- Email: [email protected]
Alterations in Membrane Fatty Acid Unsaturation and Chain Length in Hypertension as Observed by 1H NMR Spectroscopy 1
AJH
1998;11:340 –348
Alterations in Membrane Fatty Acid Unsaturation and Chain Length in Hypertension as Observed by 1 H NMR Spectroscopy Yuling Chi and Raj K. Gupta
Alterations in fatty acids of membrane phospholipids in essential hypertension may account for altered membrane ion transport, elasticity, and contractility properties of hypertensive tissues. To investigate the abnormalities in membrane fatty acids in essential hypertension, the degree of fatty acid unsaturation ([–CHACH–]/[–CH3]), the average carbon chain length, ratio of glycerol to fatty acyl chains, ratio of phosphatidylcholine to fatty acyl chains, and the ratio of free and acylated cholesterol to fatty acyl chains in fatty acid fractions of membrane phospholipids of aorta, kidney, and heart were determined in spontaneously hypertensive rats (SHR) and Wistar-Kyoto (WKY) rats by 1H nuclear magnetic resonance (NMR) spectroscopy. The degrees of fatty acid unsaturation in the aorta and the kidney membranes were significantly lower in SHR than in WKY rats (aorta, 0.53 6 0.01 v 0.63 6 0.01, n 5 5, P 5 .01; kidney, 0.70 6 0.01 v 0.84 6 0.03, n 5 10, P 5 .01). No significant difference could be detected in fatty acid unsaturation in heart membranes between these two strains. For aorta, kidney, and heart membranes, the average
carbon chain lengths of fatty acid fractions of membrane phospholipids were significantly shorter for SHR than for WKY rats (aorta, 15.1 6 0.2 v 18.3 6 0.7, n 5 5, P 5 .02; kidney, 14.5 6 0.2 v 16.4 6 0.4, n 5 10, P 5 .01; heart, 17.3 6 0.5 v 18.8 6 0.6, n 5 10, P 5 .05). The lower unsaturated fatty acid content in membrane phospholipids of the aorta and the kidney, with concomitant reduction in average chain length, may arise from increased oxidation of fatty acid double bonds in hypertensive tissues and may account, in part, for the increased aortic stiffness and abnormal kidney function associated with essential hypertension. Whether the lower unsaturated fatty acid content and decreased carbon chain length of phospholipid membranes in the aorta and the kidney are a cause or a consequence of the high blood pressure, however, remains unknown. Am J Hypertens 1998;11:340 –348 © 1998 American Journal of Hypertension, Ltd.
Received July 7, 1997. Accepted September 29, 1997. From the Department of Physiology and Biophysics, Albert Einstein College of Medicine of Yeshiva University, Bronx, New York. This work was supported by National Institutes of Health Research Grant DK32030.
Address correspondence and reprint requests to Dr. Raj K. Gupta, Department of Physiology and Biophysics, Albert Einstein College of Medicine of Yeshiva University, Bronx, NY 10461; e-mail: [email protected]
© 1998 by the American Journal of Hypertension, Ltd. Published by Elsevier Science, Inc.
Hypertension, fatty acid, 1H nuclear magnetic resonance. KEY WORDS:
0895-7061/98/$19.00 PII S0895-7061(97)00456-1
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
H
ypertension has long been associated with alterations in cell membranes of various tissues. Fatty acid fractions are important constituents of phospholipids in cell membranes. Alterations in fatty acids affect the membrane properties and functions, such as phase transition temperature, elasticity, and ion permeability. Dietary supplementation with polyunsaturated fatty acids (PUFA), which can increase membrane elasticity, can lower blood pressure (BP) in both hypertensive and normotensive humans as well as in animal models.1– 4 We have therefore hypothesized that abnormalities in fatty acid contents of phospholipids in cell membranes of various tissues may be associated with hypertension. It is well accepted that an increase in arterial stiffness and peripheral resistance is the major mechanism responsible for hypertension.5 Meaney et al observed a strong positive correlation between systolic pressure and arterial stiffness.5 Patients who have isolated systolic hypertension have an increase in peripheral resistance.5 Peripheral resistance is in turn determined by arterial diameters, the smaller the arterial diameter, the higher the peripheral resistance. It has been found in hypertensives that, although the diastolic blood pressure was increased, the diameter of the aorta, essentially a large artery, was significantly decreased.6 Aortic stiffness is also significantly increased in hypertension.6 The inelasticity of smooth muscle membranes contributes to aortic stiffness, which has been shown to be significantly related to both blood pressure and the content of atherogenic lipids.7 Changes in fatty acids of phospholipids in membranes result in differences in membrane fluidity and permeability, and may lead to differences in aortic stiffness. Other major target organs for hypertension are the kidney and the heart. Kidney function and hypertension are well known to be intimately related. Hypertension can be both cause and consequence of renal injury. An abnormality in the kidney is often associated with the development of hypertension. Hypertension induced cardiac hypertrophy and other cardiomyopathies have also been described in the literature. We have hypothesized that alterations in membrane fatty acids may occur in these organs as well. In view of the importance of kidney, heart, and vasculature in the development and maintenance of hypertension, and the important role of membranes in their biologic actions, we felt it important to investigate fatty acids in the membranes of hypertensive aorta, kidney, and heart. Proton nuclear magnetic resonance (NMR) spectroscopy was used to quantitate fatty acid unsaturation ratio, average carbon chain length, ratio of glycerol to fatty acyl chains, ratio of phosphatidylcholine to fatty acyl
MEMBRANE FATTY ACIDS IN HYPERTENSION
341
chains, and the ratio of free and acylated cholesterol to fatty acyl chains in fatty acid fractions of the aorta, kidney, and heart membranes from SHR and WKY rats. 1H NMR offers a rapid and comprehensive technique for monitoring lipid composition of cells, body fluids, and tissues. One of us, along with others, recently used it to demonstrate that extracellular Mg21 modulates membrane lipids in vascular smooth muscle.8 The same technique is now being applied to investigate alterations in membrane fatty acids in hypertension. MATERIALS AND METHODS Materials Methanol (anhydrous) (MeOH) and deuterated chloroform (CDCl3) were supplied by Aldrich (Milwaukee, WI). Chloroform (CHCl3) and potassium chloride (KCl) were purchased from Sigma Chemical Company (St. Louis, MO). Phospholipid Preparation Male SHR (250 to 400 g) were used as hypertensive models. Male WKY rats (250 to 400 g) were used as the normotensive controls. Pentobarbital anesthesia (65 mg/kg intraperitoneally) was administered; laparotomy was performed. Fresh aortae, kidneys, and hearts were taken out and frozen immediately by using liquid nitrogen or dry ice. Total lipid was extracted by a modification of the methods of Meneses and Glonek9 and of Bligh and Dyer.10 Frozen tissues were weighed and homogenized, using homogenizer, in a 5 mL mixture of chloroform-methanol (2:1) for 2 min. The homogenates were transferred to flask, and more chloroform-methanol (2:1) mixture was added up to 20 volumes/unit tissue weight (20 mL/g). The mixture was stirred for 20 min and filtered through a sintered glass funnel. The same volume of solution was used to rinse the residuals through the funnel. The filtrate was mixed with 0.2 times its total volume with 0.74% KCl to remove all nonlipid impurities and was allowed to separate thoroughly overnight. The bottom (chloroform) layer was collected and dried in a rotary evaporator at 30°C.9 –12 Dry extracts were dissolved in CDCl3 and used for NMR experiments. Although this method extracts total tissue lipid, membrane phospholipids constitute the predominant component of the lipid extracts. H NMR Measurements 1H NMR spectra for analyzing fatty acid fraction in the phospholipid extracts of aorta, kidney, and heart membranes were obtained at 200 MHz on a Varian XL-200 NMR spectrometer using a spectral width of 5000 Hz, 16 K data points, a flip angle of 60° (10 msec), an acquisition time of 1.64 sec, a 4 sec delay, and 512 transients, at room temperature. Chemical shifts are referenced to tetramethylsilane (TMS) set at 0 ppm.13 Resonance intensities were measured as peak areas. Relative intensities remained unchanged upon doubling the pulse recycle time, 1
342
CHI AND GUPTA
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 1. Typical 1H NMR spectra of lipid extracts from a spontaneously hypertensive rat aorta (upper trace) and a normotensive rat aorta (lower trace). Chemical shifts are referenced to TMS set at 0 ppm. Region A, protons on double bonded carbons. Region B, protons in acylated glycerols, and in phosphatidylcholine and phosphatidylethanolamine. Region C, protons of methylene groups in fatty acid backbone. Region D, protons of terminal methyl groups. Region E, cholesterol C-18 methyl group.
