methacholine or sodium nitroprusside at baseline and after 2 and 4 h. Forearm blood flow was measured by venous occlusion plethysmography. Lipid oxidation ...
Clinical Science (2000) 99, 169–174 (Printed in Great Britain)
The impairment in endothelial function induced by non-esterified fatty acids can be reversed by insulin Lars LIND*, Andreas FUGMANN*, Stefan BRANTH†, Bengt VESSBY‡, Jonas MILLGA/ RD*, Christian BERNE* and Hans LITHELL‡ *Department of Internal Medicine, University Hospital of Uppsala, S-75185 Uppsala, Sweden, †Department of Nutrition, University Hospital of Uppsala, S-75185 Uppsala, Sweden, and ‡Department of Geriatrics, University Hospital of Uppsala, S-75185 Uppsala, Sweden
A
B
S
T
R
A
C
T
Dyslipidaemia, with elevations of circulating triacylglycerols (triglycerides) and non-esterified (free) fatty acids, and hyperinsulinaemia are often found in the same subjects, the so-called ‘ insulin resistance syndrome ’. The present study aims to investigate how elevated levels of nonesterified fatty acids, hyperinsulinaemia and the combination of these factors affects endothelium-dependent vasodilatation (EDV). Ten volunteers were examined on two occasions. Intralipid2 (plus heparin) or saline (0.9 % NaCl) was infused for 4 h. During the final 2 h, euglycaemic hyperinsulinaemia (80p4 m-units/l) was imposed. EDV and endothelium-independent vasodilatation were evaluated in the forearm by local intra-arterial infusion of methacholine or sodium nitroprusside at baseline and after 2 and 4 h. Forearm blood flow was measured by venous occlusion plethysmography. Lipid oxidation was determined by measuring plasma malondialdehyde levels. Infusion of Intralipid2 plus heparin increased the concentration of non-esterified fatty acids to 2.6p1.2 mmol/l and decreased EDV from 27.6p8.7 to 21.0p5.7 ml:min−1:100 ml−1 tissue (P 0.01). This effect was completely reversed by hyperinsulinaemia (P 0.01). Hyperinsulinaemia alone increased EDV (to 30.4p 9.5 ml:min−1:100 ml−1 tissue ; P 0.01), while endothelium-independent vasodilatation was unaltered by the interventions. Infusion of Intralipid2 plus heparin increased malondialdehyde levels from 0.67p0.22 to 1.2p0.37 µmol/l (P 0.001), while hyperinsulinaemia did not change the malondialdehyde level. In conclusion, an acute increase in serum levels of nonesterified fatty acids increased lipid oxidation and decreased EDV. The effect on EDV of non-esterified fatty acids could be reversed by hyperinsulinaemia.
INTRODUCTION More than 10 years ago it was discovered that the endothelium plays an important role not only in regulating vascular tone, but also in leucocyte adhesion, thrombocyte activation and vascular remodelling, due to
secretion of substances such as nitric oxide (NO), prostacyclin and endothelin [1–3]. The endothelium is likely, therefore, to play a central role in the development of cardiovascular diseases associated with atherosclerosis plaque evolution and thrombus formation. Consequently, an understanding of factors governing en-
Key words : blood flow, endothelium, fatty acids, insulin, lipids. Abbreviations : EIDV, endothelium-independent vasodilatation ; EDV, endothelium-dependent vasodilatation ; FBF, forearm blood flow ; MDA, malondialdehyde ; NEFA, non-esterified fatty acids ; SNP, sodium nitroprusside. Correspondence : Dr Lars Lind (e-mail lars.lind!medsci.uu.se).
