Chronic exposure to cyclosporine affects endothelial and smooth muscle reactivity in the rat aorta

Chronic Exposure to Cyclosporine Affects Endothelial and Smooth Muscle Reactivity in the Rat Aorta Raymond Cartier, MD, Francois Dagenais, MD, Charleen Hollmann, RN, Helen Cambron, and Josie Buluran, BS Department of Cardiovascular Surgery, Montreal Heart Institute, Montreal, Quebec, Canada

Chronic exposure to cyclosporine affects vascular reactivity. Experiments were designed to characterize the endothelium-dependent and endothelium-independent vascular reactivity of rats exposed to oral cyclosporin A (CyA). Two subsets of rats (n = 6) were treated with CyA (20 mglkg/day) and olive oil (cyclosporine vehicle), respectively, for a period of 8 weeks. Aortic rings (4-5 mm) were suspended for isometric force measurement in organ chambers containing Krebs Ringer solution (37°C, 95% 02' 5% CO 2). The maximal endothelium-dependent relaxation to cumulative doses of acetylcholine was significantly decreased in the CyA-treated aortic rings compared to olive oil-treated ones (data expressed as percent of initial contraction; CyA, 50% ± 3% versus olive oil, 37% ± 7%; P < 0.05). However, endothelium-dependent relaxations to histamine and adenosine diphosphate and

endothelium-independent relaxation to sodium nitroprusside were not affected in both groups. An endothelium-dependent contraction to serotonin and aggregating platelets were observed in the CyA group, but not in the control group. The endothelium-independent contraction to norepinephrine was enhanced in the Cy A group (CyA ED so, log -7.66 ± 0.18 mollL versus olive oil ED so, log -7.01 ± 0.11 mollL; p < 0.01). These experiments suggest that chronic exposure to cyclosporine A could contribute to augmenting vascular tone by (1) decreased release of endothelial relaxing factor mediated by muscarinic receptors, (2) increased production of endothelium-related constricting factor mediated by serotoninergic receptors, and (3) greater vascular smooth muscle sensitivity to circulating catecholamine. (Ann Thorae Surg 1994;58:789-94)

C

in vitro, the effects of long-term administration of CyA on endothelial and smooth muscle reactivity of the rat aorta.

yclosp orin A (CyA) commonly is accepted as the immunosuppressive drug of choice in human organ transplantation. Even though this drug has improved survival of organ recipients, significant adverse side effects have been encountered. Systemic hypertension and nephrotoxicity are two major side effects that seem to be related to CyA vascular toxicity [1-3]. Although the etiology of CyA toxicity is not thoroughly understood, mechanisms have been suggested. In animals, renal infusion of CyA enhances release of endothelin and thromboxane and decreases the endothelial production of relaxant factors such as nitric oxide, thus contributing to increased renal vessel tone [4-6]. In peripheral arteries, CyA increases sympathetic tone and vascular smooth muscle contractile responsiveness to circulating catecholamine [7, 8]. Furthermore, CyA could also enhance platelet aggregation and thromboxane A 2 release [9]. All of these effects may contribute not only to systemic hypertension and nephrotoxicity, but also to graft deterioration (by promoting thrombotic complications), endothelial toxicity, and accelerated coronary atherosclerosis. Therefore, the present study was designed to characterize, Accepted for publication Jan 18, 1994. Address reprint requests to Dr Cartier, Department of Cardiovascular Surgery, Montreal Heart Institute, 5000 Belanger St East, Montreal, Que, Canada HIT 1C8.

© 1994 by The Society of Thoracic Surgeons

Material and Methods Animal Preparation Male Sprague-Dawley rats (initial body weight, 100 to 150 g) were maintained on a standard laboratory diet. In group I (n = 6), rats were treated daily with oral CyA (20 mg/kg) for a period of 8 weeks. In group II (n = 6), rats were given a similar volume of olive oil (cyclosporine vehicle) for the same period of time. Both CyA and olive oil were administered daily to the animals using a feeding tube. CyA was obtained from Sandoz Ltd (Dorval, Que, Canada) and dissolved in pure olive oil. At the end of the treatment period, animals were heparinized and anesthetized (pentobarbital, 20 mg/kg), and then exsanguinated and the thoracic aorta carefully harvested. The aorta was divided into four rings of 4-5 mm each and placed in chilled Krebs Ringer solution. A blood sample was sent for lipid profile and cyclosporine level. Cyclosporine dosage was determined by the TDx TDx FLx cyclosporine monoclonal whole blood assay using fluorescence polarization immunoassay technology (Abbott Laboratories, Abbott Park, IL).