showing that they were unaffected by relaxation times. Calculations Figures 1 and 2 compare the typical 1H NMR spectra of the total lipid extracted from normotensive (lower trace) and hypertensive (upper trace)
rat aorta. The peaks corresponding to protons on double bonded carbons appear at ;5.3 ppm (region A). The protons associated with the fatty acid backbone (–CH2–)n are located in region C (;1.25 ppm), whereas the terminal methyl groups resonate in re-
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 2.
MEMBRANE FATTY ACIDS IN HYPERTENSION
343
H NMR spectra of Figure 1 replotted with vertical scale increased ;5 times.
1
gion D (;0.85 ppm). The phosphatidylcholine methyl group resonates at 3.23 ppm. Assignments are based on those published by Pollesello et al,13 Esclassan et al,14 Pearlman et al,15 and Sze and Jardetzky16 and confirmed in our laboratory. Average unsaturation ratios (a measure of double bonds) were obtained as ratios of intensities of region A over region D in Fig-
ures 1. Average fatty acyl carbon chain lengths were determined by dividing the sum of appropriately normalized proton intensities of all fatty acid carbons by the normalized proton intensity of the methyl carbons (the sum of the integration of region A times 3, the integration of region C times 1.5, the integration of region D and the contribution of –CO2 2 , which is same
344
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
CHI AND GUPTA
TABLE 1. LIPID COMPOSITIONS OF MEMBRANES FROM NORMOTENSIVE AND HYPERTENSIVE RATS Parameter of Lipid Composition
Parameter Definition
Unsaturation ratio
Average chain length
–CHACH– –CH3 –CH3 1 –CH2– 1 –CHA 1 –CO2 2 –CH3
Ratio of glycerol to fatty acyl chains
glycerol –CH3
Ratio of PC to fatty acyl chains
(CH3)3N1– –CH3
Ratio of free and acylated cholesterol to fatty acyl chains
C-18 (cholesterol) –CH3
Organ
N
Normotensive
Hypertensive
P
Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart
5 10 10 5 10 10 5 10 10 5 10 10 5 10 10
0.63 6 0.01 0.84 6 0.03 0.98 6 0.03 18.3 6 0.7 16.4 6 0.4 18.8 6 0.6 0.44 6 0.04 0.27 6 0.04 0.39 6 0.03 0.086 6 0.018 0.31 6 0.02 0.33 6 0.03 0.023 6 0.007 0.071 6 0.003 0.032 6 0.003
0.53 6 0.01 0.70 6 0.01 0.92 6 0.02 15.1 6 0.2 14.5 6 0.2 17.3 6 0.5 0.41 6 0.02 0.32 6 0.02 0.37 6 0.02 0.155 6 0.022 0.30 6 0.02 0.32 6 0.03 0.033 6 0.004 0.059 6 0.002 0.027 6 0.002
.01 .01 .13 .02 .01 .05 .41 .36 .56 .04 .92 .96 .25 .01 .17
Values are mean 6 SEM. Statistical analysis performed with t test. P # .05 was considered significantly different.
as region D, divided by the integration of region D). Ratios of glycerol to fatty acyl chains were the integration of region B minus the integration of peak at 3.23 ppm over the integration of region D. Ratios of PC to fatty acyl chains were the integration of peak at 3.23 ppm over the integration of region D. Ratios of free and acylated cholesterol to fatty acyl chains were the integration of region E over D. Statistical Analysis All data are reported as the mean 6 standard error (SEM). The statistical significance of the differences between SHR and WKY rats was analyzed by the nonpaired t test. A P value # .05 was considered significantly different. RESULTS 1
H NMR spectra of phospholipid extracts from aorta of hypertensive and normotensive rats are compared in Figures 1 and 2. As mentioned previously, terminal methyl groups (–CH3) on fatty acid chains are located in region D (0.85 ppm). All resonance intensities have been normalized relative to the resonances of terminal methyl groups, after taking into account differences in the number of protons in –CHA, –CH2–, and –CH3 groups. These normalized intensities were used to determine all of the parameters in Table 1. Peaks in region A (5.28 to 5.5 ppm) represent the ACH– proton in acyl chains.13–15,17 For the same intensity of region D, the intensity of region A is lower for hypertensive rats than for normotensive rats. As shown in Table 1, the unsaturation ratio (a measure of number of double bonds) in aortic membranes was significantly lower in SHR than in WKY rats (0.53 6 0.01 v 0.63 6 0.01, n 5 5 (two aortae were used to obtain each measurement), P 5 .01). 1H NMR spectra of phospholipid extracts
from kidney and heart were analyzed in a similar manner. Unsaturation ratio (double bonds) was also significantly lower in kidney membranes from SHR than from WKY rats (0.70 6 0.01 v 0.84 6 0.03, n 5 10, P 5 .01) (Table 1). A significant difference in unsaturation ratio for the heart membrane phospholipids between these two strains, however, could not be established. Peaks in region C (Figures 1 to 4) are methylene groups, corresponding to (–CH2–)n (1.2 to 1.7 ppm), and –CH2– groups in –CH2–CHACH2, –CHACH2, and ACH–CH2–CHA (1.8 to 2.9 ppm).13–15,17 The (–CH2–)n carbons make the dominant contribution to the average chain length. From Figure 2, it is not difficult to see that the intensity of region C in the spectra of phospholipid extracts from the aorta is somewhat lower for SHR than for WKY. For all aorta, kidney, and heart membranes, the average chain length in fatty acid fractions of phospholipids was significantly smaller for SHR than for WKY rats (aorta, 15.1 6 0.2 v 18.3 6 0.7, n 5 5, P 5 .02; kidney, 14.5 6 0.2 v 16.4 6 0.4, n 5 10, P 5 .01; heart, 17.3 6 0.5 v 18.8 6 0.6, n 5 10, P 5 .05) (Table 1). Peaks from 3.95 ppm to 5.28 ppm belong to protons on the glycerol backbone (Figures 1 to 4).13–14,17 There was no significant difference between the SHR and WKY rat in ratio of glycerol to fatty acyl chains in phospholipids of aorta, kidney, and heart membranes (Table 1). The peak at 3.2 ppm was assigned to protons in N(CH3)3 groups of PC (Figure 1).13–14 The intensity of this PC resonance in aortic membrane phospholipids from hypertensive rats appeared higher than that from normotensive animals. Statistical analysis
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
MEMBRANE FATTY ACIDS IN HYPERTENSION
345
FIGURE 3. Typical 1H NMR spectra of lipid extract from a rat kidney. The peak assignments are the same as in Figure 1.
