Biol Chem 1992;267:3637-3643.
8.
Moore RD, Webb GD: The Κ Factor: Reversing and Pre venting High Blood Pressure Without Drugs. New York, MacMillan Publishing Company, 1986.
9.
Addison WLT, Clark HG: Calcium and potassium chlo rides in treatment of arterial hypertension. Can Med Assoc J 1925;15:913-915.
10.
Addison WLT: Use of sodium chloride, potassium chlo ride, sodium bromide, and potassium bromide in cases of arterial hypertension which are amenable to potassium chloride. Can Med Assoc J 1928;18:281-285.
11.
Krishna GG, Kapoor SC: Potassium depletion exacer bates essential hypertension. Ann Intern Med 1991; 115:77-83.
12.
Cappuccio FP, MacGregor GA: Does potassium supple mentation lower blood pressure? A meta-analysis of published trials. J Hypertens 1991;9:465-473.
13.
Langford HG: Sodium-potassium interaction in hyper tension and hypertensive cardiovascular disease. Hyper tension 1991;17(suppl I):I-155-I-157.
14.
Barden A, Beilin LJ, Vandongen R: Effect of potassium supplementation on blood pressure and vasodilator mechanisms in spontaneously hypertensive rats. Clin Sci 1988;75:527-534.
15.
Sato Y, Ando K, Ogata E, Fujita T: High-potassium diet attenuates salt-induced acceleration of hypertension. Am J Physiol 1991;260:R21-R26.
16.
Ganguli M, Tobian L: Dietary Κ determines NaCl sensi tivity in NaCl-induced rises of blood pressure in sponta neously hypertensive rats. Am J Hypertens 1990;3:482484.
17.
Tobian L: High potassium diets markedly protect against stroke deaths and kidney disease in hypertensive rats, a possible legacy from prehistoric times. Can J Physiol Pharmacol 1986;64:840-848.
18.
Tobian L, MacNeill D, Johnson MA, et al: Potassium protection against lesions of the renal tubules, arteries, and glomeruli and nephron loss in salt-loaded hyperten sive Dahl S rats. Hypertension 1984;6(suppl I):I1701176.
19.
Davis DG, Murphy E, London RE: Uptake of cesium ions by human erythrocytes and perfused rat heart: a cesium133 study. Biochem 1988;27:3547-3551.
20.
Williamson MP: Measurement of cromakalim-induced Rb flux in intact cells by NMR. FEBS Lett 1989; 254:171-173.
1
+
+
+
+
6
+
+
34
+
+
36
+
+
+
+
37
+
REFERENCES 1.
Postnov GO, Orlov SN: Ion transport across plasma membrane in primary hypertension. Physiol Rev 1985;65:904-945.
2.
Resnick LM, Gupta RK, Laragh JH: Intracellular free mag nesium in erythrocytes of essential hypertension: rela tion to blood pressure and serum divalent cations. Proc Natl Acad Sci USA 1984;81:6511-6515.
3.
4.
5.
Resnick LM, Gupta RK, Sosa RE, et al: Intracellular pH in human and experimental hypertension. Proc Natl Acad Sci USA 1987;84:7663-7667. Resnick LM: Cellular calcium and magnesium metabo lism in the pathophysiology and treatment of hyperten sion and related metabolic disorders. Am J Med (in press). Resnick LM, Gupta RK, Bhargava KK, et al: Cellular ions
5, NO. 10
1992-VOL
87
21.
Allis JL, Snaith CD, Seymour AM, Radda GK: Rb NMR studies of the perfused rat heart. FEBS Lett 1989; 242:215-217.
22.
Allis JL, Endre ZH, Radda GK: Rb, Na and P nuclear
87
87
23
31
AJH-OCTOBER
1992-VOL
5, NO. 10
739
NMR MEASUREMENT OF POTASSIUM IN HYPERTENSIVE A O R T A
magnetic resonance spectroscopy of the perfused rat kid ney. Renal Physiol Biochem 1989;12:171-180.
30.
23.
Endre ZH, Allis JL, Radda GK: Toxicity of dysprosium shift reagents in the isolated perfused rat kidney. Magn Reson Med 1989;11:267-274.
Boulanger Y, Vinay P: NMR visibility of K and C1 in erythrocytes and kidney tubules. Magn Reson Med 1990;16:246-251.