# 2000 The Biochemical Society and the Medical Research Society
169
170
L. Lind and others
dothelial function will be important for the prevention of cardiovascular disorders. In addition to high levels of insulin being regarded as a cardiovascular risk factor, it has been shown that insulin can induce vasodilatation [4,5]. Insulin-mediated vasodilatation is attenuated by concomitant infusion of the L-arginine analogue NG-monomethyl-L-arginine, which inhibits NO synthesis [6,7], suggesting that insulinmediated vasodilatation is endothelium-dependent. Insulin infusion has also been found to enhance methacholine-induced endothelium-dependent vasodilatation (EDV) [6], providing further evidence that the vasodilatation evoked by insulin is endothelium-mediated. Insulin is anti-lipolytic, lowering the serum concentration of non-esterified (free) fatty acids (NEFA), an action that is impaired in insulin-resistant states. Oleic acid has been found to attenuate acetylcholine-induced endothelial-dependent relaxation of precontracted rabbit aortic rings in vitro [8], and infusion with Intralipid2 attenuated EDV in humans [9]. Thus a decrease in serum NEFA induced by insulin might also be involved in the potentiation of EDV seen during hyperinsulinaemia. Dyslipidaemia, with elevated serum NEFA and triacylglycerols, often co-exists with insulin resistance and hyperinsulinaemia in subjects with cardiovascular disorders [10–13]. In order to understand in more detail how NEFA and insulin affect EDV under normal conditions and how these factors act together on the endothelium, acute changes in these metabolic variables were induced in young, healthy volunteers. Based on previous data on the action of NEFA and insulin on EDV, we tested whether insulin could counteract the expected negative effect of NEFA on EDV, at least in part. As it has been suggested that the impairment in EDV induced by certain NEFA in vitro could be due to lipid oxidation [8], we also measured the circulating levels of plasma malondialdehyde (MDA), a marker of lipid oxidation, during the studies.
MATERIALS AND METHODS Ten young, healthy volunteers [mean age 23 years (range 20–26 years) ; six males and four females ; mean body mass index 23.2p2.4 kg\m#], without any known metabolic or cardiovascular abnormalities, were randomly allocated to either of two investigations with a 2–3-week interval. All studies began in the morning after an overnight fast. During the experiments, an intravenous infusion of Intralipid2 (200 mg\ml ; Pharmacia & Upjohn, Uppsala, Sweden) plus heparin or of saline (0.9 % NaCl) was administered for 4 h (Figure 1). During the final 2 h of the infusions a euglycaemic hyperinsulinaemic clamp was added. EDV and endothelium-independent vasodilatation (EIDV) were examined before the start of the # 2000 The Biochemical Society and the Medical Research Society
Figure 1
Experimental protocol
The protocol was carried out twice in each individual : once with Intralipid2 plus heparin and once with saline. The thin horizontal arrows indicate the different infusions. The vertical arrows indicate the time points at which EDV and EIDV were measured. infusions and after 2 h and 4 h of infusion. The protocol was repeated after 2–3 weeks, with the exception that saline was given if Intralipid2 had been given on the first occasion, and vice versa. Intralipid2 was given as a bolus injection of 0.5 ml\kg over 10 min and thereafter as a continuous intravenous infusion at a rate of 90 ml\h. Saline was infused in a similar manner. For the Intralipid2 infusions, intermittent injections of heparin (Pharmacia & Upjohn, Uppsala, Sweden) were also given intravenously every 5 min, at a dose corresponding to a rate of 150 units\h. The euglycaemic hyperinsulinaemic clamp was performed according to DeFronzo et al. [14]. After a priming dose given over 10 min, insulin (Actrapid Human2 ; Novo Nordisk, Bagsvaerd, Denmark) was infused at a constant rate of 56 m-units:min−":m−# body surface area. The arterial plasma glucose concentration was determined at 5 min intervals (Beckman Glucose Analyzer II ; Beckman Instruments, Fullerton, CA, U.S.A.). The rate of infusion of a 20 % (w\v) glucose solution was adjusted in order to achieve a target concentration of arterial plasma glucose of 5.1 