Inhibitory Effects of Insulin on Cytosolic Ca2+ Level and Contraction in the Rat Aorta
Endothelium-Dependent and -Independent Mechanisms
Abstract To determine the mechanism of the inhibitory effect of insulin on vascular tone, contraction was measured simultaneously with endothelial and smooth muscle cytosolic Ca2+ level ([Ca2+]i) in the isolated rat aorta. Insulin (200 mU/mL) increased endothelial [Ca2+]i and decreased resting muscle tone. The removal of endothelium abolished the effects of insulin. In the aorta precontracted with norepinephrine, insulin (3 to 120 mU/mL) induced concentration-dependent inhibition of contraction. The relaxant effect followed the increase in endothelial [Ca2+]i and decrease in smooth muscle [Ca2+]i. The relaxant effect was attenuated by removal of endothelium or by the addition of 10−5 mol/L NG-monomethyl-l-arginine but not by 10−5 mol/L indomethacin. In the absence of endothelium, the relaxant effect of insulin followed the decrease in smooth muscle [Ca2+]i. These results suggest that insulin inhibits vascular contraction by dual mechanisms in the isolated rat aorta: (1) Insulin acts on vascular endothelium by increasing endothelial [Ca2+]i and releasing NO, which decreases smooth muscle [Ca2+]i and the Ca2+ sensitivity of the contractile elements. (2) Insulin also directly acts on smooth muscle and decreases smooth muscle [Ca2+]i.
- cytosolic Ca2+
- endothelium-derived relaxing factor
- rat aortic smooth muscle
- rat aortic endothelium
Hyperinsulinemia has long been postulated to contribute to the pathogenesis of hypertension associated with obesity and diabetes because of its stimulating action on sympathetic activity and renal sodium reabsorption. Recently, however, it was observed that plasma insulin levels were increased in hypertensive patients and that the ability of insulin to produce vasodilation was impaired, suggesting a state of insulin resistance in peripheral tissues.1 2 Thus, it is postulated that insulin resistance is the primary anomaly in hypertension associated with obesity and diabetes.3 4 5 6 7 8
Previous studies have shown that insulin causes vasodilatation in humans1 2 9 10 11 12 and inhibits the vasoconstricting action of norepinephrine, phenylephrine, and angiotensin II in isolated arteries and veins.12 13 14 15 As for the mechanism of vasodilator action of insulin, it has been suggested to be related to a β-adrenergic mechanism,9 hyperpolarization of the cell membrane,16 stimulation of the plasma membrane Ca2+-ATPase activity,14 inhibition of Ca2+ channels,17 18 and endothelium-dependent19 and -independent20 mechanisms. Since smooth muscle contraction is regulated by the cytosolic Ca2+ level ([Ca2+]i), insulin would decrease smooth muscle [Ca2+]i. If insulin released NO from vascular endothelium, insulin would increase endothelial [Ca2+]i, thereby activating NO synthase activity.21 22 23 However, the effects of insulin on smooth muscle and endothelial [Ca2+]i have not been clarified.
The purpose of the present study was to clarify the role of vascular endothelium in the vasodilator action of insulin. For this purpose, we compared the effects of insulin on contraction in the isolated rat aorta with and without endothelium and measured the effects of insulin on [Ca2+]i in endothelial and smooth muscle cells.
Materials and Methods
The chemicals used in the present study were porcine insulin crystal (Sigma Chemical Co), NE (Sigma), ACh (Sigma), L-NMMA (Hoechst Japan), fura 2-AM (Dojindo Laboratories), EDTA (Dojindo), cremophor EL (Nacalai Tesque), indomethacin (Hoechst Japan), and DMSO (Wako). Porcine insulin was dissolved in 0.01N HCl and then diluted to a desired concentration with PSS.
Muscle Preparations and Solution
Male Wistar rats (4 to 5 weeks old) were killed by a blow on the neck and exsanguination. The thoracic aorta was rapidly removed and placed into normal PSS at 22°C. The aortas were dissected free of fat and connective tissue. For measurements of contractile force, the aorta was cut into 2-mm-long ring segments. For measurements of [Ca2+]i, the aorta was cut into spiral strips (2 mm wide and 10 mm long). In some experiments, vascular endothelium was removed by gently rubbing the intimal surface with a finger moistened with PSS. PSS contained (mmol/L) NaCl 136.9, KCl 5.4, glucose 5.5, NaHCO3 23.8, CaCl2 1.5, MgCl2 1.2, and EDTA 0.01. High K+ (65.4 mmol/L) solution was made by replacing 60 mmol/L NaCl in PSS with equimolar KCl. These solutions were saturated with a mixture of 95% O2/5% CO2 at 37°C (pH 7.4).