31.
24.
Seo Y, Murakami M, Suzuki E, et al: NMR characteristics of intracellular Κ in the rat salivary gland: a K NMR study using double-quantum filtering. Biochem 1990; 29:599-603.
Joseph PM, Summers, RM: The flip-angle effect: a method for detection of sodium-23 quadrupole splitting in tissue. Magn Reson Med 1987;4:67-77.
32.
Hiraishi T, Seo Y, Murakami M, Watari H: Detection of biexponential relaxation in intracellular Κ in the rat heart by double-quantum K NMR. J Magn Reson 1990;87: 169-173.
Fossel ET, Hoefeler H: Observation of intracellular potas sium and sodium in the heart by NMR: a major fraction of potassium is "invisible." Magn Reson Med 1986; 3:534-540.
33.
Burstein D, Fossel ET: Nuclear magnetic resonance stud ies of intracellular ions in perfused frog heart. Am J Phy siol 1987;252:H1138-H1146.
34.
Syme PD, Dixon RM, Aronson JK, et al: Evidence for increased in vivo sodium-potassium pump activity and potassium efflux in skeletal muscle of spontaneously hy pertensive rats. J Hypertens 1990;8:1161-1166.
35.
Orlov SN, Postnov IY, Pokudin NI, et al: Na -H ex change and other ion-transport systems in erythrocytes of essential hypertensives and spontaneously hyperten sive rats. J Hypertens 1989;7:781-788.
36.
Syme PD, Arnolda L, Green Y, et al: Evidence for in creased in vivo Na -H antiporter activity and an altered skeletal muscle contractile response in the spontane ously hypertensive rats. J Hypertens 1990;8:1027-1036.
39
25.
39
26.
Gupta RK, Gupta P: Direct observation of resolved reso nances from intra- and extracellular sodium-23 ions in NMR studies of intact cells and tissues using dysprosium(III)tripolyphosphate as paramagnetic shift reagent. J Magn Reson 1982;47:344-350.
27.
Hoffman JF: Active transport of Na and K by red blood cells, in Andreoli TE, Hoffman JF, Fanestil DD, Schultz SG, (eds): Physiology of Membrane Disorders, ed 2. New York, Plenum Publishing Corp., 1986, pp 221-234.
28.
Ogino T, Shulman GI, Avison MJ, et al: Na and K NMR studies of ion transport in human erythrocytes. Proc Natl Acad Sci USA 1985;82:1099-1103.
29.
Brophy PJ, Hayer MK, Riddell FG: Measurement of in tracellular potassium ion concentration by n.m.r. Bio chem J 1983;210:961-963.
+
+
23
39
39
+
+
37.
35
+
+
Cook NS: The pharmacology of potassium channels and their therapeutic potential. Trends Pharmacol Sci 1988;9:21-28.
1992;
5:733-739
NMR Measurement of Intracellular Potassium in the Perfused Normotensive and Spontaneously Hypertensive Rat Aorta by a Multinuclear Subtraction Procedure Joseph C Veniero and Raj K. Gupta
Intracellular free potassium ion concentrations have been measured in the isolated perfused aortae from Wistar-Kyoto and spontaneously hypertensive rats by a novel multinuclear NMR procedure, which allows the estimation and subtraction of the extracellular potassium signal from the total K resonance. When tested on a sample of erythro cytes, the method yielded an intracellular potas sium concentration of 149 mmol/L cell water, which is similar to the previously reported average of 140 mmol/L cell water, indicating full (100%) NMR "visibility" of red cell K . In contrast, the K resonance of the perfused rat aorta preparation indicated very low (—15%) NMR visibility of intra cellular potassium in the aortic tissue, similar to 3 9
3 9
3 9
+
+
+
T
that previously reported for perfused hearts. More importantly, aortae from spontaneously hyperten sive rats yielded a corrected (for low NMR visibil ity) intracellular potassium concentration of only 90 ± 10 (mean ± SE; η = 6) mmol/L cell water, which was significantly lower (40%; Ρ = .02) than that in aortae from normotensive rats. The NMR data thus reveal a significant depletion of intracel lular potassium in hypertensive arterial smooth muscle cells. Am J Hypertens 1992;5:733-739
here is a large body of evidence demonstrating the existence, in essential hypertension, of an abnormality in the plasma membrane of many cell types, including erythrocytes, " vascular smooth muscle cells, and kidney cells. The normal 1
2
6
5
7
KEY WORDS: Nuclear magnetic resonance, intracel lular potassium, vascular smooth muscle, spontane ously hypertensive rat, aorta.