mmol\l. The coefficient of variation for plasma glucose was 5 %. During the studies, the subjects rested supine in a room at constant temperature (20–22 mC). An arterial catheter (1.0 mm ; Ohmeda, Swindon, U.K.) was inserted into the brachial artery for regional infusion of methacholine or sodium nitroprusside (SNP). Resting forearm blood flow (FBF) was measured before the two separate infusions (methacholine and SNP). Both infusions were given in two dosage steps, of 5 min for each step, with a 20 min wash-out period between the two different vasodilators. The infused doses were 2 and 4 µg\min for methacholine (for evaluation of EDV) and 5 and 10 µg\min for SNP (for evaluation of EIDV). The two infusions were given in a random order and at a rate of 1 ml\min. Previous control studies have shown that these infusion rates do not induce any
Non-esterified fatty acids, insulin and endothelial function
alterations in systemic haemodynamics, or in blood flow in the contra-lateral arm [15]. We have recently found, in a pilot study in healthy volunteers, that methacholine (4 µg\min) induced a significant increase in forearm venous plasma nitrite and nitrate concentrations (L. Lind and A. Larsson, unpublished work). When plasma nitrite and nitrate concentrations were measured in both arterial blood and forearm venous blood samples, the forearm release of these two breakdown products of NO increased by more than 8-fold, indicating that the vasodilatory capacity of methacholine is mediated by an increase in NO production. FBF was measured by venous occlusion plethysmography. A mercury-in-silastic strain gauge was placed on the upper third of the forearm, which rested comfortably slightly above the level of the heart. The strain gauge was coupled to a calibrated plethysmograph. Venous occlusion was achieved by a blood pressure cuff applied proximal to the elbow and inflated to 40 mmHg by a rapid cuff inflator. FBF was calculated from the mean of at least five consecutive recordings. The reproducibility of this test of EDV and EIDV was evaluated in the present study. Both after 2 h of saline infusion and when repeated after 3 weeks during baseline conditions, resting FBF showed a variation of 10 %. The variation in FBF during the two vasodilatory infusions was 5 %, after both 2 h and 3 weeks. In addition to measuring absolute FBF values obtained during infusion with methacholine or SNP, an endothelium function index was calculated. This index was defined as the ratio between FBF seen during infusion of methacholine at the highest dose (4 µg\min) and FBF obtained with the highest dose of SNP (10 µg\min). As SNP induces vasodilatation independently of the endothelium, this index expresses the contribution of the endothelium to the vasodilatory process. Serum NEFA, serum triacylglycerols, plasma insulin and plasma MDA were measured before the start of the infusion in the fasting state, and subsequently after 2 and 4 h of infusion with Intralipid2jheparin or saline. Due to technical problems, MDA was only measured in eight of the subjects. Insulin was assayed in EDTA plasma in duplicate, using an enzymic immunological assay (Enzymmun2 ; Boehringer, Mannheim, Germany) performed in an ES300 automatic analyser (Boehringer). NEFA and triacylglycerol concentrations in serum were assayed by enzymic techniques in a Monarch 2000 centrifugal analyser (Instrumentation Laboratories, Lexington, MA, U.S.A.). MDA was determined as described by O$ hrvall et al. [16]. In brief, 750 µl of 0.15 M phosphoric acid and 300 µl of thiobarbituric acid were added to 200 µl of plasma. After boiling for 60 min, the mixture was cooled on ice. Methanol was added to the MDA\thiobarbituric acid mixture, and MDA in 20 µl of the mixture was measured
on an HPLC system using a fluorescence detector with an excitation wavelength of 532 nm and an emission wavelength of 553 nm. Calibration was carried out using a standard solution of 1,1,3,3-tetraethoxypropane (Sigma Chemical Co., St. Louis, MO, U.S.A.). The study was approved by the Human Ethics Committee of the Medical Faculty of Uppsala University, and the participants gave their informed consent. Differences in EDV and EIDV under the different metabolic conditions were evaluated by ANOVA ; P 0.05 was regarded as significant.