In Vitro Experiments Experiments were conducted in 25-mL organ chambers (Fig 1) filled with 37°C oxygenated (95% 02' 5% CO2 ) 0003-4975/94/$7.00

790

CARTIER ET AL CYCLOSPORIN A AND ENDOTHELIAL REACTIVITY

Ann Thorae Surg 1994;58:789 -94

Drugs and Platelet Preparations

Aortic ring on stirrups

Organ chamber

95%°2 5% CO

Fig 1. Organ chamber for study of isolated blood vessels.

Acetylcholine, ADP, 5-HT, HIS, NE, and SNP were obtained from Sigma (St. Louis, MO). Drugs were prepared daily with distilled water except for indomethacin, which was dissolved in Na 2C03 (10-s mol/L). Concentrations always represent the final drug concentration in the 25-mL organ chamber and are expressed in mol/L. Fresh plateletrich solutions were obtained from healthy human volunteers and mixed with anticoagulant citrate solution 00% vol/vol). The solution then was centrifuged twice for 15 minutes at a speed of 1,600 rpm. The platelet-rich plasma was collected and diluted (50% vol/vol) with a modified citrate solution and centrifuged for 20 minutes (1,200 rpm). The platelet pellets were collected and a small sample was used to determine the concentration of the platelets. Platelet concentrations always represent the final concentration in the 25-mL organ chamber and are expressed by platelets per microliter.

Data Analysis Krebs Ringer solution ([mmol/L] NaCl, 118.3; KCl, 4.7; MgS04 , 1.2; KH 2P04 , 1.22; CaCl 2 , 1.3; NaHC03 , 25, and glucose 15; pH 7.4 at 37°C). In randomly selected rings in which vascular smooth muscle function was to be studied in the absence of endothelium, the luminal surface was gently rubbed with a pair of watchmaker's forceps. Vessel rings were suspended between two stainless steel stirrups connected to a force transducer for the measurement of isometric tension. Rings were progressively stretched to their optimal length-tension ratio obtained by contraction at each level of distension with potassium ions (KCl, 20 mmol/L). After this was obtained, vessels were rested for 45 minutes before administration of drugs.

Protocols All rings were precontracted with KCl, 40 mmol/L, which was shown to induce sustained contraction (80% of maximal contraction) of the vessel. Vessels exposed to 5-hydroxy-triptamine (5-HT) were precontracted to KCl, 20 mmol/L to induce a 50% optimal contraction and to allow the rings to relax or contract. Pairs of rings (with and without endothelium) from the same animal were studied in parallel and used for one of the following protocols: PROTOCOL 1. After precontraction, rings were exposed to cumulative doses of acetylcholine (ACh) (10- 9-10- 4 mol/L), histamine (HIS) (10- 9-10- 4 molz l.), and human platelets (25,000-100,000 platelets/ f.LL, final bath concentration). PROTOCOL 2. Rings were precontracted and exposed to cumulative doses of adenosine diphosphate (ADP) (10- 910- 4 mol zL), 5-hydroxytriptamine (5-HT) 00- 9-10- 4 rnol z'L), and norepinephrine (NE) 00-9-1O- s mol z'L), followed by sodium nitroprusside (SNP) (10-9-1O- s mol /L). In all experiments, indomethacin (10-s mol/L) was added to the organ chambers to prevent synthesis of endogenous prostanoids. Each experiment was separated by a 30-minute period during which rings were washed out with KR solution and were allowed to equilibrate.

Results are expressed as mean and standard error of the mean. In all experiments n refers to the number of rats from which rings of aorta were harvested. Relaxation is expressed as the percentage of the maximum contraction obtained with KCl preconstriction (which represents 100%). The negative logarithm of the effective molar dosage of agonists causing 50% relaxation or contraction (ED so) was calculated for all dose-response curves to compare rings' sensitivity to the different drugs used. Rings' maximum tension (t m a x ) were recorded in grams (gm). Statistical analysis of the data was achieved with the Student's t test for either paired or unpaired observations. Statistical significance was considered when p was less than 0.05.