Measurement of Contractile Tension
Muscle tension was measured isometrically with a force displacement transducer (Nihon Kohden). Passive tension of 10 mN was initially applied, and tissues were allowed to equilibrate for 60 minutes before the beginning of the experimental period. The endothelial cells were considered to be intact if 10−6 mol/L ACh, a potent stimulant of NO release,24 almost completely (>80%) relaxed the contraction induced by 10−7 mol/L NE. In the strips from which the endothelium was removed, the 10−6 mol/L ACh–induced relaxation was <10%. Muscle tension was expressed as percentage of the contraction induced by 65.4 mmol/L KCl, which is less sensitive to the relaxant effect of NO than are the contractions induced by NE.25 In experiments in which NE pretreatment preceded the addition of insulin, NE-stimulated contraction before insulin administration was taken as 100%.
In some experiments, NE was added cumulatively, with insulin (12 or 120 mU/mL) being administered 30 minutes before the addition of NE. In other experiments, insulin (1 to 120 mU/mL) was applied cumulatively when the contraction induced by NE reached a steady level. The concentration of NE or insulin was increased only after the response to the previous concentration had attained a steady level. The experiment on the control strips, which received the same concentration of the vehicle, was performed at the same time.
To examine whether endogenous prostaglandins are involved in the inhibitory effect of insulin, the muscle strips with endothelium were treated with 10−5 mol/L indomethacin dissolved in DMSO or with the same concentration of the vehicle for 15 minutes, and then NE was added followed by a cumulative addition of insulin. L-NMMA, a specific inhibitor of NO synthesis, was used to assess the NO-dependent component of the relaxing response to insulin. In the strips with endothelium, the relaxation caused by a cumulative application of insulin (1 to 120 mU/mL) was measured in the presence and absence of 10−5 mol/L L-NMMA. In some experiments, 10−5 mol/L L-NMMA was added when the insulin-induced relaxation reached a steady level.
Measurement of Endothelial [Ca2+]i and Smooth Muscle [Ca2+]i
[Ca2+]i was measured simultaneously with contraction, as described previously,21 26 27 28 by using a fluorescent Ca2+ indicator, fura 2. Briefly, the muscle strips were loaded with 7 μmol/L fura 2-AM for 4 to 5 hours in the presence of 0.02% cremophor EL at room temperature. The fura 2–loaded strips were washed with normal PSS for 30 minutes in a tissue bath at 37°C to remove unhydrolyzed fura 2-AM. Measurements of [Ca2+]i were performed with a fluorometer (CAF-100, JASCO). The muscle strip was held horizontally on a silicon rubber sheet in an organ bath (containing 5 mL PSS at 37°C) that was attached to the fluorometer and bubbled with 95% O2/5% CO2. One end of the strip was pinned to the silicon rubber sheet, and the other end was connected to a force transducer (Orientec). The resting tension was adjusted to 10 mN. The muscle strip was illuminated alternately with 340- and 380-nm lights, and 500-nm emission was detected. The intensity of the 500-nm fluorescence induced by the 340-nm excitation (F340) and that induced by the 380-nm excitation (F380) was measured, and the ratio of these two fluorescence values (F340/F380) was calculated. Because the fura 2 content and dissociation constant of fura 2 for Ca2+ in the endothelial cells may be different from those in smooth muscle cells, it is difficult to calculate the absolute amounts of [Ca2+]i, especially in the strips containing both types of cells.21 Therefore, in the present experiments, we used the relative F340/F380 ratio as an indicator of [Ca2+]i, taking the ratio in the resting muscle as 0% and that in the presence of 65 mmol/L K+ (a selective stimulant of smooth muscle) or 10−6 mol/L ACh (a selective stimulant of endothelium in the rat aorta)21 as 100%.