RESULTS A 2 h saline infusion did not significantly alter serum NEFA, serum triacylglycerol or plasma insulin levels (Table 1). A 2 h infusion of Intralipid2 plus heparin increased the circulating level of NEFA from 0.43p0.25 to 2.6p1.2 mmol\l (P 0.001). Serum triacylglycerols increased similarly, while the plasma levels of insulin did not change. Euglycaemic hyperinsulinaemia alone increased plasma insulin to 80p4 m-units\l, while serum NEFA (P 0.001), but not serum triacylglycerols, were significantly suppressed. The combination of hyperinsulinaemia and Intralipid2 infusion did not result in significantly different levels of serum NEFA and triacylglycerols when compared with Intralipid2 infusion alone. Similarly, the combination did not produce significantly different levels of plasma insulin compared with those induced by hyperinsulinaemia alone (Table 1). After 2 h of saline infusion, no significant changes were seen in resting FBF or in the vasodilatation produced by methacholine or SNP (Table 2). In contrast, 2 h of infusion of Intralipid2 plus heparin induced a significant increase in resting FBF, and also produced a significant decline in the vasodilatation induced by the two different doses of methacholine (P 0.01), but not in that induced by SNP. Hyperinsulinaemia alone also increased resting FBF, but, in contrast with the response to Intralipid2 infusion, FBF was significantly enhanced at both doses of methacholine (P 0.01). Hyperinsulinaemia had no significant effects on the vasodilatation induced by SNP (Table 2). When infusion of Intralipid2 plus heparin was combined with euglycaemic hyperinsulinaemia, the impaired methacholine-induced vasodilatation induced by the acute elevation of NEFA was completely reversed at both doses of methacholine (P 0.01) (Figure 2). No significant effects on SNP-induced vasodilatation were seen when Intralipid2 infusion was combined with hyperinsulinaemia (Table 2). When blood flow was expressed as the endothelial function index, rather than as absolute FBF values, during methacholine and SNP infusions, no change was induced by saline infusion (from 1.09p0.17 at baseline to # 2000 The Biochemical Society and the Medical Research Society
171
172
L. Lind and others
Table 1 Serum triacylglycerols, serum NEFA and plasma insulin at baseline, during saline infusion and during the different metabolic conditions
Values are meanspS.D. Significance of differences : ***P 0.001 compared with baseline and saline infusion at 120 min ; †††P 0.001 compared with hyperlipidaemia at 120 min. Conditions
Time (min)
Serum triacylglycerols (mmol/l)
Serum NEFA (mmol/l)
Plasma insulin (m-units/l)
Baseline Saline Hyperlipidaemia Hyperinsulinaemia Hyperlipidaemiajhyperinsulinaemia
0 120 120 240 240
1.03p0.27 0.93p0.29 3.9p0.29*** 0.76p0.28 4.7p2.0***
0.43p0.25 0.45p0.11 2.6p1.2*** 0.05p0.08*** 2.6p1.5***
9.9p3.8 8.6p4.2 9.6p4.6 80p4*** 84p29***†††
Table 2 Resting FBF and vasodilatation during local infusions of methacholine or SNP at baseline, during saline infusion and under the different metabolic conditions
Values are meanspS.D. Significance of differences : **P compared with hyperlipidaemia at 120 min.
0.01 compared with baseline and saline at 120 min ; ††P
0.01
FBF (ml:min−1:100 ml−1 tissue)
Conditions Baseline Saline Hyperlipidaemia Hyperinsulinaemia Hyperlipidaemiaj hyperinsulinaemia
Figure 2
Time (min)
Resting
jMethacholine (2 µg/min)
jMethacholine (4 µg/min)
jSNP (5 µg/min)
jSNP (10 µg/min)
0 120 120 240 240
3.9p1.1 4.2p1.2 5.7p1.6** 6.3p1.6** 8.4p2.9**††
22.4p7.1 21.9p6.9 17.3p6.7** 24.2p7.7** 23.9p7.3††
27.6p8.7 27.3p8.4 21.0p5.7** 30.4p9.5** 28.9p6.7††
18.7p4.2 19.3p4.5 14.9p6.5 19.8p5.6 21.4p5.8
23.7p5.8 23.5p6.7 24.5p7.9 24.7p7.7 26.5p5.0
Effects of different treatments on FBF
