(Circulation. 1997;96:3042-3047.)
© 1997 American Heart Association, Inc.
Articles |
From the Centre for Clinical Pharmacology, Cruciform Project for Strategic Medical Research and Department of Medicine, University College London (UK).
Correspondence to Kiran Bhagat, BSc, PhD, MRCP, Centre for Clinical Pharmacology, University College London, London, UK. E-mail k.bhagat{at}ucl.ac.uk
| Abstract |
|---|
|
|
|---|
Methods and Results Subjects lay with one hand placed on an
angled support. The diameter of a vein was recorded by measuring
the linear displacement of a probe placed on the skin overlying the
vein when the pressure in a congesting cuff placed around the upper arm
was deflated from 40 to 0 mm Hg. A length of the vein was
isolated by two wedges. TNF-
(1 ng), IL-1ß (1 ng), or IL-6 (100
pg) were instilled for 1 hour, either individually or together. At the
end of the hour, the wedges were removed and the vein reconnected with
the circulation. Dose-response curves (bradykinin: 2, 4, and 8
pmol/min; arachidonic acid: 0.2, 2, and 20 nmol/min;
and glyceryl trinitrate 1, 2, and 4 pmol/min) were constructed before
and 1, 6, 24, and 48 hours after instillation. In another study,
hydrocortisone (100 mg) was given 2 hours before the study. In a
different study, subjects were given oral aspirin (75 mg or 1 g) 2
hours before the study. TNF-
and IL-1ß alone but not IL-6
attenuated the dilatation to bradykinin and arachidonic
acid; the response was greatest at 1 hour with recovery occurring by 6
hours. Combination of IL-1ß and TNF-
prolonged the
endothelial dysfunction, resulting in recovery at 24
hours. Hydrocortisone and high-dose aspirin prevented
endothelial dysfunction.
Conclusions The results demonstrate that pro-inflammatory cytokines induce transient and reversible endothelial dysfunction and indicate that cyclooxygenase activity may contribute to the genesis of the effect. If other vessels behave similarly, this may provide further insight into the mechanisms precipitating acute cardiovascular events after inflammatory disorders.
Key Words: endothelium vasodilation bradykinin interleukins prostaglandins
| Introduction |
|---|
|
|
|---|
Consistent with the link between endothelial dysfunction and cardiovascular disease, smoking,4 hypertension,5 diabetes,6 and hyperlipidemia7 have all been associated with impaired endothelium-dependent dilatation in studies in animals and humans. However, these are chronic and often relatively stable risk factors for cardiovascular disease; recent epidemiological studies have recognized another more acute risk factor for unstable angina, myocardial infarction, and stroke: that of infection.11 A number of reports have indicated that a preceding febrile illness or a bacteremic episode is associated with a markedly increased risk of an acute cardiovascular event for several weeks after the illness.12 13 In a previous study in healthy volunteers, we demonstrated that a brief exposure to bacterial endotoxin impairs endothelium-dependent relaxation for many days14 and suggested that transient "stunning" of endothelial function might provide a mechanism linking infection to increased risk of infarction.
Cytokines mediate many of the biological effects of
endotoxin,15 and local concentrations of certain
pro-inflammatory cytokines are significantly elevated in
patients with unstable angina and myocardial
infarction.16 17 18 Indeed, it has been suggested that an
inflammatory response associated with cytokine production might trigger
the transition from stable to unstable
atheroma.8 In the present study we report
on the effects of TNF-
, IL-1ß, and IL-6 on
endothelial function in healthy volunteers and assess
the effects of anti-inflammatory drugs on the responses seen.
| Methods |
|---|
|
|
|---|
![]() |
Instillation of Cytokines
To instill cytokines, a length of the vein under study
was isolated from the circulation by means of two wedges placed 2 to 3
cm apart on the skin overlying the vessel as described
previously.19 TNF-
(1 ng in 1 mL saline), IL-1ß (1 ng
in 1 mL saline), and IL-6 (100 pg in 1 mL saline) were instilled for 1
hour, either individually or together. At the end of the period of
instillation, the wedges were removed and the vein was reconnected with
the circulation for assessment of reactivity. This method of
instillation produces local changes in the study vein but adjacent
vessels remain unaffected.19 The volume of blood in the
isolated vein is in the order of 1 to 2 mL, and the calculated
concentration of cytokine was in the order of 300 to 1000
pg/mL (TNF-
and IL-1ß) and 30 to 100 pg/mL
(IL-6).