Smooth muscle [Ca2+]i was measured by using the strips without endothelium. In some experiments, smooth muscle [Ca2+]i was measured by using the strips with endothelium, detecting the fluorescence from the adventitial surface of the strips. Because ultraviolet light is absorbed by vascular tissues, fluorescence derived from endothelium was not detected by this method.21
Endothelial [Ca2+]i was measured by detecting the fura 2 fluorescence from the endothelial surface of the strips simultaneously with contractile tension. Since the fluorescence derived from underlying smooth muscle cells is also detected by this method, results were compared with smooth muscle [Ca2+]i, and changes in endothelial [Ca2+]i were estimated.21
Data were analyzed by one-factor ANOVA. When statistically significant effects were found, a Newman-Keuls test was performed to isolate the differences between the groups. Student’s t test for unpaired data was used when appropriate. A value of P<.05 was considered significant. All data are presented as mean±SEM.
Effects of Insulin on Vascular Contraction
Fig 1⇓ shows the effect of insulin on the contraction induced by the cumulative addition of NE in the rat aortic strips without endothelium. The cumulative addition of NE in control strips (without insulin) induced a concentration-dependent contraction with an EC50 (concentration needed to induce a 50% maximum contraction) of 5×10−9 mol/L. In the presence of insulin (12 or 120 mU/mL, 30-minute preincubation), NE-induced contraction was inhibited, shifting the concentration-response curve for NE to the right and downward.
Fig 2⇓ shows the effect of the cumulative addition of insulin on NE-induced contraction in the rat aortic strips with endothelium. NE (5×10−8 mol/L) caused a sustained contraction in normal PSS (7.4±1.1 mN, n=8) (Fig 2a⇓). The cumulative application of insulin (1 to 120 mU/mL) after NE induced a concentration-dependent relaxation of the contraction (Fig 2b⇓). The concentration-response curve for insulin is shown in Fig 2c⇓. The threshold concentration for insulin to inhibit the contraction was 1 mU/mL, and the contraction was inhibited by 64.3±3.5% (n=8) in the presence of 120 mU/mL insulin. The IC50 (concentration needed to produce a 50% maximum inhibition) for insulin was 17.1±1.3 mU/mL (n=8).
In the strips without endothelium, the concentration-response curve for NE was shifted to the left, decreasing the EC50 to 1×10−9 mol/L (data not shown). In the absence of endothelium, the contraction induced by 1×10−8 mol/L NE (7.8±1.3 mN, n=8) was inhibited by the cumulative addition of insulin (3 to 120 mU/mL) in a concentration-dependent manner (Fig 2c⇑). Compared with the effect of insulin in the presence of endothelium, the effect in the absence of endothelium was significantly smaller. The IC50 for insulin in the aorta without endothelium was 35.6±1.5 mU/mL (n=8, P<.01 versus aorta with endothelium), and 120 mU/mL insulin inhibited the contraction by only 31.3±3.2% (P<.01 versus with endothelium).
The effect of 10−5 mol/L L-NMMA on the relaxation induced by insulin in the strips with endothelium is also shown in Fig 2c⇑. In the aortic strips pretreated with 1×10−5 mol/L L-NMMA for 15 minutes, 5×10−8 mol/L NE induced sustained contraction (7.0±1.5 mN, n=6). This contraction was inhibited by insulin, although more weakly than in the absence of L-NMMA. The IC50 for insulin was 17.1±1.2 mU/mL in the absence of the L-NMMA and 34.4±1.1 mU/mL in the presence of L-NMMA (n=8 each, P<.01). It was also observed that when the inhibitory effect of 60 mU/mL insulin reached a steady level, the addition of 10−5 mol/L L-NMMA reversed the inhibition (Fig 3⇓, n=6). In contrast, in the absence of endothelium, L-NMMA was ineffective (data not shown).
We also examined the effect of indomethacin on the insulin-induced inhibition of NE contraction in the strips with endothelium. The results indicated that pretreatment with 10−5 mol/L indomethacin for 15 minutes did not affect the insulin-induced inhibition. The IC50 values for insulin in the absence and presence of indomethacin were 17.1±1.3 and 15.8±1.7 mU/mL (P>.05), respectively, and the maximal inhibition was 64.4±4.5% and 59.6±4.8%, respectively (P>.05, n=8 each).