FBF was measured during local infusion with methacholine (4 µg/min) in order to assess EDV during the four different experimental conditions : control (saline) ; elevated serum NEFA (FFA) during infusion of Intralipid2 plus heparin ; the same protocol combined with euglycaemic hyperinsulinaemia ; and a euglycaemic insulin clamp with concomitant saline infusion. Significance of differences : **P 0.01. 1.10p0.16). Intralipid2 plus heparin infusion reduced the endothelial function index (from 1.11p0.14 at baseline to 0.90p0.22 ; P 0.05), whereas, in contrast, euglycaemic hyperinsulinaemia increased this index of endothelial vasodilatory function (from 1.10p0.16 to 1.18p0.20 ; P 0.05). When euglycaemic hyperinsulinaemia was combined with the Intralipid2 plus heparin # 2000 The Biochemical Society and the Medical Research Society
Figure 3 MDA
Effects of different treatments on plasma levels of
The four different experimental conditions were : control (saline) ; elevated serum NEFA (FFA) during infusion of Intralipid2 plus heparin ; the same protocol combined with euglycaemic hyperinsulinaemia ; and a euglycaemic insulin clamp with concomitant saline infusion. Significance of difference : ***P 0.001. infusion, an improvement in the endothelial function index was seen compared with that with infusion alone (from 0.90p0.22 to 1.09p0.17 ; P 0.05), resulting in the same level as was seen at baseline. Whereas euglycaemic hyperinsulinaemia alone did not affect plasma levels of MDA, infusion of Intralipid2 plus heparin induced a significant rise in MDA levels (P 0.001) (Figure 3). A tendency for a decrease in plasma
Non-esterified fatty acids, insulin and endothelial function
MDA levels was seen when hyperinsulinaemia was combined with Intralipid2 plus heparin infusion (P l 0.07) (Figure 3). A weak, non-significant relationship was seen between the change in MDA levels induced by Intralipid2 plus heparin infusion and the change in methacholine-induced vasodilatation (r l k0.37).
DISCUSSION The present study shows that an acute elevation in NEFA and serum triacylglycerols, produced by infusion of Intralipid2 plus heparin, lowered EDV but not EIDV, in accordance with a previous study [9]. Furthermore, this effect of an acute elevation of NEFA was accompanied by an increase in plasma MDA levels, and was completely reversed by an elevation in plasma insulin. An acute rise in circulating NEFA impaired EDV in humans in vivo, similar to what was shown previously in the rabbit aorta in vitro [8]. There are several steps at which EDV could be influenced by NEFA. First, infusion of Intralipid2 lowers the circulating levels of most amino acids, including arginine [17], which may render the substrate for formation of NO less readily available and thereby impair NO production. Impaired EDV due to inhibition of NO synthesis by fatty acids is a second alternative [18]. In both in vitro experiments [19] and human studies [20], formation of oxygen free radicals has been linked to exposure to fatty acids. This heterogeneous group of highly reactive compounds interacts with NO and thereby inhibits vasodilatation caused by NO. This is thought to be one mechanism by which oleic acid impairs endothelium-dependent relaxation, as endothelial dysfunction in the presence of oleic acid was reversed by the addition of the oxygen free radical scavenger superoxide dismutase [8]. In the present study, plasma levels of MDA, formed as a product of lipid oxidation, increased substantially during administration of Intralipid2, suggesting that an interaction between NO and oxygen free radicals could be the cause of the impaired EDV. When the change in MDA levels was related to the change in EDV during administration of Intralipid2 plus heparin, however, only a non-significant inverse correlation was seen. This might be due to the fact either that lipid oxidation, as measured by MDA levels, only plays a minor role in the impairment of EDV induced by an acute rise in NEFA, or that the sample size of the present study was too small to detect a significant relationship. EDV is impaired in subjects with hypercholesterolaemia [21,22]. In experimental studies this has been attributed to the presence of oxidized LDL [23,24], and other studies have shown oleic acid to be a major component of the oxidized LDL fraction [8]. Taken together, these results suggest some common mechanisms by which high levels of NEFA and hyper-