Dilator Dose-Response Curves
Vessels were preconstricted to approximately 50% of resting
diameter by a continuous infusion of norepinephrine (5 to
640 pmol/min). To construct cumulative dose-response curves,
bradykinin (2, 4, and 8 pmol/min, each dose for 5 minutes),
arachidonic acid (0.2, 2, and 20 nmol/min, each
dose for 5 minutes), or GTN (1, 2, and 4 pmol/min, each dose for
5 minutes) was co-infused with the norepinephrine and
relaxation recorded. In these vessels, bradykinin activates
nitric oxide synthase in the
endothelium,21 22
arachidonic acid stimulates prostanoid
production,19 and GTN is a nitric oxide donor that
acts directly on smooth muscle. In all studies, a 10 to 15-minute
washout period (infusion of norepinephrine alone) separated
the dose response to different agonists.
Effects of Cytokines
In 40 subjects dilator dose-response curves were constructed
before and at 1, 6, 24, and 48 hours after instillation of either
TNF-
alone, IL-1ß alone, IL-6 alone, TNF-
and IL-1ß together,
or TNF-
, IL-1ß, and IL-6 together. To construct a time course,
each individual was studied on several occasions. To ensure that the
same site of the study vein was assessed at different time points, an
indelible mark was placed at the site of the needle insertion and the
point at which measurements were taken.
Effects of Anti-inflammatory Drugs on the Response to
Cytokines
Hydrocortisone
Subjects were given oral hydrocortisone (100 mg) 2 hours before
the study, and dilator dose-response curves were constructed before and
1 hour after instillation of TNF-
alone (n=5), IL-1ß alone (n=5),
or a combination of TNF-
, IL-1ß, and IL-6 (n=5).
Aspirin
Subjects were given oral aspirin (75 mg or 1 g) 2 hours
before the study and dilator dose-response curves were constructed
before and 1 hour after instillation of TNF-
alone (n=5; highest
dose of aspirin only) or a combination of TNF-
, IL-1ß, and IL-6
(n=5; each subject studied twice, once with high-dose and once with
low-dose aspirin).
Drugs
IL-1ß (10 µg), TNF-
(10 µg), and IL-6 (10 µg) were
obtained from Bachem, bradykinin from Clinalpha AG, hydrocortisone (20
mg/tablet) from MSD, norepinephrine (2
mg/vial) from Sanofi Winthrop, ascorbic acid (100 mg/mL)
from Evans Medical Ltd, and heparin (100 U/mL) from CP Pharmaceuticals
Ltd. Sodium arachidonate (5 mg/vial) stored under
nitrogen was obtained from Sigma. Vials were stored at -20°C and a
single vial was used for each study. Sodium arachidonate (5
mg) was dissolved in 154 µL absolute alcohol to produce a stock
solution of 1 mmol/L. Subsequent dilution was in saline,
and the final concentration of alcohol in the infusate was 0.0001%.
Arachidonic acid always was used within 1 hour of
preparation. Ascorbic acid was added to norepinephrine
stock solutions to prevent auto-oxidation. Heparin (100 U/mL) was added
to the cytokine solution before administration to prevent thrombus
formation. All solutions of cytokines and bradykinin were
filtered through a 0.2-µm bacterial filter (Acrodisc PF, Gelman
Science).