Effects of Insulin on Endothelial [Ca2+]i and Smooth Muscle [Ca2+]i
The effect of insulin on endothelial [Ca2+]i is shown in Fig 4⇓. High K+ (65 mmol/L) induced a sustained increase in [Ca2+]i and a sustained contraction in the strips with (Fig 4a⇓) and without (Fig 4b⇓) endothelium. In the resting strips with endothelium, 10−6 mol/L ACh increased [Ca2+]i to 59.3±3.8% of high K+–stimulated [Ca2+]i and decreased resting tone by 13.4±2.2% of high K+–induced contraction (n=7). After several washes with PSS, the addition of 200 mU/mL insulin increased [Ca2+]i to 70.0±6.0% of the ACh-induced increase in [Ca2+]i and caused a small relaxation (68.0±5.0% of ACh-induced relaxation, n=7) (Fig 4a⇓). In the strips without endothelium, neither insulin (200 mU/mL) nor ACh (10−6 mol/L) changed [Ca2+]i and resting tone (Fig 4b⇓).
The effect of insulin on smooth muscle [Ca2+]i in the strips with endothelium in the presence of NE is shown in Fig 5a⇓ and 5b⇓. Since fura 2–Ca2+ fluorescence was detected from the adventitial surface of the strips, [Ca2+]i derived from endothelial cells was not detected, and only [Ca2+]i derived from smooth muscle cells was detected (see “Materials and Methods”). NE (5×10−8 mol/L) induced a sustained increase in smooth muscle [Ca2+]i and a sustained contraction (Fig 5a⇓). Sequential addition of 12 and 120 mU/mL insulin decreased both [Ca2+]i and contraction (Fig 5b⇓). At a concentration of 12 mU/mL, insulin inhibited contraction by 25.3±1.2% and the [Ca2+]i by 17.1±0.8%, whereas 120 mU/mL insulin inhibited contraction by 72.5±2.1% and [Ca2+]i by 39.7±1.8% (n=8). These results indicate that insulin inhibits the contraction more strongly than the [Ca2+]i in the strips with endothelium (P<.05).
Fig 5c⇑ shows the effects of insulin on the mixed (endothelial and smooth muscle) [Ca2+]i in the aortic strips stimulated by 5×10−8 mol/L NE. In this experiment, fura 2–Ca2+ fluorescence was measured from the endothelial surface; therefore, both endothelial [Ca2+]i and smooth muscle [Ca2+]i were measured simultaneously. It is shown that NE induced a sustained increase in [Ca2+]i. The addition of 120 mU/mL insulin induced an additional transient increase in [Ca2+]i (to 19.4±2.3% of the NE-induced increase), followed by a decrease (by 29.2±3.0%) and relaxation (by 57.1±3.6%, n=5). Comparing the results in Fig 5b⇑ (which shows measurement of smooth muscle [Ca2+]i) and Fig 5c⇑ (which shows measurement of mixed [Ca2+]i), it is suggested that insulin increased endothelial [Ca2+]i and decreased smooth muscle [Ca2+]i.
The effect of insulin on the smooth muscle [Ca2+]i in the aorta without endothelium is shown in Fig 6⇓. The fura 2–Ca2+ fluorescence was derived from luminal surface from which the endothelium was removed. The addition of 10−8 mol/L NE induced a sustained increase in [Ca2+]i and a sustained contraction. Insulin (12 and 120 mU/mL) inhibited both the [Ca2+]i and contraction stimulated by 10−8 mol/L NE. At a concentration of 12 mU/mL, insulin inhibited the contraction by 16.4±2.2% and [Ca2+]i by 12.5±1.8%, whereas 120 mU/mL insulin inhibited the contraction by 29.2±2.3% and [Ca2+]i by 24.7±2.5% (n=8).
In the present study, we found that NE-induced contraction was inhibited in the presence of insulin in the rat aorta. We also observed that the addition of insulin during the sustained contraction induced by NE caused a concentration-dependent relaxation in the aortic strips with and without endothelium and that the relaxation in the aorta with endothelium was significantly greater than that in the aorta without endothelium (Figs 1⇑ and 2⇑). Previously, it has been shown that insulin produces endothelium-dependent relaxation in human coronary, pulmonary, and radial arteries.19 These findings led us to examine the mechanism of insulin-induced endothelium-mediated relaxation. It has been shown that endothelial cells release locally acting vasodilating prostaglandins, such as prostaglandin E2 and prostaglandin I2. In the present experiment, however, indomethacin did not change the insulin-induced relaxation, suggesting that release of the prostaglandins is not responsible for the relaxation induced by insulin. In contrast, a specific inhibitor of NO synthesis, L-NMMA, significantly inhibited the insulin-induced relaxation in the presence of endothelium (Figs 2⇑ and 3⇑). These results suggest that insulin releases NO from endothelial cells, thereby inhibiting smooth muscle contraction.