physiology and regulation of various cations appear al tered in hypertension. Previous work has demon strated significant increases in the intracellular concen trations of free calcium and sodium ions, and suggested a decrease in the intracellular free magnesium ion con centration in the spontaneously hypertensive rat aorta. However, little is known about the derangement of in tracellular potassium in hypertension, despite increas ing evidence for the involvement of potassium ions in the pathophysiology of essential hypertension. The significance of dietary potassium and its potential use as a therapeutic agent is discussed in the early litera ture. ' More recently, total body potassium was found to be significantly decreased in patients with essential hypertension. Limiting dietary potassium raises the blood pressure of patients with essential hypertension; 1
6
Received February 21, 1992. Accepted July 2, 1992. From the Departments of Physiology and Biophysics (JCV, RKG), and Biochemistry (RKG), Albert Einstein College of Medicine, Bronx, New York. This research was supported by Public Health Service National Institutes of Health research grant DK32030. The Einstein NMR Re search Facility is supported in part by the National Cancer Institute Core Grant CA13330. Address correspondence and reprint requests to Dr. Raj K. Gupta, Albert Einstein College of Medicine, 1300 Morris Park Avenue, De partment of Physiology and Biophysics, Bronx, NY 10461.
© 1992 by the American Journal of Hypertension, Inc.
8
9
10
8
11
0895-7061/'92/'$5.00
734
VENIERO AND GUPTA
AJH-OCTOBER
and in an analysis of published clinical trials, Cappuccio and MacGregor have shown that potassium supple ments result in a significant reduction in both systolic and diastolic blood pressure. This effect is more pro nounced in hypertensive patients than in normotensive subjects and it is increased with a longer duration of supplementation. However, with extended potassium supplementation, adrenal secretion of aldosterone should increase, so that potassium induced natriuresis is inhibited and sodium losses are reduced. Thus, further decreases in blood pressure are inhibited by the normal homeostatic response to increased potassium intake. In spontaneously hypertensive rats, dietery supplemen tation with a 1 % (w/v) potassium chloride solution sig nificantly diminished the rise in blood pressure that normally occurs in this strain with age. Also, the accel eration of the rise in blood pressure precipitated by a sodium load in these rats is attenuated by a high oral potassium intake. In addition, Ganguli and Tobian have shown that a "Nad-resistant" strain of SHR on a high K diet loses its NaCl-resistance and has a higher mortality rate when placed on a low potassium diet. Studies have also shown that high potassium diets re duce arteriolar hypertensive hypertrophy, protect against strokes, and reduce the extent of kidney disease in hypertensive r a t s . The measurement of the intracellular free potassium ion concentration ([K ]i) by NMR has been less popular than that of N a , C a , and M g ions because of its very low sensitivity and the large deviation of its reso nance frequency from those of commonly measured nuclei. Instead, investigators have often used C s or R b NMR to measure potassium fluxes because these ions are reported to follow potassium changes in v i t r o . " These ions, however, must be preloaded into the cell system being investigated and, therefore, are inappropriate for studies in which a measurement of the steady state [K ]i is desired. While the direct NMR mea surements of [Κ \ are currently performed using anionic paramagnetic shift reagents that usually are able to ade quately separate intracellular and extracellular K reso nances, these reagents may be toxic and therefore inap propriate for some systems. Attempts have also been made to directly observe [K ]i using double quantumfiltered K NMR. However, the sensitivity of double quantum-filtered K NMR is an order of magnitude lower than that of single quantum K N M R , which limits its applicability. To circumvent the above deficiencies, we have devel oped a multinuclear NMR procedure to measure [K ] by directly and noninvasively estimating and subtracting the contribution of the extracellular ions. In this paper, we describe this subtraction method and report compar ative measurements of [K ]i in the aortae of normoten sive and spontaneously hypertensive rats using this method. 12