Calculations and Statistics
Changes in vein size were measured in arbitrary units and
converted to millimeters after calibration of the transducer at the end
of each experiment. The response of the resting vein to drugs is
expressed as a reduction in diameter from that measured during infusion
of saline alone. The response of the
norepinephrine-preconstricted vein to drugs is expressed as
percentage reversal of the induced constriction as described
above.14 Results are compared using Student's
t test for paired data or ANOVA of the means as appropriate;
a value of P<.05 is considered statistically
significant.
| Results |
|---|
|
|
|---|
alone nor IL-6 alone affected the constrictor
response to norepinephrine. However, IL-1ß alone and
combinations of IL-1ß with the other cytokines decreased the
constrictor responses; to maintain the same degree of preconstriction,
the norepinephrine dose was increased as necessary
(Table
|
Effects of Cytokines on Dilator Response
TNF-
and IL-1ß attenuated the dilatation to bradykinin and
arachidonic acid (Figs 1
and 2
). IL-6 was without effect. The
duration of the effect of TNF-
alone or IL-1ß alone was
short-lived, and the dilatation to bradykinin and
arachidonic acid returned to normal by 6 hours.
Combining TNF-
, IL-1ß, and IL-6 did not increase the magnitude of
the effect on endothelium-dependent dilatation but
increased the duration (Fig 3
). When all
three cytokines were instilled together for 1 hour, the
impairment of endothelium-dependent dilatation
persisted for at least 24 hours (Fig 3
). In contrast to the effects of
cytokines on endothelium-dependent dilatation,
the response to GTN was unaltered by instillation of cytokines
(Fig 4
).
|
|
|
|
Effects of Anti-inflammatory Drugs on the Response to
Cytokines
Hydrocortisone
Prior administration of oral hydrocortisone prevented the effects
of cytokines on dilatation to bradykinin and
arachidonic acid (Fig 5
).
For example, in the hydrocortisone pretreatment group the dilatation to
bradykinin (2, 4, and 8 pmol/min) was 19±5%, 43±4%, and
81±9% before, and 21±5%, 64±11%, and 89±7% 1 hour after
instillation of TNF
, IL-1ß, and IL-6. Similarly, the dilatation
to arachidonic acid before instillation of the
cytokines was 19±6%, 42±10%, and 81±9%, and 1 hour after
instillation of TNF-
, IL-1ß, and IL-6 was 81±4%, 50±6%, and
94±4%.
|
Aspirin
Prior administration of high-dose aspirin (1 g) significantly
attenuated (P<.05) the inhibitory effects of
either TNF
alone or the combination of TNF-
, IL-1ß, and IL-6
(Fig 6A
and 6B
). However, pretreatment
with low-dose aspirin was without effect (Fig 6C
).
|
| Discussion |
|---|
|
|
|---|
In our earlier study we demonstrated that a brief exposure to endotoxin
impairs endothelium-dependent relaxation for several
days.14 This effect also was seen when human saphenous
vein was incubated with endotoxin in vitro and was not due to
structural damage to the endothelial
layer.14 Similarly, it has been reported that acute
exposure to endotoxin impairs the production of nitric oxide in
response to bradykinin in bovine endothelial cells in
culture.27 In the present study, we have explored the
effects of three cytokines that have been implicated in
mediating inflammatory response and that are found in elevated local
concentrations in the plasma of patients with acute myocardial
infarction or unstable angina.16 17 18 25 26 In our vein
model,19 IL-1ß and TNF-
impaired the relaxant
responses to bradykinin (a mediator that acts through stimulation of
nitric oxide production in these vessels21 22 ) and
to arachidonic acid (the precursor of prostanoid
synthesis28 ). The effect was specific for
endothelium-dependent dilators, and the response to the
nitric oxide donor GTN was unaltered. IL-6 had no effect on its own,
but the combination of all three cytokines produced the most
long-lasting impairment of endothelium-dependent
relaxation, with recovery occurring only by 48 hours. The manouver of
isolating and wedging the vein for 1 hour in the absence of
cytokines does not by itself affect responses to
norepinephrine, bradykinin, or arachidonic
acid.19 29
We took care to ensure that the degree of preconstriction was identical
for each study (Table
). IL-1ß induced hyporesponsiveness to
norepinephrine have described previously,29
and we needed to use more norepinephrine to produce the
same degree of constriction after instillation of this cytokine.