showed that the ratio of PC to fatty acyl chains was significantly higher for SHR than for WKY rats (0.155 6 0.022 v 0.086 6 0.018, P 5 .04) (Table 1). However, for phospholipids in kidney and heart membranes, the ratios of PC to fatty acyl chains from SHR and WKY rats were not significantly different (Table 1). The methyl signal appearing at 0.67 ppm, which was indicated as E, was assigned to the C-18 methyl groups of total cholesterol.13,18,19 The ratio of free and acylated cholesterol to fatty acyl chains in kidney membranes is lower in SHR than in WKY rats (Table 1). We did not find a significant difference between these two strains in the ratio of free and acylated cholesterol to fatty acyl chains in either aorta or heart membranes (Table 1). DISCUSSION Phospholipids, which contain fatty acyl chains, are liquid crystalline materials and are able to undergo thermotropic transitions among distinct intermediate phases between the liquid and solid states.15 Calorimetry experiments have shown that bilayers with saturated fatty acids exist in the relatively rigid gel phase; conversely, bilayers composed of monounsaturated fatty acyl chains exist in the liquid crystalline state.20 The degree of unsaturation of acyl chains affects the
phase behavior of phospholipids. Unsaturation induces chain disordering and increasing the degree of unsaturation lowers the phase transition energy; therefore, it lowers the melting point and consequently increases membrane fluidity and elasticity. In general, a single double bond is located at the C-9, 10 position of the parent fatty acid. Additional double bonds may exist at 3-carbon intervals down the chain.21 Our measurement of the unsaturation ratio reflects the average unsaturation in mono-, di-, and polyunsaturated acids. Baenziger et al claimed that phospholipids with diunsaturated acyl chains had higher mobility, because diunsaturated chains can undergo two-side jump motion.20 But Engler et al22 also found that dietary supplementation with n-3 PUFA decreased blood pressure, which might be at least in part explained by increased membrane fluidity.4,23 Thus both di- and polyunsaturated fatty acid chains would appear to lead to increased membrane fluidity. In this study, we found that the unsaturation of fatty acyl chains significantly decreased in aortic membranes from hypertensive rats. This is probably one of the reasons for the increase in stiffness of the aortic wall in hypertensive rats, which results in reduced aortic compliance associated with hypertension.5 Decreased unsaturation in fatty acyl chains, by decreasing fluidity, might alter ion channels in membranes
346
CHI AND GUPTA
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 4. Typical 1H NMR spectra of lipid extract from a rat heart. The peak assignments are the same as in Figure 1.
and thereby alter ion permeability of membranes. Because the function of the kidney is critically dependent on cellular ion balance, lower unsaturation in kidney membranes in hypertensive rats, which was found in this study, may contribute to the derangement of kidney function in hypertension. Our results are in agreement with those of Puddey et al, who found that phospholipids of aorta showed a significant decrease in the polyunsaturated to saturated fatty acid ratio in ethanol induced hypertension in rats.24 These researchers also reported that in these rats blood pressure was significantly negatively correlated to linoleic acid (an unsaturated fatty acid) in aorta.24 In the kidney from ethanol induced hypertensive rats, Puddey et al reported that blood pressure was significantly negatively correlated to arachidonic acid (an unsaturated fatty acid) level.24 Further, in the membrane phospholipids of kidney from ethanol induced hypertensive rats, the ratio of polyunsaturated to saturated (P/S) fatty acids decreased.24 Consistent with our work, Huang et al also found that the percentage of n-3 polyunsaturated fatty acids was significantly lower in SHR than in WKY rats,25 although the percentage of n-9 polyunsaturated fatty acids was higher in SHRs than in WKY rats.25 However, no significant difference in P/S ratio could be established
between these two strains by Huang et al.25 Although our 1H NMR data fail to demonstrate a change in fatty acid unsaturation in cardiac membrane phospholipids, in earlier biochemical work on cultured heart myocytes, the percentages of linoleic acid and eicosapentaenoic acid, respectively, in the fatty acid content were found significantly lower in SHR as compared to WKY rats.26 The ratio of polyunsaturated to saturated fatty acids was also lower.26 However, in cultured heart fibroblasts, there was no significant difference in fatty acids between these two strains.26 For ethanol induced hypertensive rats, in contrast to observations on aorta and kidney, Puddey et al observed a significant increase in the ratio of polyunsaturated to saturated fatty acids of phospholipids in heart membranes.24 Nevertheless, they did not find significant correlations between blood pressure and fatty acids.24 Huang et al reported that the percentage of n-6 polyunsaturated fatty acids was higher in heart membranes from SHR than from WKY rats and that there was no significant difference in P/S ratio between these two strains.25 Unfortunately, our results fail to resolve these controversial findings on cardiac phospholipids. Acyl side chains in naturally occurring glycerophospholipids are derived from unbranched fatty acids, typically 14 to 20 carbon atoms in length.21 The aver-
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
age chain lengths we got from NMR measurements are in this range (Table 1). The average chains are shorter in aorta, kidney, and heart membrane phospholipids from hypertensive rats than from normotensive rats. Longer chains can undergo wider conformational swings,27 whereas shorter chains may be more immobile and lead to greater rigidity in membrane phospholipids. We have found that in the aorta and the kidney, a decrease in both the level of fatty acid unsaturation (double bonds) and the average fatty acid chain length is associated with hypertension. There is ;16% decrease in fatty acid unsaturation and ;2 carbon decrease in average chain length. These changes occurred because of altered turnover or modification of endogenous membrane lipids, even when both normotensive and hypertensive animals were fed identical diets. The hypertension induced decreases in both fatty acid chain length and number of double bonds may reflect enhanced oxidation of double bonds via reactive oxygen radicals, and concomitant chain shortening in the sn-2 position (middle carbon) of the glycerol backbone. This may arise from decreased intracellular free Mg21 in hypertension,28 as Morrill et al8 have shown that Mg21 modulates membrane phospholipids, with reduced Mg21 leading to a loss of unsaturation in fatty acids. An oxidative modification of the polyunsaturated fatty acyl residues of membrane phospholipids in hypertension may also be caused by the increase in intracellular free calcium that occurs in both the aorta and the kidney in hypertension.29,30 An increase in intracellular free calcium is often correlated with increased generation of reactive oxygen species.31 The studies outlined here suggest a novel mechanism for the involvement of membrane phospholipids in hypertension. Fatty acid unsaturation (double bonds) and average fatty acid chain length decrease in both kidney and aorta with hypertension. The beneficial effects of increased dietary intake of polyunsaturated fatty acids in hypertension may arise from increased incorporation of the polyunsaturated fatty acids into membrane phospholipids, which lead to more elastic membranes. On the other hand, increased fatty acid oxidation in hypertension due to decreased intracellular free Mg21 or increased free Ca21 will cause decreased unsaturation ratios and shorter fatty acid chain lengths, leading to more rigid, inelastic membranes. In summary, this study, by using state-of-the-art NMR techniques, identifies structural differences in various membrane lipid components of blood vessels, kidneys, and heart of SHR compared with WKY rats. The significance of this study derives mainly from previous work by our laboratory and by others demonstrating 1) potential functional consequences of al-
MEMBRANE FATTY ACIDS IN HYPERTENSION
347
tered membrane structure, such as changes in ion channel activity and gross tissue (blood vessel) stiffness, as being attributable to the differences observed in the degree of fatty acid unsaturation and in fatty acid acyl chain length in SHR and WKY rat models; 2) conversely, that these very membrane lipid changes may themselves be the result of altered intracellular free magnesium concentrations reported earlier,28 thus linking cellular ionic and lipid abnormalities in hypertension; and lastly 3) therapeutically, that these membrane lipid alterations may help to explain, at least in part, the blood pressure effects of dietary formulations containing greater amounts of unsaturated fatty acids. REFERENCES 1.