13
14
15
+
16
17/18
+
+
2 +
2 +
1 3 3
87
19
22
1992-VOL
5, NO. 10
MATERIALS AND M E T H O D S For each run, the aortae from either two male WistarKyoto (WKY) rats weighing 200 to 400 g or two age matched male spontaneously hypertensive (SHR) rats of similar weight were used. Aortae were perfused as previously described. Briefly, after an injection of hepa rin (1500 U/kg, intraperitoneally) to reduce clotting, the aorta was removed under pentobarbital anesthesia (100 mg/kg, intraperitoneally), trimmed of visible fat and other extravascular tissue; two aortae were simulta neously perfused via a Y-connector. The aortae were tied off at the bottom so that the perfusate had to exit through the intercostal arteries. The preparations were suspended in moist air, rather than submerged in perfu sion medium. The perfusion medium was a Kreb's buffer consisting of (mmol/L) NaCl, 120; KCl, 4.8; M g S 0 , 1 . 2 ; C a C l , 1 . 3 ; N a H C 0 , 2 4 ; glucose, 10; pyru vate, 10, and 1 % H 0 , maintained at 37°C, and bub bled with a 9 5 % 0 / 5 % C 0 gas mixture. The aprtae were placed in a 10 mm (outside diameter) NMR tube and the perfusate flow was set at 6 mL/min. A diagram of the perfusion setup is shown in Figure 1. All NMR spectra were run on a Varian VXR-500 spectrometer equipped with a low band probe which allowed the recording of K and C1. Using an inductor stick in parallel with the tuned circuit made it possible to record the H resonance on the same probe's observe coil. This permitted all measurements to be done without a probe change and assured that all measurements were made on the same section of the aorta. For each K spectrum 30,000 free induction decays (FID) were accumulated at the resonance frequency of 23.3 MHz using 90° pulses, a receiver deadtime of 30 //sec and a recycle time of 0.205 sec with a spectral width of 10,000 Hz. All C1 spectra were acquired using 90° pulses at a frequency of 49.0 MHz. A recycle time of 0.245 sec and a spectral 6
4
2
3
2
2
2
39
2
35
2
39
35
+
+
39
23
+
39
39
39
24,25
+
{
+
FIGURE 1. Diagram of the aorta perfusion apparatus.
AJH-OCTOBER
1992-VOL
NMR MEASUREMENT OF POTASSIUM IN HYPERTENSIVE AORTA
5, NO. 10
width of 5200 Hz were used to accumulate 1000 FID. The N a spectra were recorded using 90° pulses with a spectral.width of 10,000 Hz, and a recycle time of 0.205 sec; a total of 100 FID were recorded for each spectrum. The H spectra were recorded at 76.7 MHz using 90° pulses with a spectral width of 5000 Hz, an acquisition time of 0.5 sec, and a recycle time of 1.0 sec; for each spectrum 100 FID were accumulated. In order to estimate [K \ from the observed K reso nance, it is necessary to subtract the contribution of extracellular K . The extracellular contribution to the K signal of the perfused aorta preparation can be esti mated if the fractional extracellular volume of the prepa ration is known. This fractional volume was calculated from the ratio of the (extracellular) C1 resonance of the aorta to the C1 resonance of a perfusate sample which fills the sensitive volume of the coil. The ratio of the extracellular to perfusate C1 resonance represents the extracellular water as a fraction of the active volume. The intracellular C1~ concentration in tissues is too small and the C1 resonance too broad to be detectable under our experimental conditions. This assumption was checked by independently determining extracellu lar N a space using N a NMR, and comparing the N a and C1 measurements. Since [Na \ is small and the intracellular volume is small compared to the extra cellular volume in the perfused aorta preparation, the intracellular N a represents less than 1 % of the tissue N a resonance in this preparation, and thus the ob served N a resonance gave a good estimate of extracel lular space. The fractional extracellular spaces calcu lated using N a and C1 NMR were essentially identical, and directly allowed calculation of the extra cellular K signal from the measured perfusate medium K signal. Figure 2 shows the extracellular K contribu tion (upper trace) to the aorta K signal. This was calcu lated by multiplying the perfusate K signal by a factor which represents the extracellular chloride or sodium space as a fraction of the total active volume. This extra cellular contribution was subtracted from the K reso nance of the perfused aorta preparation (lower trace) to obtain the true intracellular K signal.