However, it is unlikely that the impaired relaxation we observed was
due to differences in the amount of norepinephrine required
to preconstrict the vessels for the different parts of the study, since
TNF-
alone did not alter the constrictor response to
norepinephrine but was the most effective cytokine
at producing selective impairment of
endothelium-dependent relaxation. Furthermore,
high-dose aspirin abolished the endothelial stunning
induced by the combination of TNF-
, IL-1ß, and IL-6 without
affecting the hyporesponsiveness to norepinephrine. The
transient hyporesponsiveness to norepinephrine produced by
IL-1ß is due to induction of basal nitric oxide generation and can be
reversed by nitric oxide synthase
inhibitors.29 30
Prior administration of hydrocortisone prevented the inhibitory actions of the cytokines on endothelial dilator function, suggesting that expression of inflammatory enzymes and generation of local inflammatory mediators were responsible for the effects seen. An anti-inflammatory dose (1 g) of aspirin significantly reduced the effects of cytokines on endothelial function, whereas a low cardioprotective dose (75 mg) of aspirin had no effect. We have previously shown that this anti-inflammatory dose of aspirin (1 g) abolishes vascular prostanoid synthesis in the hand veins, whereas a low dose of 75 mg of aspirin abolishes arachidonic acid-induced platelet aggregation without affecting arachidonic acid-induced venodilatation.28 Although it is possible that hydrocortisone produced some mineralocorticoid effects and that aspirin induced some degree of acute sodium reabsorption, the most likely explanation is that these agents exerted anti-inflammatory effects on the vessel wall and that the endothelial stunning induced by cytokines depends on local prostanoid synthesis. One possibility is that inducible cyclooxygenase (COX)-II was expressed in response to cytokines, and that the activity of this enzyme contributes to the endothelial stunning. It seems unlikely that COX-I activity in the endothelium was responsible, since the usual dilatation to arachidonic acid28 was not reduced after cytokine treatment.
TNF-
appears to be a key cytokine inducing
endothelial dysfunction. Administration of TNF-
depresses endothelium-dependent relaxation in
vivo,31 and in vitro TNF-
reduces the half-life of mRNA
coding for nitric oxide synthase.32 In addition, in
patients with heart failure, significantly elevated levels of TNF-
have been documented33 and, in experimental heart failure,
reduced gene expression of endothelial nitric oxide
synthase and COX-I activity has been reported.34 It is not
known whether COX-II activity contributes to these effects of TNF-
;
however, generation of free radicals as a byproduct of COX
activity35 might affect endothelial
function. Studies in animals have shown that the
endothelial dilator dysfunction that occurs during
endotoxemia is significantly restored in the presence of free radical
scavengers.36 Although it is possible that induction of
nitric oxide generation contributed to the effects produced by IL-1ß,
it is unlikely that increased synthesis of nitric oxide due to
expression of the inducible isoform of nitric oxide synthase made a
major contribution to the effects we observed, since TNF-
alone does
not induce nitric oxide synthesis in this model.29 Further
studies will be required to explore these mechanisms and to understand
why such a brief (1 hour) exposure to cytokines causes such
prolonged (lasting at least 24 hours) stunning of normal
endothelial dilator activity.
The model we have developed allows the safe study of the effects of cytokines on a human blood vessel in situ. The cytokines were instilled in doses sufficient to produce a local concentration similar to those found in patients with certain infections,37 acute myocardial infarction, and unstable angina.16 17 18 25 26 The effects are generated locally within the blood vessel, and our previous studies have indicated that even an adjacent vessel on the same hand is unaffected by agents instilled into the isolated segment. Of course, we do not know whether the observations made in the superficial hand vein are relevant to what might happen in the coronary or carotid artery or other clinically important vessels, and it would be important to extend these studies to determine whether arterial endothelium is also affected by exposure to cytokines. However, the pharmacology and physiology of the hand veins are similar to those of the saphenous vein,22 38 39 40 a vessel widely used for bypass grafting,41 and we studied the vessels in situ in the usual physiological environment.