cholesterolaemia might impair EDV. It seems unlikely that the elevation in serum triacylglycerols induced by infusion of Intralipid2 plus heparin would result in impaired EDV, since infusion with Intralipid2 without heparin, which results in an elevation of serum triacylglycerols without a change in NEFA levels, did not impair EDV [9]. The present study corroborates previous findings which showed that hyperinsulinaemia enhanced EDV produced by methacholine or acetylcholine [6]. Furthermore, hyperinsulinaemia reversed the adverse effects of elevated NEFA levels on EDV. This is most likely to be due to a stimulation of NO synthase activity, since the vasodilatory effect of insulin seems to be mediated by NO [6,7]. The possibility that insulin enhances EDV through its anti-lipolytic effect [9] is unlikely, since serum NEFA levels were unaltered when hyperinsulinaemia restored the EDV during Intralipid2 infusion. Furthermore, circulating MDA levels were only lowered to a small extent during hyperinsulinaemia, suggesting that the improvement in EDV induced by the addition of insulin was not solely due to a decrease in the formation of oxidation products. The two metabolic abnormalities (elevated levels of NEFA and hyperinsulinaemia) affected endothelial function in different ways in this sample of young, healthy individuals. Insulin-resistant subjects are also resistant to the vasodilatory action of insulin [25], and in these subjects the effect of NEFA on EDV might be dominant. It remains to be shown, however, how insulin-resistant subjects respond to elevations in NEFA and insulin. Furthermore, in the present study, acute major alterations within the physiological range of NEFA and insulin levels were induced in order to explore the physiological effects of these metabolic conditions on endothelial function. It must be remembered, therefore, that similar effects on endothelial function might not be found in insulin-resistant subjects, in whom more modest and chronic elevations of NEFA and insulin are seen. In the present study, Intralipid2 plus heparin, as well as hyperinsulinaemia alone or in combination with elevated serum NEFA levels, all increased basal FBF. We have data indicating that the evaluation of endothelial function is not related to basal FBF [15]. In the present study the effect of SNP was virtually unaffected by the different metabolic conditions, despite the fact that basal FBF changed. This observation suggests that changes following acute intervention should be evaluated by the absolute levels of FBF and not in relation to basal FBF. Furthermore, we also used the endothelial function index in the evaluation of the effects of NEFA and insulin. This index was defined as the ratio between FBF during infusion of methacholine and FBF during infusion of SNP. As SNP induces vasodilatation that is independent of the endothelium, this index expresses the contribution of the endothelium to the vasodilatory # 2000 The Biochemical Society and the Medical Research Society
173
174
L. Lind and others
process. This index has the advantage that effects on endothelial vasodilatory function could be evaluated independently of changes in basal FBF. Similar conclusions regarding the effects of NEFA and insulin were obtained using this index of endothelial function as when the absolute levels of FBF during methacholine and SNP treatments were used. In conclusion, the present study shows that an acute elevation of NEFA induced by intravenous infusion of Intralipid2 plus heparin decreased EDV and increased lipid oxidation in young, healthy subjects. This effect on EDV was fully reversed by hyperinsulinaemia under euglycaemic conditions.