Synthesis and release of NO are regulated by endothelial [Ca2+]i, which stimulates NO synthase.21 22 23 In the present study, we found that insulin increased endothelial [Ca2+]i and decreased resting tone (Figs 4a⇑ and 5c⇑). These effects were similar to those of muscarinic receptor agonists.21 These results, together with the fact that insulin-induced relaxation was inhibited by L-NMMA, suggest that insulin increases endothelial [Ca2+]i, activates NO synthase, releases NO, and inhibits smooth muscle contraction.
In smooth muscle, NO activates soluble guanylate cyclase to increase the synthesis of cGMP.29 Elevation of the cGMP accelerates the Ca2+ pump and/or inhibits Ca2+ channels, thereby decreasing intracellular Ca2+ levels.26 cGMP also decreases Ca2+ sensitivity of myosin light chain phosphorylation in smooth muscle cells.26 27 30 Therefore, the increase in cGMP results in a greater inhibition on contraction than that predicted by the decrease in [Ca2+]i. In the present study, insulin (12 and 120 mU/mL) inhibited NE-induced contraction more strongly than smooth muscle [Ca2+]i (Fig 5b⇑), suggesting that the endothelium-dependent relaxation caused by insulin is due not only to the decrease in smooth muscle [Ca2+]i but also to the decrease in Ca2+ sensitivity of the contractile elements. The findings in the present study are consistent with a recent report showing that insulin attenuates NE-induced vasoconstriction in healthy normotensive volunteers by a mechanism that involves a cGMP-dependent pathway.12
In the present study, we also found that insulin inhibits the NE-induced contraction even in the absence of endothelium, although the inhibition was smaller than that observed in the presence of endothelium (Fig 2⇑). The inhibitory effect was associated with a decrease in the NE-stimulated [Ca2+]i (Fig 6b⇑). Insulin has been shown to inhibit inward Ca2+ current17 and to stimulate the Na+-K+ pump, which leads to hyperpolarization of the cell membrane, thereby decreasing Ca2+ influx via voltage-dependent Ca2+ channels.16 It has also been reported that insulin increases Ca2+-ATPase activity in the plasma membrane, thereby increasing Ca2+ extrusion from the cell.14 These effects may explain how insulin decreases smooth muscle [Ca2+]i. Hyperpolarization of endothelial cells increases Ca2+ influx by increasing the electrical gradient,31 and this may be the mechanism by which insulin increases endothelial [Ca2+]i.
In the present study, insulin increased endothelial [Ca2+]i, decreased smooth muscle [Ca2+]i, and relaxed the muscle within 5 minutes. It has been shown that there are specific binding sites for insulin in the arterial endothelial and smooth muscle cells of several mammalian species.32 33 The vasodilator action of insulin may be mediated by these receptors. Since we have examined the effects of short-term application of high concentrations of insulin in vitro, our results may not be directly related to the in vivo chronic action of hyperinsulinemia. However, a recent study18 has shown that physiological concentrations of insulin reduced vascular contraction and [Ca2+]i both in short-term (20-minute) and long-term (7-day) experiments. Thus, it is postulated that the decreased vascular effects of insulin may contribute to the development of hypertension in insulin-resistant states such as obesity and diabetes. Further experiments are needed to examine this possibility.
In summary, our results indicate that insulin inhibits vascular contraction by dual mechanisms in the isolated rat aorta: (1) Insulin acts on vascular endothelium by increasing endothelial [Ca2+]i and releasing NO, which decreases smooth muscle [Ca2+]i and the Ca2+ sensitivity of the contractile elements. (2) Insulin directly acts on smooth muscle to decrease smooth muscle [Ca2+]i.
Selected Abbreviations and Acronyms
This study was supported by a grant-in-aid for Scientific Research from the Ministry of Education, Science, and Culture, Japan, and a grant from the Japan Milk Association.
- Received March 20, 1995.
- Accepted June 22, 1995.
- © 1995 American Heart Association, Inc.