Results Endothelium-Dependent Reactivity Endothelium-dependent relaxation to HIS and ADP were comparable in both treated and control groups (Table 1). However, in the CyA-treated group (Fig 2), acetylcholine-mediated relaxation was significantly decreased as summarized in Tabie 1. Maximal relaxation obtained with ACH (expressed as % of initial precontraction) in CyA-treated aortic rings was 50% ::': 3% compared with 37% ::'::: 7% for control rings (p < 0.05). Acetylcholine RDso was also decreased in the CyA group (CyA, log -6.54 ::'::: 0.05 mol/L versus control, log -6.86 ::'::: 0.03 mol/L, p < 0.0l). ACH, HIS AND ADP.

Rings with and without endothelium displayed similar contraction after administration of 5-HT in the control group (Fig 3). However, a significant endotheliumdependent vasoconstriction was observed with aortic rings of CyA-treated rats stimulated with high dose of 5-HT. Maximal contraction obtained in aortic rings with preserved endothelium was 276% ::'::: 65% of the initial value compared with 168% ::'::: 35% in rings with mechanically denuded endothelium (p < 0.05). These findings suggest an endothelium-dependent constricting mechanism.

5-HT.

Ann Thorac Surg 1994;58:789 -94

CARTIER ET AL CYCLOSPORIN A AND ENDOTHELIAL REACTIVITY

791

Table 1. Maximal Relaxation and ED so in Response to ACH, AD?, HIS, and SNP" Maximal Relaxation (%) Drug ACH ADP HIS SNP ENDO+ ENDO-

ED so (logM)

CyA

Control (Olive Oil)

p Value

50 ± 3 73 ± 4 61 ± 4

36 ± 7 77 ± 3 64 ± 4

<0.05 NS NS

0 0

0 0

Control (Olive Oil)

p Value

-6.54 ± .05 -5.87 ± .09 -5.35 ± .09

-6.86 ± .03 -6.20 ± 0.12 -5.59 ± 0.17

<0.01 NS NS

-8.31 ± 0.10 -8.45 ± 0.10

-8.36 ± 0.10 -8.42 ± 0.14

NS NS

CyA

Endothelium-dependent relaxation was studied on rings with preserved endothelium (ACH, ADP, HIS) and endothelium-independent relaxation on rings with endothelium or on rings with mechanically denudated endothelium.

a

ACH = acetylcholine; ADP = adenosine diphosphate; CyA = cycIosporine A; contraction; ENOO+ = with endothelium; ENDO- = without endothelium; nitroprusside. PLATELETS. Cyclosporine-treated rings (with endothelium) vasoconstricted after being exposed to aggregating platelets, whereas olive oil-treated rings vasodilated slightly at low concentrations before contracting at higher concentrations (Fig 4). However, these changes reached statistical significance only at the lowest concentration (25 X 103 platelets/ j.LL), (CyA, 112% ± 6% versus control, 93% ± 4%; p < 0.05).

ED 5 0 ~ effective molar dose of agonist causing 50% relaxation or HIS = histamine; NS = not significant; SNP = sodium

Endothelium-independent relaxation to sodium nitroprusside remained unchanged in CyA-treated rats and was not influenced by the removal of the endothelium (Table 1).

% OF CONTRACTION 150

Endothelium-Independent Reactivity The dose-response curve for NE in rings with mechanically removed endothelium was enhanced in the CyAtreated group compared with the control group (Fig 5). Maximal tension generated with NE was similar in both groups (CyA, 4.40 ± 0.43 gm versus control, 3.69 ± 0.57 g; p = NS); however, the NE ED so was significantly decreased in the CyA group (CyA, log -7.66 ± 0.18 mol/L and control, log -7.01 ± 0.11 mol/L; p < 0.01), suggesting an increased sensitivity to NE. The dose-response curves of endothelium-preserved vascular rings for NE were comparable in both groups (Table 2).