REFERENCES 1 Moncada, S., Herman, A. G. and Vanhoutte, P. M. (1987) Endothelium-derived relaxing factor is identified as nitric oxide. Trends Pharmacol. Sci. 8, 365–368 2 Furchgott, R. F. and Vanhoutte, P. M. (1989) Endothelium-derived relaxing and contracting factors. FASEB J. 3, 2007–2018 3 Cohen, R. A. and Vanhoutte, P. M. (1995) Endotheliumdependent hyperpolarization. Circulation 92, 3337–3349 4 Anderson, E. A., Hoffman, R. P., Balon, T. W., Sinkey, C. A. and Mark, A. L. (1991) Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J. Clin. Invest. 87, 2246–2252 5 Baron, A. D., Brechtel-Hook, G., Johnson, A. and Hardin, D. (1993) Skeletal muscle blood flow. A possible link between insulin resistance and blood pressure. Hypertension 21, 129–135 6 Steinberg, H. O., Brechtel, G., Johnson, A., Fineberg, N. and Baron, A. D. (1994) Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent. A novel action of insulin to increase nitric oxide release. J. Clin. Invest. 94, 1172–1179 7 Scherrer, U., Randin, D., Vollenweider, P., Vollenweider, L. and Nicod, P. (1994) Nitric oxide release accounts for insulin’s vascular effects in humans. J. Clin. Invest. 94, 2511–2515 8 Niu, X. L., Liu, L. Y., Hu, M. L. and Chen, X. (1995) Some similarities in vascular effects of oleic acid and oxidized low-density lipoproteins on rabbit aorta. J. Mol. Cell. Cardiol. 27, 531–539 9 Steinberg, H. O., Tarshoby, M., Monstrel, R. et al. (1997) Elevated circulating free fatty acid levels impair endothelium-dependent vasodilatation. J. Clin. Invest. 100, 1230–1239 10 Lind, L., Berne, C. and Lithell, H. (1995) Prevalence of insulin resistance in essential hypertension. J. Hypertens. 13, 1457–1462
11 Reaven, G. (1988) Role of insulin resistance in human disease. Diabetes 37, 1595–1607 12 Modan, M., Halkin, H., Almog, S. et al. (1985) Hyperinsulinemia. A link between hypertension, obesity and glucose intolerance. J. Clin. Invest. 75, 809–817 13 Pollare, T., Lithell, H. and Berne, C. (1990) Insulin resistance is a characteristic feature of primary hypertension independent of obesity. Metab. Clin. Exp. 39, 167–174 14 DeFronzo, R. A., Tobin, J. D. and Andres, R. (1979) Glucose clamp technique : a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237, 214–223 15 Lind, L., Sarabi, M. and Millga/ rd, J. (1998) Methodological aspects of the evaluation of endothelium-dependent vasodilatation in the human forearm. Clin. Physiol. 18, 81–87 16 O$ hrvall, M., Tengblad, S., Ekstrand, B., Siegbahn, A. and Vessby, B. (1994) Malondialdehyde concentration in plasma is inversely correlated to the proportion of linoleic acid in serum lipids. Atherosclerosis 108, 103–110 17 Ferrannini, E., Barrett, E. J., Bevilacqua, S. et al. (1986) Effect of free fatty acids on blood amino acid levels in human. Am. J. Physiol. 250, E686–E694 18 Davda, R. K., Stepniakowski, K. T., Lu, G., Ullian, M. E., Goodfriend, T. L. and Egan, B. M. (1995) Oleic acid inhibits endothelial nitric oxide synthase by a protein kinase C-independent mechanism. Hypertension 26, 764–770 19 Toborek, M. and Henning, B. (1994) Fatty acid-mediated effects on the glutathione redox cycle in cultured endothelial cells. Am. J. Nutr. 59, 60–65 20 Paolisso, G., Gambardella, A., Tagliamonte, M. R. et al. (1996) Does free fatty acid infusion impair insulin action also through an increase in oxidative stress ? J. Clin. Endocrinol. Metab. 81, 4244–4248 21 Creager, M. A., Gallagher, S. J., Girerd, X. J., Coleman, S. M., Dzau, V. J. and Cooke, J. P. (1992) L-arginine improves endothelium-dependent vasodilatation in hypercholesterolemic humans. J. Clin. Invest. 90, 1248–1253 22 Gilligan, D., Sack, M. N., Guetta, V. et al. (1994) Effect of antioxidant vitamins on low density lipoprotein oxidation and impaired endothelium-dependent vasodilatation in patients with hypercholesterolemia. J. Am. Coll. Cardiol. 24, 1611–1617 23 Mangin, Jr, E. L., Kugiyama, K., Nguy, J. H., Kerns, S. A. and Henry, P. D. (1993) Effects of lysolipids and oxidatively modified low density lipoprotein on endothelium-dependent relaxation of rabbit aorta. Circulation 72, 161–166 24 Simon, B. C., Cunningham, L. D. and Cohen, R. A. (1990) Oxidized low density lipoproteins cause contraction and inhibit endothelium-dependent relaxation in the pig coronary artery. J. Clin. Invest. 86, 75–79 25 Laakso, M., Edelman, S. V., Brechtel, G. and Baron, A. D. (1990) Decreased effect of insulin to stimulate skeletal muscle blood flow in obese man. J. Clin. Invest. 85, 1844–1852 Received 9 December 1999/3 March 2000; accepted 4 May 2000
# 2000 The Biochemical Society and the Medical Research Society