Knapp HR, FitzGerald GA: The antihypertensive effects of fish oil, a controlled study of polyunsaturated fatty acid supplements in essential hypertension. N Engl J Med 1989;320:1037–1043.
2.
Bønna KH, Bjerve KS, Straume B, et al: Effects of eicosapentaenoic and docosahexaenoic acids on blood pressure in hypertension. N Engl J Med 1990;322:795– 801.
3.
Levinson PD, Iosiphidis AH, Saritelli AL, et al: Effects of n-3 fatty acids in essential hypertension. Am J Hypertens 1990;3:754 –760.
4.
Watanabe Y, Huang Y, Simmons VA, et al: The effect of dietary n-6 and n-3 polyunsaturated fatty acids on blood pressure and tissue fatty acid composition in spontaneously hypertensive rats. Lipids 1989;24:638 – 644.
5.
Meaney E, Soltero E, Samaniego V, et al: Vascular dynamics in isolated systolic arterial hypertension. Clin Cardiol 1995;18:721–725.
6.
Handa SP, Wong J, Wolf HK: The relationship of plasma norepinephrine levels and aortic distensibility in elderly with isolated systolic hypertension. Am J Hypertens 1995;8:422– 425.
7.
Lehmann ED, Hopkins KD, Parker JR, et al: Hyperlipidaemia, hypertension, and coronary heart disease. Lancet 1995;345:863.
8.
Morrill GA, Gupta RK, Kostellow AB, et al: Mg21 modulates membrane lipids in vascular smooth muscle: a link to atherogenesis. FEBS Lett 1997;408:191–194.
9.
Meneses P, Glonek T: High resolution 31P NMR of extracted phospholipids. J Lipid Res 1988;29:679 – 689.
10.
Bligh EG, Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37:911–917.
11.
Rose HG, Oklander M: Improved procedure for the extraction of lipids from human erythrocytes. J Lipid Res 1965;6:428 – 431.
12.
Merchant TE, Glonek T: 31P NMR of tissue phospholipids: competition for Mg21, Ca21, Na1 and K1 cations. Lipids 1992;27:551–559.
13.
Pollesello P, Masutti F, Croce` LS, et al: 1H NMR spectroscopic studies of lipid extracts from human fatty liver. Biochem Biophys Res Comm 1993;192:1217–1222.
348
14.
Esclassan J, Chemin-Thomas C, Cre´ach Y, et al: Highresolution 1H NMR study of membrane phospholipids in rat pancreas stimulated by caerulein. Pancreas 1994; 9:263–269.
15.
Pearlman JD, Zajicek J, Merickel MB, et al: High-resolution 1H NMR spectral signature from human atheroma. Magn Reson Med 1988;7:262–279.
16.
Sze DY, Jardetzky O: Characterization of lipid composition in stimulated human lymphocytes by 1H-NMR. Biochem Biophys Acta 1990;1054:198 –206.
17.
18.
19.
20.
polyunsaturated fatty acids on glucose homeostasis and blood pressure in essential hypertension. Ann Intern Med 1995;123:911–918. 24.
Puddey IB, Burke V, Croft K, et al: Increased blood pressure and changes in membrane lipids associated with chronic ethanol treatment of rats. Clin Exp Pharm Physiol 1995;22:655– 657.
25.
Wang Y, Hollingsworth RI: A solvent system for the high-resolution proton nuclear magnetic resonance spectroscopy of membrane lipids. Anal Biochem 1995; 225:242–251.
Huang Y, Mills DE, Ward RP, et al: Effect of essential fatty acid depletion on tissue phospholipid fatty acids in spontaneously hypertensive and normotensive rats. Lipids 1989;24:565–571.
26.
Sparling ML, Zidovetzki R, Muller L, et al: Analysis of membrane lipids by 500 MHz 1H NMR. Anal Biochem 1989;178:67–76.
Brussel EM, Simon J, Freyss-Be´guin M: Altered phospholipid fatty acid content and metabolism in heart cell cultures from newborn spontaneously hypertensive rats. Am J Hypertens 1994;7:953–959.
27.
Dea P, Chan SI, Dea FJ: High-resolution proton magnetic resonance spectra of a rabbit sciatic nerve. Science 1972;175:206 –209.
Morrow MR, Singh D, Lu D, et al: Glycosphingolipid fatty acid arrangement in phospholipid bilayers: cholesterol effects. Biophys J 1995;68:179 –186.
28.
Resnick LM, Gupta RK, Laragh JH: Intracellular free magnesium in erythrocytes of essential hypertension: relation to blood pressure and serum divalent cations. Proc Natl Acad Sci 1984;81:6511– 6515.
29.
Jelicks LA, Gupta RK: NMR measurement of cytosolic free calcium, free magnesium, and intracellular sodium in the aorta of the normal and spontaneously hypertensive rat. J Biol Chem 1990;265:1394 –1400.
30.
Dowd TL, Gupta RK: Multinuclear NMR studies of intracellular cations in perfused hypertensive rat kidney. J Biol Chem 1992;267:3637–3643.
31.
Greene EL, Paller MS: Calcium and free radicals I hypoxia/reoxygenation injury of renal epithelial cells. Am J Physiol 1994;266:F13–F20.
Baenziger JE, Jarrell HC, Hill RJ, et al: Average structural and motional properties of a diunsaturated acyl chain in a lipid bilayer: effects of two cis-unsaturated double bonds. Biochem 1991;30:894 –903.
21.
Pearce JM, Komoroski RA: Resolution of phospholipid molecular species by 31P NMR. Magn Reson Med 1993; 29:724 –731.
22.
Engler MB, Engler MM, Ursell PC: Vasorelaxant properties of n-3 polyunsaturated fatty acids in aortas from spontaneously hypertensive and normotensive rats. J Cardiovasc Risk 1994;1:75– 80.
23.