75 -----

50

--0--

ENDOENDO+

25 Ol+-----.-~-~---~---~--~

-9

-8

-7 -6 CONCENTRATION (log M)

A

-5

-4

% OF CONTRACTION

350

% OF CONTRACTION

300

100 ~==rt===t't

250

80

60

150 100 6---ej---Q--Q--9..r- ~ * P < 0.05

50

----- Olive oil --0-- Cyclosporin

O+---------,--~--r---~~~--,

-9

o+-----.-~-~---~---~--~

-9

ENDOENDO+

200

* p < 0.05

40

20

------0--

-8

-7

-6

-5

-4

CONCENTRATION (log M)

Fig 2. Concentration-response curves for acetylcholine in olive oil(.) and cyclosporine-treated (D) rat aortic segments. Rings were precontracted with KCI (40 mEqjU to produce a contraction of 80% of maximal contraction for that compound. (*Significant difference between rat treated with olive oil and cyclosporine: p < 0.05.)

-8

-7 -6 CONCENTRATION (log M)

-5

-4

B

Fig 3. Concentration-response curves to 5-hydroxytriptamine in olive oil- (A) and cyclosporine-treated (B) rat aortic segments with (ENDO+) and without (ENDO-) endothelium. Rings were precontraeted with KCl (20 mEqjU to produce a contraction of 50% of maximal concentration for that compound. (*Significant difference between rings with endothelium and rings without endothelium: p < 0.05.)

792

CARTIER ET AL CYCLOSPORIN A AND ENDOTHELIAL REACTIVITY

Ann Thorae Surg 1994;58:789-94

% OF CONTRACTION

140 130

_

Oliveoil

----D-

Cyclosporin

120 110 • P < 0.05

100

90 80

+--~--,---~----,--------r---~---,

o

25

50

75

100

PLT (x 1000/ml)

Fig 4. Concentration-response curves to fresh human aggregating platelets in olive oil- (e) and cyclosporine-treated (D) rat aortic segments with endothelium. Rings were precontracted with KCl (40 mEqjL) to produce a contraction of 50% of maximal contraction for that compound. ("Significant difference between rats treated with olive oil or cyclosporine: p < 0.05)

Cyclosporine Level The mean cyclosporine blood level in the treated rats was 302.5 ± 83.7 nmol/L at the termination of the study.

Lipid Profile At the end of the treatment, the lipid profile was comparable in both groups with regard to total cholesterol (control, 2.1 ± 0.2 mmol/L versus CyA, 1.7 ± 0.1 mmol/L; p = NS) and low density lipoproteins (control, 0.84 ± 0.13 versus CyA, 0.61 ± 0.1; p = NS).

Comment This study demonstrates that chronic use of CyA may interfere with local vasoregulation by altering both smooth muscle and endothelial function. Aortic rings of CyAtreated rats studied in organ chamber displayed a significant decrease in the endothelial-dependent relaxation to ACH, whereas it remained unchanged in response to HIS and ADP stimulation. These latter substances initiate vascular smooth muscle relaxation by releasing endothelial synthesized nitric oxide, which stimulates enzyme guanylate cyclase to convert guanosine triphosphate to cyclic guanine monophosphate in the smooth muscle cells [9, 10]. The increased production of cyclic guanine monophosphate reduces the intracellular calcium level, a process that is believed to underlie vasorelaxation [11, 12]. Acetylcholine induces release of endothelium-derived relaxing factor by stimulating muscarinic receptors on endothelial cells [9, 13, 14]. Adenosine diphosphate and HIS release EDRF by interacting with purinoceptors (P2y subclass) and histaminic receptors (Fl-I), respectively [15]. These receptors did not appear to be affected by CyA treatment in our study. The decrease in endotheliumderived relaxing factor release observed in CyA-treated vessels seemed to be related mostly to alteration of muscarinic receptors. Similar findings have been reported by