Laakso M, Edelman SV, Brechtel G, Baron AD. Impaired insulin-mediated skeletal muscle blood flow in patients with NIDDM. Diabetes. 1992;41:1076-1083.
Laakso M, Edelman SV, Brechtel G, Baron AD. Decreased effect of insulin to stimulate skeletal muscle blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest. 1990;85:1844-1853.
Gans ROB, Bilo HJG, Nauta JJP, Heine RJ, Doner AJM. Acute hyperinsulinemia induces sodium retention and blood pressure decline in diabetes mellitus. Hypertension. 1992;20:199-209.
Anderson EA, Hoffman RP, Balon TW, Sinkey CA, Mark AL. Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal human. J Clin Invest. 1991;87:2246-2252.
Epstein M, Sowers JR. Diabetes mellitus and hypertension. Hypertension. 1992;19:403-418.
Creager MA, Liang CS, Coffman JD. Beta-adrenergic mediated vasodilator response to insulin in the human forearm. J Pharmacol Exp Ther. 1985;235:709-714.
Anderson EA, Hoffman RP, Balon TW, Sinkey CA, Mark AL. Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J Clin Invest. 1991;87:2246-2252.
Sakai K, Imaizumi T, Masaki H, Takeshita A. Intra-arterial infusion of insulin attenuates vasoreactivity in human forearm. Hypertension. 1993;22:67-73.
Grover A, Padington C, Wilson MF, Sung BH, Izzo JL Jr, Dandona P. Insulin attenuates norepinephrine-induced venoconstriction: an ultrasonographic study. Hypertension. 1995;25(pt 2):779-784.
Alexander WD, Oake RJ. The effect of insulin on vascular reactivity to norepinephrine. Diabetes. 1977;26:611-614.
Zemel MB, Johnson BA, Ambrozy SA. Insulin-stimulated vascular relaxation: role of Ca2+-ATPase. Am J Hypertens. 1992;5:637-641.
Yagi S, Takata S, Kiyokawa H, Yamamoto M, Noto Y, Ikeda T, Hattori N. Effects of insulin on vasoconstrictive responses to norepinephrine and angiotensin II in rabbit femoral artery and vein. Diabetes. 1988;37:1064-1067.
Prakash TR, MacKenzie SJ, Ram JL, Sowers JR. Insulin (INS) stimulates gene transcription and activity of Na+-K+ ATPase in vascular smooth muscle cells (VSMC). Hypertension. 1992;20:443. Abstract.
Standley PR, Zhang F, Ram LJ, Zemel MB, Sowers JR. Insulin attenuates vasopressin-induced calcium transients and a voltage dependent calcium response in rat vascular smooth muscle cells. J Clin Invest. 1991;88:1230-1236.
Kahn AM, Seidel CL, Allen JC, O’Neil RG, Shelat H, Song T. Insulin reduces contraction and intracellular calcium concentration in vascular smooth muscle. Hypertension. 1993;22:735-742.
Thom S, Hughes A, Sever PS. Endothelium-dependent responses in human arteries. In: Vanhoutte PM, ed. Relaxing and Contracting Factors. Clifton, NJ: Humana Press; 1988.
Sato K, Ozaki H, Karaki H. Differential effects of carbachol cytosolic calcium levels in vascular endothelium and smooth muscle. J Pharmacol Exp Ther. 1990;255:114-119.
Sato K, Ozaki H, Karaki H. Changes in cytosolic calcium level in vascular smooth muscle strips measured simultaneously with contraction using fluorescent calcium indicator fura 2. J Pharmacol Exp Ther. 1988;246:294-300.
Rapoport RM, Murad F. Agonist-induced endothelium-dependent relaxation in rat thoracic aorta may be mediated through cGMP. Circ Res. 1983;52:352-357.
Rembold CM. Regulation of contraction and relaxation in arterial smooth muscle. Hypertension. 1992;20:129-137.
Adam DJ. Ionic channels in vascular endothelial cells. Trends Cardiovasc Med. 1994;4:18-26.
Kaiser N, Vlodavsky I, Tur-Sinai A, Fuks Z, Cerasi E. Binding, internalization and degradation of insulin in vascular endothelial cells. Diabetes. 1982;31:1077-1083.
Kaiser N, Tur-Sinai A, Hasin M, Cerasi E. Binding, degradation, and biological activity of insulin in vascular smooth muscle cells. Am J Physiol. 1985;249:E292-E298.