735
39
23
2
3 9
+
+
39
3 9
391^
35
35
35
35
35
2 3
+
23
23
35
+
6
+
23
FIGURE 2. The K resonance of an isolated perfused rat aorta (lower trace, labeled K + K ) along with the extracellular contribution (upper trace, labeled K ). The extracellular signal was derived by multiplying the resonance of the perfusate sample with the factor representing the extracellular volume as a fraction of the sensitive volume, as described in the text. 39
39
39
in
0Ut
39
0Ut
23
23
35
+
39
+
resonances in these spectra can then be used to calculate [ K ] i using the following equation, which can be derived in a straightforward manner. +
3 9
39
3 9
A
K
- ( a 5 x § ) A Αο'Χ^Γ C1
Aput
3 9
In order to calculate [K \ on a cell water basis, the ratio of the H resonances of the aorta preparation and the perfusate, which represents the total water in the aorta preparation as a fraction of the sensitive volume of the coil, was measured. Figure 3 shows representative spec tra of various nuclei; the spectra in panel A show K signals of the aorta preparation (left trace) and perfusate medium (right trace), panel Β contains spectra showing N a signals of the perfused aorta preparation (left trace) and perfusion medium (right trace), the spectra in panel C show H signals of the aorta preparation (left trace) and the perfusate medium (right trace), and the spectra in panel D show extracellular (left trace) and perfusate C1 resonances (right trace). The integrated areas of the
X
+
2
3 9
23
2
35
A ?
A 1
D
*aorta Ao
^ Γ
χ
[
κ
+
] ο
(i)
AqCI
where A denotes the area of a resonance peak, the su perscript denotes the nucleus, the subscript identifies the sample ("aorta" refers to the perfused aorta prepara tion; " o " refers to measurements when the NMR tube is completely filled with perfusate with no aortae present; " o u t " refers to the extracellular volume of the aorta preparation); and [ K ] is the [K ] in the perfusate. As discussed above, the C1 resonance (A^ ) measures ex tracellular volume because the intracellular C1~ is un detectable. The basis for the calculation of the concen tration is essentially the same as that presented by +
35
0
+
t
35
736
VENIERO AND GUPTA
i 9
K + i n
3 9
AJH-OCTOBER
Na + Na i n
2 3
0 1
U
ι η
5, NO. 10
K 2 3
Ή
1992-VOL
J5
+\
,5
J
Κ.
ci
ol
J
ci
0
v.
FIGURE 3. Panel A shows K NMR spectra of an isolated perfused rat aorta preparation (left trace) and a perfusate sample run under the same conditions (right trace). Panel Β contains Na spectra of the same aorta preparation (left trace) and the perfusate sample (right trace). Panel C contains H spectra of the aorta preparation (left trace) and the perfusate sample (right trace). Panel D contains Cl spectra of the aorta preparation (left trace) and the perfusate sample (right trace). 39
23
2
35
Gupta and Gupta for the determination of [Na \ in cell suspensions, and by Jelicks and Gupta in an isolated perfused organ. 26
+
26
6
yielded a value of 21.4 ± 2.7 mmol/L cell water (n = 7) for the normotensive aorta (from rats with mean systolic blood pressure of 137 ± 6 mm Hg). Assuming that the K resonance represents only the central magnetic tran sition (V2 —> — Vi) for the K nucleus (I = % ) , which would lead to 4 0 % visibility for the aortic K , our mea surements would represent an intracellular free potas sium ion concentration of about 54 mmol/L cell water in the aortic cells from normal rats. These values are still far below those expected for electrochemical equilibrium, and suggest that additional mechanisms, such as the presence of quadrupole splitting, must be operative to reduce the fractional NMR visibility of intracellular K in the aortic tissue. The presence of nonvanishing quadrupolar splitting requires the existence of long range order in electric field gradients, which could occur in the interior of muscle cells because of high concentrations (—50 mEq/L) of polyanions. In the presence of a large quadrupolar splitting, even moderately soft radio fre quency pulses will only excite the central (V2 —* — Vi) transition, and, under these conditions, the K nuclear spin behaves like a "fictitious" I = Vi nucleus. However, this fictitious spin responds to radio frequency excita tion with an effective gyromagnetic ratio that is larger 39
39
+
RESULTS The multinuclear subtraction method was tested on a sample of freshly drawn and washed erythrocytes that were lightly packed. Integrating the resulting K , C1, and H spectra and applying Equation 1 yielded a [K ] of 149 mmol/L cell water. This value is very close to the literature value of approximately 140 mmol/L cell w a t e r and demonstrates that in a system in which intracellular potassium is fully NMR v i s i b l e " this method can give an accurate value for [K ] . Figure 4 shows the K resonances of isolated per fused hypertensive (left trace) and normotensive rat aortae (right trace). The K resonance of the hyperten sive aorta preparation is significantly smaller than that of the normotensive aorta preparation. The ratio of ex tracellular and intracellular water volumes of the per fused aorta preparations are given in Table 1. Un corrected [K ]| measurements using the above method 39
35
2
+
27
28
+
3 9
39
+
{
30
{
31
+
39
AJH-OCTOBER
1992-VOL.