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
CHI AND GUPTA
Toft I, Bonaa KH, Ingebretsen OC, et al: Effects of n-3
1998;11:340 –348
Alterations in Membrane Fatty Acid Unsaturation and Chain Length in Hypertension as Observed by 1 H NMR Spectroscopy Yuling Chi and Raj K. Gupta
Alterations in fatty acids of membrane phospholipids in essential hypertension may account for altered membrane ion transport, elasticity, and contractility properties of hypertensive tissues. To investigate the abnormalities in membrane fatty acids in essential hypertension, the degree of fatty acid unsaturation ([–CHACH–]/[–CH3]), the average carbon chain length, ratio of glycerol to fatty acyl chains, ratio of phosphatidylcholine to fatty acyl chains, and the ratio of free and acylated cholesterol to fatty acyl chains in fatty acid fractions of membrane phospholipids of aorta, kidney, and heart were determined in spontaneously hypertensive rats (SHR) and Wistar-Kyoto (WKY) rats by 1H nuclear magnetic resonance (NMR) spectroscopy. The degrees of fatty acid unsaturation in the aorta and the kidney membranes were significantly lower in SHR than in WKY rats (aorta, 0.53 6 0.01 v 0.63 6 0.01, n 5 5, P 5 .01; kidney, 0.70 6 0.01 v 0.84 6 0.03, n 5 10, P 5 .01). No significant difference could be detected in fatty acid unsaturation in heart membranes between these two strains. For aorta, kidney, and heart membranes, the average
carbon chain lengths of fatty acid fractions of membrane phospholipids were significantly shorter for SHR than for WKY rats (aorta, 15.1 6 0.2 v 18.3 6 0.7, n 5 5, P 5 .02; kidney, 14.5 6 0.2 v 16.4 6 0.4, n 5 10, P 5 .01; heart, 17.3 6 0.5 v 18.8 6 0.6, n 5 10, P 5 .05). The lower unsaturated fatty acid content in membrane phospholipids of the aorta and the kidney, with concomitant reduction in average chain length, may arise from increased oxidation of fatty acid double bonds in hypertensive tissues and may account, in part, for the increased aortic stiffness and abnormal kidney function associated with essential hypertension. Whether the lower unsaturated fatty acid content and decreased carbon chain length of phospholipid membranes in the aorta and the kidney are a cause or a consequence of the high blood pressure, however, remains unknown. Am J Hypertens 1998;11:340 –348 © 1998 American Journal of Hypertension, Ltd.
Received July 7, 1997. Accepted September 29, 1997. From the Department of Physiology and Biophysics, Albert Einstein College of Medicine of Yeshiva University, Bronx, New York. This work was supported by National Institutes of Health Research Grant DK32030.
Address correspondence and reprint requests to Dr. Raj K. Gupta, Department of Physiology and Biophysics, Albert Einstein College of Medicine of Yeshiva University, Bronx, NY 10461; e-mail: [email protected]
© 1998 by the American Journal of Hypertension, Ltd. Published by Elsevier Science, Inc.
Hypertension, fatty acid, 1H nuclear magnetic resonance. KEY WORDS:
0895-7061/98/$19.00 PII S0895-7061(97)00456-1
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
H
ypertension has long been associated with alterations in cell membranes of various tissues. Fatty acid fractions are important constituents of phospholipids in cell membranes. Alterations in fatty acids affect the membrane properties and functions, such as phase transition temperature, elasticity, and ion permeability. Dietary supplementation with polyunsaturated fatty acids (PUFA), which can increase membrane elasticity, can lower blood pressure (BP) in both hypertensive and normotensive humans as well as in animal models.1– 4 We have therefore hypothesized that abnormalities in fatty acid contents of phospholipids in cell membranes of various tissues may be associated with hypertension. It is well accepted that an increase in arterial stiffness and peripheral resistance is the major mechanism responsible for hypertension.5 Meaney et al observed a strong positive correlation between systolic pressure and arterial stiffness.5 Patients who have isolated systolic hypertension have an increase in peripheral resistance.5 Peripheral resistance is in turn determined by arterial diameters, the smaller the arterial diameter, the higher the peripheral resistance. It has been found in hypertensives that, although the diastolic blood pressure was increased, the diameter of the aorta, essentially a large artery, was significantly decreased.6 Aortic stiffness is also significantly increased in hypertension.6 The inelasticity of smooth muscle membranes contributes to aortic stiffness, which has been shown to be significantly related to both blood pressure and the content of atherogenic lipids.7 Changes in fatty acids of phospholipids in membranes result in differences in membrane fluidity and permeability, and may lead to differences in aortic stiffness. Other major target organs for hypertension are the kidney and the heart. Kidney function and hypertension are well known to be intimately related. Hypertension can be both cause and consequence of renal injury. An abnormality in the kidney is often associated with the development of hypertension. Hypertension induced cardiac hypertrophy and other cardiomyopathies have also been described in the literature. We have hypothesized that alterations in membrane fatty acids may occur in these organs as well. In view of the importance of kidney, heart, and vasculature in the development and maintenance of hypertension, and the important role of membranes in their biologic actions, we felt it important to investigate fatty acids in the membranes of hypertensive aorta, kidney, and heart. Proton nuclear magnetic resonance (NMR) spectroscopy was used to quantitate fatty acid unsaturation ratio, average carbon chain length, ratio of glycerol to fatty acyl chains, ratio of phosphatidylcholine to fatty acyl
MEMBRANE FATTY ACIDS IN HYPERTENSION
341
chains, and the ratio of free and acylated cholesterol to fatty acyl chains in fatty acid fractions of the aorta, kidney, and heart membranes from SHR and WKY rats. 1H NMR offers a rapid and comprehensive technique for monitoring lipid composition of cells, body fluids, and tissues. One of us, along with others, recently used it to demonstrate that extracellular Mg21 modulates membrane lipids in vascular smooth muscle.8 The same technique is now being applied to investigate alterations in membrane fatty acids in hypertension. MATERIALS AND METHODS Materials Methanol (anhydrous) (MeOH) and deuterated chloroform (CDCl3) were supplied by Aldrich (Milwaukee, WI). Chloroform (CHCl3) and potassium chloride (KCl) were purchased from Sigma Chemical Company (St. Louis, MO). Phospholipid Preparation Male SHR (250 to 400 g) were used as hypertensive models. Male WKY rats (250 to 400 g) were used as the normotensive controls. Pentobarbital anesthesia (65 mg/kg intraperitoneally) was administered; laparotomy was performed. Fresh aortae, kidneys, and hearts were taken out and frozen immediately by using liquid nitrogen or dry ice. Total lipid was extracted by a modification of the methods of Meneses and Glonek9 and of Bligh and Dyer.10 Frozen tissues were weighed and homogenized, using homogenizer, in a 5 mL mixture of chloroform-methanol (2:1) for 2 min. The homogenates were transferred to flask, and more chloroform-methanol (2:1) mixture was added up to 20 volumes/unit tissue weight (20 mL/g). The mixture was stirred for 20 min and filtered through a sintered glass funnel. The same volume of solution was used to rinse the residuals through the funnel. The filtrate was mixed with 0.2 times its total volume with 0.74% KCl to remove all nonlipid impurities and was allowed to separate thoroughly overnight. The bottom (chloroform) layer was collected and dried in a rotary evaporator at 30°C.9 –12 Dry extracts were dissolved in CDCl3 and used for NMR experiments. Although this method extracts total tissue lipid, membrane phospholipids constitute the predominant component of the lipid extracts. H NMR Measurements 1H NMR spectra for analyzing fatty acid fraction in the phospholipid extracts of aorta, kidney, and heart membranes were obtained at 200 MHz on a Varian XL-200 NMR spectrometer using a spectral width of 5000 Hz, 16 K data points, a flip angle of 60° (10 msec), an acquisition time of 1.64 sec, a 4 sec delay, and 512 transients, at room temperature. Chemical shifts are referenced to tetramethylsilane (TMS) set at 0 ppm.13 Resonance intensities were measured as peak areas. Relative intensities remained unchanged upon doubling the pulse recycle time, 1
342
CHI AND GUPTA
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 1. Typical 1H NMR spectra of lipid extracts from a spontaneously hypertensive rat aorta (upper trace) and a normotensive rat aorta (lower trace). Chemical shifts are referenced to TMS set at 0 ppm. Region A, protons on double bonded carbons. Region B, protons in acylated glycerols, and in phosphatidylcholine and phosphatidylethanolamine. Region C, protons of methylene groups in fatty acid backbone. Region D, protons of terminal methyl groups. Region E, cholesterol C-18 methyl group.