What are the potential clinical implications of our study? The
incidence of acute myocardial events or stroke appears to rise
significantly after a febrile or bacteremic
illness.11 12 13 We have reported previously that bacterial
endotoxin impairs endothelial function for several
days.19 We now show that certain cytokines that
might be generated in a much wider range of infectious or inflammatory
conditions, and which have been implicated in the pathogenesis of acute
cardiovascular events,16 17 18 25 26 also
impair endothelium-dependent dilatation in healthy
volunteers in vivo. TNF-
appears to be the critical cytokine
in this process. The impairment is not only confined to the
L-arginine-nitric oxide pathway but also affects dilator
prostanoid production. If other human vessels are affected by
these inflammatory cytokines in the same way as the hand veins
are, our results demonstrate a mechanism by which an acute inflammatory
response (for example, due to infection or surgery) might be linked to
transiently increased risk of acute cardiovascular
events including deep vein thrombosis. It remains to be determined
whether the protective effects of aspirin we observed contribute to the
efficacy of this drug in unstable angina or acute myocardial
infarction.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received April 4, 1997; revision received May 22, 1997; accepted May 28, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. K.-T. Li, B. C.-H. Kwan, C. C. Szeto, and G. T.-C. Ko Metabolic syndrome in peritoneal dialysis patients NDT Plus, June 27, 2008; (2008) sfn073v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Higashi, C. Goto, D. Jitsuiki, T. Umemura, K. Nishioka, T. Hidaka, H. Takemoto, S. Nakamura, J. Soga, K. Chayama, et al. Periodontal Infection Is Associated With Endothelial Dysfunction in Healthy Subjects and Hypertensive Patients Hypertension, February 1, 2008; 51(2): 446 - 453. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Vlachopoulos, K. Aznaouridis, A. Dagre, C. Vasiliadou, C. Masoura, E. Stefanadi, J. Skoumas, C. Pitsavos, and C. Stefanadis Protective effect of atorvastatin on acute systemic inflammation-induced endothelial dysfunction in hypercholesterolaemic subjects Eur. Heart J., September 1, 2007; 28(17): 2102 - 2109. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. C.H. Kwan, M. A. Murtaugh, and S. Beddhu Associations of Body Size with Metabolic Syndrome and Mortality in Moderate Chronic Kidney Disease Clin. J. Am. Soc. Nephrol., September 1, 2007; 2(5): 992 - 998. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. H. Shen, L. Zhang, Y. Gan, X. Wang, J. Wang, S. A. LeMaire, J. S. Coselli, and X. L. Wang Up-regulation of PTEN (Phosphatase and Tensin Homolog Deleted on Chromosome Ten) Mediates p38 MAPK Stress Signal-induced Inhibition of Insulin Signaling: A CROSS-TALK BETWEEN STRESS SIGNALING AND INSULIN SIGNALING IN RESISTIN-TREATED HUMAN ENDOTHELIAL CELLS J. Biol. Chem., March 24, 2006; 281(12): 7727 - 7736. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Gulbins and P. L. Li Physiological and pathophysiological aspects of ceramide Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2006; 290(1): R11 - R26. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Vlachopoulos, I. Dima, K. Aznaouridis, C. Vasiliadou, N. Ioakeimidis, C. Aggeli, M. Toutouza, and C. Stefanadis Acute Systemic Inflammation Increases Arterial Stiffness and Decreases Wave Reflections in Healthy Individuals Circulation, October 4, 2005; 112(14): 2193 - 2200. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Bruunsgaard Physical activity and modulation of systemic low-level inflammation J. Leukoc. Biol., October 1, 2005; 78(4): 819 - 835. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Mittermayer, J. Pleiner, G. Schaller, S. Zorn, K. Namiranian, S. Kapiotis, G. Bartel, M. Wolfrum, M. Brugel, J. Thiery, et al. Tetrahydrobiopterin corrects Escherichia coli endotoxin-induced endothelial dysfunction Am J Physiol Heart Circ Physiol, October 1, 2005; 289(4): H1752 - H1757. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. D. Buckley, G. E. Rainger, G. B. Nash, and K. Raza Endothelial cells, fibroblasts and vasculitis Rheumatology, July 1, 2005; 44(7): 860 - 863. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. B. Tripathi, B. K. Singh, R. S. Pandey, and M. Kumar BHUx: A Patent Polyherbal Formulation to Prevent Atherosclerosis Evid. Based Complement. Altern. Med., June 1, 2005; 2(2): 217 - 221. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. H. Berg and P. E. Scherer Adipose Tissue, Inflammation, and Cardiovascular Disease Circ. Res., May 13, 2005; 96(9): 939 - 949. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. C. W. Lau, B. Dhillon, H. Yan, P. E. Szmitko, and S. Verma Adipokines: molecular links between obesity and atheroslcerosis Am J Physiol Heart Circ Physiol, May 1, 2005; 288(5): H2031 - H2041. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Mozaffarian, E. B Rimm, I. B King, R. L Lawler, G. B McDonald, and W. C Levy trans Fatty acids and systemic inflammation in heart failure Am. J. Clinical Nutrition, December 1, 2004; 80(6): 1521 - 1525. [Abstract] [Full Text] [PDF] |
||||
![]() |
A M A Shehab, R J MacFadyen, M McLaren, R Tavendale, J J F Belch, and A D Struthers Sudden unexpected death in heart failure may be preceded by short term, intraindividual increases in inflammation and in autonomic dysfunction: a pilot study Heart, November 1, 2004; 90(11): 1263 - 1268. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Fichtlscherer, S. Breuer, and A. M. Zeiher Prognostic Value of Systemic Endothelial Dysfunction in Patients With Acute Coronary Syndromes: Further Evidence for the Existence of the "Vulnerable" Patient Circulation, October 5, 2004; 110(14): 1926 - 1932. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. R. Clapp, A. D. Hingorani, R. K. Kharbanda, V. Mohamed-Ali, J. W. Stephens, P. Vallance, and R. J. MacAllister Inflammation-induced endothelial dysfunction involves reduced nitric oxide bioavailability and increased oxidant stress Cardiovasc Res, October 1, 2004; 64(1): 172 - 178. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Prasad and A. A. Quyyumi Renin-Angiotensin System and Angiotensin Receptor Blockers in the Metabolic Syndrome Circulation, September 14, 2004; 110(11): 1507 - 1512. [Full Text] [PDF] |
||||
![]() |
M. J. Quinones, M. Hernandez-Pampaloni, H. Schelbert, I. Bulnes-Enriquez, X. Jimenez, G. Hernandez, R. De La Rosa, Y. Chon, H. Yang, S. B. Nicholas, et al. Coronary Vasomotor Abnormalities in Insulin-Resistant Individuals Ann Intern Med, May 4, 2004; 140(9): 700 - 708. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Mozaffarian, T. Pischon, S. E Hankinson, N. Rifai, K. Joshipura, W. C Willett, and E. B Rimm Dietary intake of trans fatty acids and systemic inflammation in women Am. J. Clinical Nutrition, April 1, 2004; 79(4): 606 - 612. [Abstract] [Full Text] [PDF] |
||||
![]() |
L S Rallidis, M G Zolindaki, P C Pentzeridis, K P Poulopoulos, A H Velissaridou, and T S Apostolou Raised concentrations of macrophage colony stimulating factor in severe unstable angina beyond the acute phase are strongly predictive of long term outcome Heart, January 1, 2004; 90(1): 25 - 29. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Esposito, C. Di Palo, R. Marfella, and D. Giugliano The Effect of Weight Loss on Endothelial Functions in Obesity: Response to Sciacqua et al. Diabetes Care, October 1, 2003; 26(10): 2968 - 2969. [Full Text] |