others, and appear to be dose-dependent [6, 16, 17]. A minimal oral dose of 10 mg/kg per day is required to induce endothelial dysfunction in the rat aorta [16]. However, even at a concentration of 30 mg/kg per day, as-day exposure is needed to obtain significant endothelial damage [17]. The clinical significance of these observations is questionable, because,it is generally accepted that the ACh molecules released by the nerve terminals are rapidly destroyed and are notavailable in significant amounts in circulating blood. Increased sensitivity to NE was observed in the CyAtreated group, as shown by a leftward shift of the doseresponse curve of the agonist and a trend toward a higher generated maximum tension. However, this observation was made only when the endothelium was removed from the aortic segment. Contractile response induced by NE is mediated by alpha-l receptors and depends on intracellular Ca + + mobilization and extracellular Ca + + influx through voltage-dependent Ca + + channels [18-20]. In vitro studies have shown that CyA directly affects the vascular tissue by producing vasoconstriction. In dogs, Cy A causes a potentiation of NE vasoconstriction by inhibition of NE reuptake at peripheral nerve endings, thus contributing to CyA-related systemic hypertension [7]. It has been suggested that the same mechanism affects NE release in central nerve terminal and could account for the increased sympathetic nerve discharges in humans and animals submitted to CyA [11, 21]. Since the rat thoracic aorta is devoid of nerve terminals, the increased sensitivity reported with CyA has to be related to an increased CA ++ influx through the smooth cell membrane [18]. This hypothesis is supported by the fact that the calcium antagonist diltiazem has been reported to partly block CyA's direct vascular effect in dog renal artery, possibly by inhibition of Ca + + influx through membrane voltagedependent channels [1]. CyA-induced NE facilitation could be partially explained by enhanced CA + + influx through slow calcium channels. This hypothesis is sup-

% OF CONTRACTION

100

• P < 0.01

80 60 40 --0-

20

Olive oil Cyclosporin

OO-=~=~---~----~--~

-9

-8

-7 -6 CONCENTRATION (log M)

-5

Fig 5. Concentration-response curves to norepinephrine in olive oil(e) and cyclosporine-treated (D) rat aortic segments without endothelium. ("Significant difference between rats treated with olive oil or cyclosporine: p < 0.05)

793

CARTIER ET AL CYCLOSPORIN A AND ENDOTHELIAL REACTIVITY

Ann Thorac Surg 1994;58:789-94

Table 2. Maximum Tension and ED so to norepinephrine in Cyclosporin A-Treated and Olive Oil-Treated Rings With and Without Endothelium EOso OogM)

Maximum Tension (g) Condition ENOO+ ENOOCyA

=

cyclosporin A;

CyA

Control (Olive Oil)

p Value

CyA

Control (Olive Oil)

p Value

3.70 :!: 0.35 4.40 :!: 0.43

2.71 :!: 0.72 3.69 :!: 0.57

NS NS

-6.97:!: 0.10 7.66 :!: 0.18

-7.05 :!: 0.12 -7.01 :!: 0.11

NS <0.01

ENDO+

=

with endothelium;

ENDO- = without endothelium;

ported by clinical observations showing that renal toxicity and hypertension often seen in transplanted patients immunosuppressed with CyA are significantly reduced with the use of calcium-channel blockers [22-24]. Furthermore, there is clinical evidence that calcium-channel blockers such as diltiazem prevents the usual reduction in the diameter of coronary arteries normally seen in cardiac transplant recipients during the first year. Even though the mechanism underlying this process is not well understood, it has been suggested that the drug inhibits an early, immunologically mediated, diffuse vascular insult associated with the graft implantation [25]. However, interactions with CyA effects cannot be excluded. The presence of endothelium partially abolished the cyclosporine NE facilitation in our study. a-Adrenergic agonist responses are modulated by EDRF in the rat aorta [26]. This can account for the attenuation of CyA-increased sensitivity to NE seen in rings with preserved endothelium in our study. Preservation of endothelium attenuated the cyclosporine effect on vascular smooth muscle local tone; this outlines the importance of preserving coronary artery endothelium against ischemic or chemical injury during the different steps of heart transplantation such as graft harvesting and storage in preservation solutions. Hyperkalemic cardioplegic solution used to induce electromechanical arrest can significantly affect the coronary endothelial preservation and promote CyA-induced toxicity [27]. In the CyA-treated aortic segments, 5-HT stimulation resulted in an endothelium-dependent vasoconstriction especially with the use of high concentrations of the drug (10- 510- 4 mol/L), whereas no significant difference in constriction was seen in the control group between aortic rings with and without endothelium. 5-Hydroxytriptamine, one of the major vasoactive compounds released by aggregating platelets, may contribute to the vasoconstriction facilitation observed among CyA-treated vessels exposed to aggregating platelets in our study. Cyclosporine has been shown to enhance platelet adhesion both in vivo and in vitro by decreasing the amount of prostacyclin produced by the endothelial cells and promoting the release of thromboxane A z [9]. The mechanism through which 5-HT releases endothelium-dependent contracting factors in our experiment remains to be investigated. However, 5-HT has been reported to promote the release of superoxide anion in certain experimental pathologic endothelial conditions, such as regenerated rat aortic endothelium after mechanical endothelial denudation [28, 29]. Cyclosporine has been shown to induce synthesis by cultured human