showing that they were unaffected by relaxation times. Calculations Figures 1 and 2 compare the typical 1H NMR spectra of the total lipid extracted from normotensive (lower trace) and hypertensive (upper trace)
rat aorta. The peaks corresponding to protons on double bonded carbons appear at ;5.3 ppm (region A). The protons associated with the fatty acid backbone (–CH2–)n are located in region C (;1.25 ppm), whereas the terminal methyl groups resonate in re-
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 2.
MEMBRANE FATTY ACIDS IN HYPERTENSION
343
H NMR spectra of Figure 1 replotted with vertical scale increased ;5 times.
1
gion D (;0.85 ppm). The phosphatidylcholine methyl group resonates at 3.23 ppm. Assignments are based on those published by Pollesello et al,13 Esclassan et al,14 Pearlman et al,15 and Sze and Jardetzky16 and confirmed in our laboratory. Average unsaturation ratios (a measure of double bonds) were obtained as ratios of intensities of region A over region D in Fig-
ures 1. Average fatty acyl carbon chain lengths were determined by dividing the sum of appropriately normalized proton intensities of all fatty acid carbons by the normalized proton intensity of the methyl carbons (the sum of the integration of region A times 3, the integration of region C times 1.5, the integration of region D and the contribution of –CO2 2 , which is same
344
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
CHI AND GUPTA
TABLE 1. LIPID COMPOSITIONS OF MEMBRANES FROM NORMOTENSIVE AND HYPERTENSIVE RATS Parameter of Lipid Composition
Parameter Definition
Unsaturation ratio
Average chain length
–CHACH– –CH3 –CH3 1 –CH2– 1 –CHA 1 –CO2 2 –CH3
Ratio of glycerol to fatty acyl chains
glycerol –CH3
Ratio of PC to fatty acyl chains
(CH3)3N1– –CH3
Ratio of free and acylated cholesterol to fatty acyl chains
C-18 (cholesterol) –CH3
Organ
N
Normotensive
Hypertensive
P
Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart Aorta Kidney Heart
5 10 10 5 10 10 5 10 10 5 10 10 5 10 10
0.63 6 0.01 0.84 6 0.03 0.98 6 0.03 18.3 6 0.7 16.4 6 0.4 18.8 6 0.6 0.44 6 0.04 0.27 6 0.04 0.39 6 0.03 0.086 6 0.018 0.31 6 0.02 0.33 6 0.03 0.023 6 0.007 0.071 6 0.003 0.032 6 0.003
0.53 6 0.01 0.70 6 0.01 0.92 6 0.02 15.1 6 0.2 14.5 6 0.2 17.3 6 0.5 0.41 6 0.02 0.32 6 0.02 0.37 6 0.02 0.155 6 0.022 0.30 6 0.02 0.32 6 0.03 0.033 6 0.004 0.059 6 0.002 0.027 6 0.002
.01 .01 .13 .02 .01 .05 .41 .36 .56 .04 .92 .96 .25 .01 .17
Values are mean 6 SEM. Statistical analysis performed with t test. P # .05 was considered significantly different.
as region D, divided by the integration of region D). Ratios of glycerol to fatty acyl chains were the integration of region B minus the integration of peak at 3.23 ppm over the integration of region D. Ratios of PC to fatty acyl chains were the integration of peak at 3.23 ppm over the integration of region D. Ratios of free and acylated cholesterol to fatty acyl chains were the integration of region E over D. Statistical Analysis All data are reported as the mean 6 standard error (SEM). The statistical significance of the differences between SHR and WKY rats was analyzed by the nonpaired t test. A P value # .05 was considered significantly different. RESULTS 1
H NMR spectra of phospholipid extracts from aorta of hypertensive and normotensive rats are compared in Figures 1 and 2. As mentioned previously, terminal methyl groups (–CH3) on fatty acid chains are located in region D (0.85 ppm). All resonance intensities have been normalized relative to the resonances of terminal methyl groups, after taking into account differences in the number of protons in –CHA, –CH2–, and –CH3 groups. These normalized intensities were used to determine all of the parameters in Table 1. Peaks in region A (5.28 to 5.5 ppm) represent the ACH– proton in acyl chains.13–15,17 For the same intensity of region D, the intensity of region A is lower for hypertensive rats than for normotensive rats. As shown in Table 1, the unsaturation ratio (a measure of number of double bonds) in aortic membranes was significantly lower in SHR than in WKY rats (0.53 6 0.01 v 0.63 6 0.01, n 5 5 (two aortae were used to obtain each measurement), P 5 .01). 1H NMR spectra of phospholipid extracts
from kidney and heart were analyzed in a similar manner. Unsaturation ratio (double bonds) was also significantly lower in kidney membranes from SHR than from WKY rats (0.70 6 0.01 v 0.84 6 0.03, n 5 10, P 5 .01) (Table 1). A significant difference in unsaturation ratio for the heart membrane phospholipids between these two strains, however, could not be established. Peaks in region C (Figures 1 to 4) are methylene groups, corresponding to (–CH2–)n (1.2 to 1.7 ppm), and –CH2– groups in –CH2–CHACH2, –CHACH2, and ACH–CH2–CHA (1.8 to 2.9 ppm).13–15,17 The (–CH2–)n carbons make the dominant contribution to the average chain length. From Figure 2, it is not difficult to see that the intensity of region C in the spectra of phospholipid extracts from the aorta is somewhat lower for SHR than for WKY. For all aorta, kidney, and heart membranes, the average chain length in fatty acid fractions of phospholipids was significantly smaller for SHR than for WKY rats (aorta, 15.1 6 0.2 v 18.3 6 0.7, n 5 5, P 5 .02; kidney, 14.5 6 0.2 v 16.4 6 0.4, n 5 10, P 5 .01; heart, 17.3 6 0.5 v 18.8 6 0.6, n 5 10, P 5 .05) (Table 1). Peaks from 3.95 ppm to 5.28 ppm belong to protons on the glycerol backbone (Figures 1 to 4).13–14,17 There was no significant difference between the SHR and WKY rat in ratio of glycerol to fatty acyl chains in phospholipids of aorta, kidney, and heart membranes (Table 1). The peak at 3.2 ppm was assigned to protons in N(CH3)3 groups of PC (Figure 1).13–14 The intensity of this PC resonance in aortic membrane phospholipids from hypertensive rats appeared higher than that from normotensive animals. Statistical analysis
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
MEMBRANE FATTY ACIDS IN HYPERTENSION
345
FIGURE 3. Typical 1H NMR spectra of lipid extract from a rat kidney. The peak assignments are the same as in Figure 1.