5, NO. 10
N M R M E A S U R E M E N T OF POTASSIUM IN HYPERTENSIVE A O R T A
737
WKY K + K, 39
ln
FIGURE 4. Comparison of the K resonances of the isolated perfused rat aorta preparation from spontaneously hypertensive (left trace, labeled SHR) and normotensive (right trace, labeled WKY) rats. 39
than that of the K nucleus as a whole. Thus, the NMR signal resulting from the use of radio frequency excitation pulses that were optimized on standard aqueous solutions may be significantly smaller than expected from the K nucleus. This mechanism may explain the very low NMR visibility of intracellular K observed here. Low visibility of intracellular K , similar to that seen here in perfused aorta, has previously been reported for perfused h e a r t s . Assuming the cytosolic [K ] in the normal aorta to be the same as in the red blood cell (149 mmol/L cell water), the fractional NMR visibility of the aortic K resonance would be about 15%. In the aorta from spontaneously hypertensive rat (mean systolic blood pressure of 170 ± 9 mm Hg), the uncorrected [K \ according to Equation 1 was only 12.8 ± 1.4 mmol/L cell water (n = 6) (Table 1). These data demonstrate significant depletion of intracellular free potassium (P = .02) in the hypertensive tissue. The 39
39
+
+
32,33
+
3 9
+
+
invisibility factor, which is based on the subtle aspects of the physics of NMR of quadrupolar nuclei discussed above, would, however, be expected to be the same for normotensive and hypertensive aortae. This would yield, for intracellular [K ] in the hypertensive aorta, a corrected (for low NMR visibility) value of 90 ± 10 mmol/L cell water (Table 1). Irrespective of the mechanism of the NMR invisibility of K , our results clearly indicate significant depletion of intracellular potassium in a vascular smooth muscle tissue in hypertension; and this depletion can be measured noninvasively by the multinuclear subtraction procedure developed here. +
+
DISCUSSION The decrease in the intracellular potassium ion concentration in hypertension is consistent with the previously reported increases in the intracellular free sodium and free calcium ion concentrations in this hypertensive tissue. One possible mechanism for the decreased intra6
TABLE 1. INTRACELLULAR POTASSIUM IN NORMOTENSIVE AND HYPERTENSIVE RAT AORTAE
Blood pressure (mm Hg) Intracellular [K ] (mmol/L cell water) Extracellular/intracellular water +
All values presented are mean ± SEM (n = 6 for SHR; n-7
for WKY).
* Significantly different from WKY control (P < .02; unpaired t test). | Not significantly different from WKY control.