NS

=

not significant.

endothelial cells [30] of endothelin, a potent endothelialdependent vasoconstrictor [31]. In vivo, CyA infusion increases renal release of endothelin [19]. However, because endothelial cells cannot store endothelin, it is unlikely that the latter could explain acute 5-HT-dependent vasoconstriction as reported in our model [32]. Nevertheless, this 5-HTinduced vasoconstriction can promote platelet aggregation and thrombus formation during the early posttransplantation period when the endothelium is still recovering from ischemic and reperfusion injury. In conclusion, our study shows that chronic use of CyA affects the vascular reactivity of the rat aorta mainly by decreasing nitric oxide release related to endothelial muscarinic receptors, increasing smooth muscle sensitivity to circulating catecholamines such as NE, and releasing an endothelial-dependent serotoninergic-regulated vasoconstrictor agent. These observations may provide a potential explanation for the increased incidence of systemic hypertension, nephrotoxicity and thrombotic disease reported in transplanted patients treated with CyA as well as for premature coronary arteriosclerosis observed in heart transplant recipients [33, 34].

References 1. Carrier M, Trone F, Stewart 0, Pelletier LC. Dose-dependent effect of cyclosporin on renal arterial resistance in dogs. Am J PhysioI1991;261:H1791-6. 2. Curtis JJ, Luke RG, Oubovsky E, Diethelm AG, Whelchel JO, Jones P. Cyclosporin in therapeutic doses increases renal allograft vascular resistance. Lancet 1986;2:477-9. 3. Weir MR, Klassen OK, Shen SY, Sullivan 0, Buddemeyer EU, Handwerger BS. Acute effects of intravenous cyclosporine on blood pressure, renal hemodynamics, and urine prostaglandin production of healthy humans. Transplantation 1990;49:41-7. 4. Kon V, Sugiura M, Inagami T, Harvie BR,Ichikawa I, Hoover RL. Role of endothelin in cyclosporine-induced glomerular dysfunction. Kidney Int 1990;37:1487-91. 5. Paller MS. Effects of the prostaglandin E] analog misoprostol on cyclosporine nephrotoxicity. Transplantation 1988;45: 1126-31. 6. Bossaller C, Forstermann U, Hertel R, Olbricht C, Reschke V, Fleck E. Cyclosporin A inhibits endothelium-dependent vasodilatation and vascular prostacyclin production. Eur J Pharmacal 1989;165:165-9. 7. Trone F, Carrier M, Pelletier CL. Mechanism of hind limb vasoconstriction due to cyclosporin A in the dog. Circ Res 1992;71:1159-64. 8. Scherrer U, Vissing SF,Morgan BJ, et al. Cyclosporine-induced sympathetic activation and hypertension after heart transplantation. N Engl J Med 1990;323:693-9. 9. Furchgott RF, Zawadzki JV. The obligatory role of endothelial

794

10. 11.

12.

13.

14. 15. 16. 17.

18. 19.

20. 21.