showed that the ratio of PC to fatty acyl chains was significantly higher for SHR than for WKY rats (0.155 6 0.022 v 0.086 6 0.018, P 5 .04) (Table 1). However, for phospholipids in kidney and heart membranes, the ratios of PC to fatty acyl chains from SHR and WKY rats were not significantly different (Table 1). The methyl signal appearing at 0.67 ppm, which was indicated as E, was assigned to the C-18 methyl groups of total cholesterol.13,18,19 The ratio of free and acylated cholesterol to fatty acyl chains in kidney membranes is lower in SHR than in WKY rats (Table 1). We did not find a significant difference between these two strains in the ratio of free and acylated cholesterol to fatty acyl chains in either aorta or heart membranes (Table 1). DISCUSSION Phospholipids, which contain fatty acyl chains, are liquid crystalline materials and are able to undergo thermotropic transitions among distinct intermediate phases between the liquid and solid states.15 Calorimetry experiments have shown that bilayers with saturated fatty acids exist in the relatively rigid gel phase; conversely, bilayers composed of monounsaturated fatty acyl chains exist in the liquid crystalline state.20 The degree of unsaturation of acyl chains affects the
phase behavior of phospholipids. Unsaturation induces chain disordering and increasing the degree of unsaturation lowers the phase transition energy; therefore, it lowers the melting point and consequently increases membrane fluidity and elasticity. In general, a single double bond is located at the C-9, 10 position of the parent fatty acid. Additional double bonds may exist at 3-carbon intervals down the chain.21 Our measurement of the unsaturation ratio reflects the average unsaturation in mono-, di-, and polyunsaturated acids. Baenziger et al claimed that phospholipids with diunsaturated acyl chains had higher mobility, because diunsaturated chains can undergo two-side jump motion.20 But Engler et al22 also found that dietary supplementation with n-3 PUFA decreased blood pressure, which might be at least in part explained by increased membrane fluidity.4,23 Thus both di- and polyunsaturated fatty acid chains would appear to lead to increased membrane fluidity. In this study, we found that the unsaturation of fatty acyl chains significantly decreased in aortic membranes from hypertensive rats. This is probably one of the reasons for the increase in stiffness of the aortic wall in hypertensive rats, which results in reduced aortic compliance associated with hypertension.5 Decreased unsaturation in fatty acyl chains, by decreasing fluidity, might alter ion channels in membranes
346
CHI AND GUPTA
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
FIGURE 4. Typical 1H NMR spectra of lipid extract from a rat heart. The peak assignments are the same as in Figure 1.
and thereby alter ion permeability of membranes. Because the function of the kidney is critically dependent on cellular ion balance, lower unsaturation in kidney membranes in hypertensive rats, which was found in this study, may contribute to the derangement of kidney function in hypertension. Our results are in agreement with those of Puddey et al, who found that phospholipids of aorta showed a significant decrease in the polyunsaturated to saturated fatty acid ratio in ethanol induced hypertension in rats.24 These researchers also reported that in these rats blood pressure was significantly negatively correlated to linoleic acid (an unsaturated fatty acid) in aorta.24 In the kidney from ethanol induced hypertensive rats, Puddey et al reported that blood pressure was significantly negatively correlated to arachidonic acid (an unsaturated fatty acid) level.24 Further, in the membrane phospholipids of kidney from ethanol induced hypertensive rats, the ratio of polyunsaturated to saturated (P/S) fatty acids decreased.24 Consistent with our work, Huang et al also found that the percentage of n-3 polyunsaturated fatty acids was significantly lower in SHR than in WKY rats,25 although the percentage of n-9 polyunsaturated fatty acids was higher in SHRs than in WKY rats.25 However, no significant difference in P/S ratio could be established
between these two strains by Huang et al.25 Although our 1H NMR data fail to demonstrate a change in fatty acid unsaturation in cardiac membrane phospholipids, in earlier biochemical work on cultured heart myocytes, the percentages of linoleic acid and eicosapentaenoic acid, respectively, in the fatty acid content were found significantly lower in SHR as compared to WKY rats.26 The ratio of polyunsaturated to saturated fatty acids was also lower.26 However, in cultured heart fibroblasts, there was no significant difference in fatty acids between these two strains.26 For ethanol induced hypertensive rats, in contrast to observations on aorta and kidney, Puddey et al observed a significant increase in the ratio of polyunsaturated to saturated fatty acids of phospholipids in heart membranes.24 Nevertheless, they did not find significant correlations between blood pressure and fatty acids.24 Huang et al reported that the percentage of n-6 polyunsaturated fatty acids was higher in heart membranes from SHR than from WKY rats and that there was no significant difference in P/S ratio between these two strains.25 Unfortunately, our results fail to resolve these controversial findings on cardiac phospholipids. Acyl side chains in naturally occurring glycerophospholipids are derived from unbranched fatty acids, typically 14 to 20 carbon atoms in length.21 The aver-
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
age chain lengths we got from NMR measurements are in this range (Table 1). The average chains are shorter in aorta, kidney, and heart membrane phospholipids from hypertensive rats than from normotensive rats. Longer chains can undergo wider conformational swings,27 whereas shorter chains may be more immobile and lead to greater rigidity in membrane phospholipids. We have found that in the aorta and the kidney, a decrease in both the level of fatty acid unsaturation (double bonds) and the average fatty acid chain length is associated with hypertension. There is ;16% decrease in fatty acid unsaturation and ;2 carbon decrease in average chain length. These changes occurred because of altered turnover or modification of endogenous membrane lipids, even when both normotensive and hypertensive animals were fed identical diets. The hypertension induced decreases in both fatty acid chain length and number of double bonds may reflect enhanced oxidation of double bonds via reactive oxygen radicals, and concomitant chain shortening in the sn-2 position (middle carbon) of the glycerol backbone. This may arise from decreased intracellular free Mg21 in hypertension,28 as Morrill et al8 have shown that Mg21 modulates membrane phospholipids, with reduced Mg21 leading to a loss of unsaturation in fatty acids. An oxidative modification of the polyunsaturated fatty acyl residues of membrane phospholipids in hypertension may also be caused by the increase in intracellular free calcium that occurs in both the aorta and the kidney in hypertension.29,30 An increase in intracellular free calcium is often correlated with increased generation of reactive oxygen species.31 The studies outlined here suggest a novel mechanism for the involvement of membrane phospholipids in hypertension. Fatty acid unsaturation (double bonds) and average fatty acid chain length decrease in both kidney and aorta with hypertension. The beneficial effects of increased dietary intake of polyunsaturated fatty acids in hypertension may arise from increased incorporation of the polyunsaturated fatty acids into membrane phospholipids, which lead to more elastic membranes. On the other hand, increased fatty acid oxidation in hypertension due to decreased intracellular free Mg21 or increased free Ca21 will cause decreased unsaturation ratios and shorter fatty acid chain lengths, leading to more rigid, inelastic membranes. In summary, this study, by using state-of-the-art NMR techniques, identifies structural differences in various membrane lipid components of blood vessels, kidneys, and heart of SHR compared with WKY rats. The significance of this study derives mainly from previous work by our laboratory and by others demonstrating 1) potential functional consequences of al-
MEMBRANE FATTY ACIDS IN HYPERTENSION
347
tered membrane structure, such as changes in ion channel activity and gross tissue (blood vessel) stiffness, as being attributable to the differences observed in the degree of fatty acid unsaturation and in fatty acid acyl chain length in SHR and WKY rat models; 2) conversely, that these very membrane lipid changes may themselves be the result of altered intracellular free magnesium concentrations reported earlier,28 thus linking cellular ionic and lipid abnormalities in hypertension; and lastly 3) therapeutically, that these membrane lipid alterations may help to explain, at least in part, the blood pressure effects of dietary formulations containing greater amounts of unsaturated fatty acids. REFERENCES 1.