SHR
WKY
170 ± 9 * 90 ± 10* 3.73 ± 0.43f
137 ± 6 149 ± 19 4.42 ± 0.76
738
AJH-OCTOBER
VENIERO AND GUPTA
cellular potassium concentration is a decreased N a - K ATPase activity in essential hypertension, although there is recent evidence to the contrary. Another expla nation would be an increased loss of intracellular potas sium ions via calcium-activated potassium channels due to the higher concentration of intracellular free C a in the hypertensive tissue. An increased passive leak of potassium ions in hypertensive vascular smooth muscle could also account for the observed decrease in [K+Jj. Since the concentration of intracellular K is thought to reflect the magnitude of the membrane po tential, our data would indicate significant depolariza tion of the plasma membrane in hypertension. This could occur as a consequence of either decreased activity of the N a / K p u m p or an increase in N a permeability. Both explanations are consistent with the increase in the intracellular sodium concentration previously reported by Jelicks and Gupta. Syme et al have reported a threefold increase over that of normotensive controls in the efflux rate of Rb , a substitute for K , in the skeletal muscle of SHRs in vivo. They propose that this may be in response to a reported increase in the activity of the N a - H antiporter in SHR. Since, in cellular volume regulation, activity of the N a - H antiporter increases intracellular volume and potassium release decreases cellular vol ume, an increase in potassium efflux would be a normal homeostatic response to the increase in cell volume caused by the increase in the activity of the N a - H antiporter. The opening of potassium channels has also been associated with a relaxation of vascular tone. Potas sium channel openers are being studied as pharmacolog ical vasodilators. It is possible that an increased efflux of intracellular potassium in hypertension may be an at tempt, by the body, to regulate blood pressure in the same fashion. The low [K \ in hypertension observed in this study may be a result of this attempt at blood pres sure regulation. +
in hypertension, diabetes, and obesity: a nuclear mag netic resonance spectroscopic study. Hypertension 1991;17:951-957.
+
34
35
2 +
6
6. Jelicks LA, Gupta RK: NMR measurement of cytosolic free calcium, free magnesium, and intracellular sodium in the aorta of the normal and spontaneously hyperten sive rat. J Biol Chem 1990;265:1394-1400. 7.
Dowd TL, Gupta RK: Multinuclear NMR studies of in tracellular cations in perfused hypertensive rat kidney. J Biol Chem 1992;267:3637-3643.
8.
Moore RD, Webb GD: The Κ Factor: Reversing and Pre venting High Blood Pressure Without Drugs. New York, MacMillan Publishing Company, 1986.
9.
Addison WLT, Clark HG: Calcium and potassium chlo rides in treatment of arterial hypertension. Can Med Assoc J 1925;15:913-915.
10.
Addison WLT: Use of sodium chloride, potassium chlo ride, sodium bromide, and potassium bromide in cases of arterial hypertension which are amenable to potassium chloride. Can Med Assoc J 1928;18:281-285.
11.
Krishna GG, Kapoor SC: Potassium depletion exacer bates essential hypertension. Ann Intern Med 1991; 115:77-83.
12.
Cappuccio FP, MacGregor GA: Does potassium supple mentation lower blood pressure? A meta-analysis of published trials. J Hypertens 1991;9:465-473.
13.
Langford HG: Sodium-potassium interaction in hyper tension and hypertensive cardiovascular disease. Hyper tension 1991;17(suppl I):I-155-I-157.
14.
Barden A, Beilin LJ, Vandongen R: Effect of potassium supplementation on blood pressure and vasodilator mechanisms in spontaneously hypertensive rats. Clin Sci 1988;75:527-534.
15.
Sato Y, Ando K, Ogata E, Fujita T: High-potassium diet attenuates salt-induced acceleration of hypertension. Am J Physiol 1991;260:R21-R26.
16.
Ganguli M, Tobian L: Dietary Κ determines NaCl sensi tivity in NaCl-induced rises of blood pressure in sponta neously hypertensive rats. Am J Hypertens 1990;3:482484.
17.
Tobian L: High potassium diets markedly protect against stroke deaths and kidney disease in hypertensive rats, a possible legacy from prehistoric times. Can J Physiol Pharmacol 1986;64:840-848.
18.
Tobian L, MacNeill D, Johnson MA, et al: Potassium protection against lesions of the renal tubules, arteries, and glomeruli and nephron loss in salt-loaded hyperten sive Dahl S rats. Hypertension 1984;6(suppl I):I1701176.
19.
Davis DG, Murphy E, London RE: Uptake of cesium ions by human erythrocytes and perfused rat heart: a cesium133 study. Biochem 1988;27:3547-3551.
20.
Williamson MP: Measurement of cromakalim-induced Rb flux in intact cells by NMR. FEBS Lett 1989; 254:171-173.
1
+
+
+
+
6
+
+
34
+
+
36
+
+
+
+
37
+
REFERENCES 1.
Postnov GO, Orlov SN: Ion transport across plasma membrane in primary hypertension. Physiol Rev 1985;65:904-945.
2.