CARTIER ET AL CYCLOSPORIN A AND ENDOTHELIAL REACTIVITY

cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288:373-6. Valerie BS. Nitric oxide and vascular reactivity. Coronary Artery Dis 1991;2:293-9. Rapoport RM, Draznin MB, Murad F. Endotheliumdependent relaxation in rat aorta may be mediated through cyclic GMP-dependent protein phosphorylation. Nature 1983; 306:174-6. Gruetter CA, Gruetter DY, Lyon JE, Kadowitz PJ, Ignarro LJ. Relationship between cyclic guanosine 3':5'-monophosphate formation and relaxation of coronary arterial smooth muscle by glyceryl trinitrate, nitroprusside, nitrite and nitric oxide: effects of methylene blue and methemoglobin. J Pharmacal Exp Ther 1981;219:181-6. Angus JA, Campbell GR, Cocks TM, Manderson JA. Vasodilatation by acetylcholine is endothelium-dependent: a study by sonomicrometry in canine femoral artery in vivo. J Physiol 1983;344:209-22. Komori K, Suzuki H. Heterogeneous distribution of muscarinic receptors in the rabbit saphenous artery. Br J Pharmacol 1987;92:657-64. Searle NR, Sahab P. Endothelial vasomotor regulation in health and disease. Can J Anaesth 1992;39:838-57. Rego A, Vargas R, Foegh ML, Ramwell PW. Effect of cyclosporine A treatment on vascular reactivity of the rat thoracic aorta. Transplant Proc 1988;20(Suppl 3):572-7. Gerkens JF. Cyclosporine treatment of normal rats produces a rise in blood pressure and decreased renal vascular responses to nerve stimulation, vasoconstrictors and endotheliumdependent dilators. J Pharmacal Exp Ther 1989;250:1105-12. Hogestatt ED. Characterization of two different calcium entry pathways in small mesenteric arteries from rat. Acta Physiol Scand 1984;122:483-95. Loutzenhiser R, Epstein M. Activation mechanisms of human renal artery: effects of KCl, nore,Rinephrine and nitrendipine upon tension development and sCa influx. Eur J Pharmacol 1985;106:47-52. Godfraind T, Miller RC, Lima JS. Selective Ct,- and Ct2adrenoceptor agonist-induced contractions and 4SCa fluxes in the rat isolated aorta. Br J Pharmacol 1982;77:597-604. Morgan BI, Lyson T, Scherrer U, Victor RG. Cyclosporine causes sympathetically mediated elevations in arterial pressure in rats. Hypertension 1991;18:458-66.

Ann Thorae Surg

1994;58:789-94

22. Dawidson I, Rooth P, Fry WR, et al. Prevention of acute cyclosporine-induced renal blood flow inhibition and improved immunosuppression with verapamil. Transplantation 1989;48:575-80. 23. Kiberd BA. Cyclosporine-induced renal dysfunction in human renal allograft recipients. Transplantation 1989;48:965-9. 24. Kirk AJB, Omar I, Bateman DN, Dark JH. Cyclosporineassociated hypertension in cardiopulmonary transplantation. Transplantation 1989;48:428-30. 25. Schroeder JS, Gao SZ, Alderman EL, et al. A preliminary study of diltiazem in the prevention of coronary artery disease in heart-transplant recipients. N Engl J Med 1993;328:164-70. 26. Carrier GO, White RE. Enhancement of alpha-I and alpha-2 adrenergic agonist-induced vasoconstriction by removal of endothelium in rat aorta. J Pharmacal Exp Ther 1985;232: 682-7. 27. Cartier R, Pellerin M, Hollmann C, Pelletier LC. Effects of pressure and duration of hyperkalemic infusions on endothelial function. Ann Thorac Surg 1993;55:700-5. 28. Cartier R, Pearson PI, Lin PJ, Schaff HV. Time caurse and extent of recovery of endothelium-dependent contractions and relaxations after direct arterial injury. J Thorac Cardiovasc Surg 1991;102:371-7. 29. Lin PJ, Pearson PI, Cartier R, Schaff HV. Superoxide anion mediates the endothelium-dependent contractions to serotonin by regenerated endothelium. J Thorac Cardiovasc Surg 1991;102:378-85. 30. Bunchman TE, Brookshire CA. Cyclosporine-induced synthesis of endothelin by cultured human endothelial cells. J Clin Invest 1991;88:310-4. 31. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 1988;332:411-5. 32. Vanhoutte PM. Other endothelium-derived vasoactive factors. Circulation 1993;87(SuppI5):9-17. 33. Vanrenterghem Y, Roels L, Lerut T, et al. Thromboembolic complications and haemostatic changes in cyclosporin-treated cadaveric kidney allograft recipients. Lancet 1985;1:999-1002. 34. Carrier M, Pelletier G, Leclerc Y, et al. Accelerated coronary atherosclerosis after cardiac transplantation: major threat to long-term survival. Can J Surg 1991;34:133-6.