Knapp HR, FitzGerald GA: The antihypertensive effects of fish oil, a controlled study of polyunsaturated fatty acid supplements in essential hypertension. N Engl J Med 1989;320:1037–1043.
2.
Bønna KH, Bjerve KS, Straume B, et al: Effects of eicosapentaenoic and docosahexaenoic acids on blood pressure in hypertension. N Engl J Med 1990;322:795– 801.
3.
Levinson PD, Iosiphidis AH, Saritelli AL, et al: Effects of n-3 fatty acids in essential hypertension. Am J Hypertens 1990;3:754 –760.
4.
Watanabe Y, Huang Y, Simmons VA, et al: The effect of dietary n-6 and n-3 polyunsaturated fatty acids on blood pressure and tissue fatty acid composition in spontaneously hypertensive rats. Lipids 1989;24:638 – 644.
5.
Meaney E, Soltero E, Samaniego V, et al: Vascular dynamics in isolated systolic arterial hypertension. Clin Cardiol 1995;18:721–725.
6.
Handa SP, Wong J, Wolf HK: The relationship of plasma norepinephrine levels and aortic distensibility in elderly with isolated systolic hypertension. Am J Hypertens 1995;8:422– 425.
7.
Lehmann ED, Hopkins KD, Parker JR, et al: Hyperlipidaemia, hypertension, and coronary heart disease. Lancet 1995;345:863.
8.
Morrill GA, Gupta RK, Kostellow AB, et al: Mg21 modulates membrane lipids in vascular smooth muscle: a link to atherogenesis. FEBS Lett 1997;408:191–194.
9.
Meneses P, Glonek T: High resolution 31P NMR of extracted phospholipids. J Lipid Res 1988;29:679 – 689.
10.
Bligh EG, Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37:911–917.
11.
Rose HG, Oklander M: Improved procedure for the extraction of lipids from human erythrocytes. J Lipid Res 1965;6:428 – 431.
12.
Merchant TE, Glonek T: 31P NMR of tissue phospholipids: competition for Mg21, Ca21, Na1 and K1 cations. Lipids 1992;27:551–559.
13.
Pollesello P, Masutti F, Croce` LS, et al: 1H NMR spectroscopic studies of lipid extracts from human fatty liver. Biochem Biophys Res Comm 1993;192:1217–1222.
348
14.
Esclassan J, Chemin-Thomas C, Cre´ach Y, et al: Highresolution 1H NMR study of membrane phospholipids in rat pancreas stimulated by caerulein. Pancreas 1994; 9:263–269.
15.
Pearlman JD, Zajicek J, Merickel MB, et al: High-resolution 1H NMR spectral signature from human atheroma. Magn Reson Med 1988;7:262–279.
16.
Sze DY, Jardetzky O: Characterization of lipid composition in stimulated human lymphocytes by 1H-NMR. Biochem Biophys Acta 1990;1054:198 –206.
17.
18.
19.
20.
polyunsaturated fatty acids on glucose homeostasis and blood pressure in essential hypertension. Ann Intern Med 1995;123:911–918. 24.
Puddey IB, Burke V, Croft K, et al: Increased blood pressure and changes in membrane lipids associated with chronic ethanol treatment of rats. Clin Exp Pharm Physiol 1995;22:655– 657.
25.
Wang Y, Hollingsworth RI: A solvent system for the high-resolution proton nuclear magnetic resonance spectroscopy of membrane lipids. Anal Biochem 1995; 225:242–251.
Huang Y, Mills DE, Ward RP, et al: Effect of essential fatty acid depletion on tissue phospholipid fatty acids in spontaneously hypertensive and normotensive rats. Lipids 1989;24:565–571.
26.
Sparling ML, Zidovetzki R, Muller L, et al: Analysis of membrane lipids by 500 MHz 1H NMR. Anal Biochem 1989;178:67–76.
Brussel EM, Simon J, Freyss-Be´guin M: Altered phospholipid fatty acid content and metabolism in heart cell cultures from newborn spontaneously hypertensive rats. Am J Hypertens 1994;7:953–959.
27.
Dea P, Chan SI, Dea FJ: High-resolution proton magnetic resonance spectra of a rabbit sciatic nerve. Science 1972;175:206 –209.
Morrow MR, Singh D, Lu D, et al: Glycosphingolipid fatty acid arrangement in phospholipid bilayers: cholesterol effects. Biophys J 1995;68:179 –186.
28.
Resnick LM, Gupta RK, Laragh JH: Intracellular free magnesium in erythrocytes of essential hypertension: relation to blood pressure and serum divalent cations. Proc Natl Acad Sci 1984;81:6511– 6515.
29.
Jelicks LA, Gupta RK: NMR measurement of cytosolic free calcium, free magnesium, and intracellular sodium in the aorta of the normal and spontaneously hypertensive rat. J Biol Chem 1990;265:1394 –1400.
30.
Dowd TL, Gupta RK: Multinuclear NMR studies of intracellular cations in perfused hypertensive rat kidney. J Biol Chem 1992;267:3637–3643.
31.
Greene EL, Paller MS: Calcium and free radicals I hypoxia/reoxygenation injury of renal epithelial cells. Am J Physiol 1994;266:F13–F20.
Baenziger JE, Jarrell HC, Hill RJ, et al: Average structural and motional properties of a diunsaturated acyl chain in a lipid bilayer: effects of two cis-unsaturated double bonds. Biochem 1991;30:894 –903.
21.
Pearce JM, Komoroski RA: Resolution of phospholipid molecular species by 31P NMR. Magn Reson Med 1993; 29:724 –731.
22.
Engler MB, Engler MM, Ursell PC: Vasorelaxant properties of n-3 polyunsaturated fatty acids in aortas from spontaneously hypertensive and normotensive rats. J Cardiovasc Risk 1994;1:75– 80.
23.
AJH–MARCH 1998 –VOL. 11, NO. 3, PART 1
CHI AND GUPTA
Toft I, Bonaa KH, Ingebretsen OC, et al: Effects of n-3