Resnick LM, Gupta RK, Laragh JH: Intracellular free mag nesium in erythrocytes of essential hypertension: rela tion to blood pressure and serum divalent cations. Proc Natl Acad Sci USA 1984;81:6511-6515.
3.
4.
5.
Resnick LM, Gupta RK, Sosa RE, et al: Intracellular pH in human and experimental hypertension. Proc Natl Acad Sci USA 1987;84:7663-7667. Resnick LM: Cellular calcium and magnesium metabo lism in the pathophysiology and treatment of hyperten sion and related metabolic disorders. Am J Med (in press). Resnick LM, Gupta RK, Bhargava KK, et al: Cellular ions
5, NO. 10
1992-VOL
87
21.
Allis JL, Snaith CD, Seymour AM, Radda GK: Rb NMR studies of the perfused rat heart. FEBS Lett 1989; 242:215-217.
22.
Allis JL, Endre ZH, Radda GK: Rb, Na and P nuclear
87
87
23
31
AJH-OCTOBER
1992-VOL
5, NO. 10
739
NMR MEASUREMENT OF POTASSIUM IN HYPERTENSIVE A O R T A
magnetic resonance spectroscopy of the perfused rat kid ney. Renal Physiol Biochem 1989;12:171-180.
30.
23.
Endre ZH, Allis JL, Radda GK: Toxicity of dysprosium shift reagents in the isolated perfused rat kidney. Magn Reson Med 1989;11:267-274.
Boulanger Y, Vinay P: NMR visibility of K and C1 in erythrocytes and kidney tubules. Magn Reson Med 1990;16:246-251.
31.
24.
Seo Y, Murakami M, Suzuki E, et al: NMR characteristics of intracellular Κ in the rat salivary gland: a K NMR study using double-quantum filtering. Biochem 1990; 29:599-603.
Joseph PM, Summers, RM: The flip-angle effect: a method for detection of sodium-23 quadrupole splitting in tissue. Magn Reson Med 1987;4:67-77.
32.
Hiraishi T, Seo Y, Murakami M, Watari H: Detection of biexponential relaxation in intracellular Κ in the rat heart by double-quantum K NMR. J Magn Reson 1990;87: 169-173.
Fossel ET, Hoefeler H: Observation of intracellular potas sium and sodium in the heart by NMR: a major fraction of potassium is "invisible." Magn Reson Med 1986; 3:534-540.
33.
Burstein D, Fossel ET: Nuclear magnetic resonance stud ies of intracellular ions in perfused frog heart. Am J Phy siol 1987;252:H1138-H1146.
34.
Syme PD, Dixon RM, Aronson JK, et al: Evidence for increased in vivo sodium-potassium pump activity and potassium efflux in skeletal muscle of spontaneously hy pertensive rats. J Hypertens 1990;8:1161-1166.
35.
Orlov SN, Postnov IY, Pokudin NI, et al: Na -H ex change and other ion-transport systems in erythrocytes of essential hypertensives and spontaneously hyperten sive rats. J Hypertens 1989;7:781-788.
36.
Syme PD, Arnolda L, Green Y, et al: Evidence for in creased in vivo Na -H antiporter activity and an altered skeletal muscle contractile response in the spontane ously hypertensive rats. J Hypertens 1990;8:1027-1036.
39
25.
39
26.
Gupta RK, Gupta P: Direct observation of resolved reso nances from intra- and extracellular sodium-23 ions in NMR studies of intact cells and tissues using dysprosium(III)tripolyphosphate as paramagnetic shift reagent. J Magn Reson 1982;47:344-350.
27.
Hoffman JF: Active transport of Na and K by red blood cells, in Andreoli TE, Hoffman JF, Fanestil DD, Schultz SG, (eds): Physiology of Membrane Disorders, ed 2. New York, Plenum Publishing Corp., 1986, pp 221-234.
28.
Ogino T, Shulman GI, Avison MJ, et al: Na and K NMR studies of ion transport in human erythrocytes. Proc Natl Acad Sci USA 1985;82:1099-1103.
29.
Brophy PJ, Hayer MK, Riddell FG: Measurement of in tracellular potassium ion concentration by n.m.r. Bio chem J 1983;210:961-963.
+
+
23
39
39
+
+
37.
35
+
+
Cook NS: The pharmacology of potassium channels and their therapeutic potential. Trends Pharmacol Sci 1988;9:21-28.
