Am J Physiol Regulatory Integrative Comp Physiol
281: R1097–R1104, 2001.
Effect of nicotine on vasoconstrictor and vasodilator
responses in human skin vasculature
CLAIRE E. BLACK,1,2 NING HUANG,1 PETER C. NELIGAN,1,2 RONALD H. LEVINE,1,2
JOAN E. LIPA,1,2 STEVEN LINTLOP,3 CHRISTOPHER R. FORREST,1,2 AND CHO Y. PANG1,2,4
1
Research Institute, The Hospital for Sick Children, and Departments of 2Surgery and 4Physiology,
University of Toronto, and 3Center of Forensic Sciences, Toronto, Ontario M5G 1X8, Canada
Received 14 December 2000; accepted in final form 4 June 2001
Black, Claire E., Ning Huang, Peter C. Neligan, Ro-
nald H. Levine, Joan E. Lipa, Steven Lintlop, Christo-
pher R. Forrest, and Cho Y. Pang. Effect of nicotine on
vasoconstrictor and vasodilator responses in human skin
vasculature. Am J Physiol Regulatory Integrative Comp
Physiol 281: R1097–R1104, 2001.—Our objective was to test
the hypothesis that acute exposure of human skin vascula-
ture to nicotine may have deleterious effects on endothelial
function. Vasoconstriction and vasorelaxation in isolated per-
fused human skin flaps (⬃8 ⫻ 18 cm) derived from dermoli-
pectomy specimens were assessed by studying changes in
skin perfusion pressure measured by a pressure transducer,
and skin perfusion was assessed by a dermofluorometry
technique (n ⫽ 4 or 5). It was observed that nicotine (10⫺7 M)
amplified (P ⬍ 0.05) the norepinephrine (NE)-induced con-
centration-dependent (10⫺7-10⫺5 M) increase in skin vaso-
constriction compared with the control. This amplification
effect of nicotine in NE-induced skin vasoconstriction was not
blocked by the nicotine-receptor antagonist hexamethonium
(10⫺6 M) or the cyclooxygenase inhibitor indomethacin (10⫺5
M). It was also observed that ACh and nitroglycerin (NTG)
elicited a concentration-dependent (10⫺8-10⫺5 M) vasorelax-
ation in skin flaps preconstricted with 8 ⫻ 10⫺7 M of NE. The
vasorelaxation induced by ACh was attenuated (P ⬍ 0.05) in
the presence of nicotine (10⫺7 M) compared with the control.
However, skin vasorelaxation induced by NTG was not af-
fected by nicotine (10⫺7 M). ACh and NTG are known to
induce endothelium-dependent and -independent vasorelax-
ation, respectively. The present findings were interpreted to
indicate that acute exposure of human skin vasculature to
nicotine was associated with 1) amplification of NE-induced
skin vasoconstriction and 2) impairment of endothelium-
dependent skin vasorelaxation. Cyclooxygenase products and
nicotine receptors blocked by hexamethonium were not in-
volved in the amplification of NE-induced skin vasoconstric-
tion by nicotine. These findings may provide further insight
into the pathogenesis of skin vasospasm in skin flap surgery
and skin ischemic disease associated with cigarette smoking
or use of smokeless tobacco.
norepinephrine; acetylcholine; nitroglycerin
THE ASSOCIATION OF CIGARETTE smoking or use of smoke-
less tobacco (e.g., chewing tobacco) with arterioscle-
rotic peripheral vascular disease and thromboangiitis
obliterans (Buerger’s disease) has been identified for
Address for reprint requests and other correspondence: C. Y. Pang,
The Hospital for Sick Children, 555 Univ. Ave., Toronto, Ontario
M5G 1X8, Canada (E-mail: pang@sickkids.on.ca).
http://www.ajpregu.org 0363-6119/01 $5.00 Copyright © 2001 the American Physiological Society
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
some time (16, 21, 31, 38, 43). More recently, there is
also the clinical impression that cigarette smoking
increases the incidence of skin ischemic necrosis in
skin flap surgery (45). This clinical impression is sup-
ported by the observations that cigarette smoke inten-
sified skin ischemic necrosis in skin flap surgery in the
rat and hamster (11, 23, 28, 42). Skin vasospasm asso-
ciated with cigarette smoking and or use of smokeless
tobacco implies that the by-products of tobacco may
have detrimental effects on the skin vasculature, re-
sulting in reduction in skin circulation, but the mech-
anism is unclear. Epidemiological studies on the dele-
terious effects of cigarette smoking in humans
indicated that the risk in developing cigarette smok-
ing-related cardiovascular disease was related to the
nicotine content of the cigarette (2, 3, 18). Therefore,
much of the research on the pathogenesis of cigarette
smoking-related cardiovascular diseases has been fo-
cused on the deleterious effect of nicotine on the vas-
culature. To date, there is experimental evidence to
indicate that chronic nicotine treatment has deleteri-
ous effects on the hemodynamics of the cardiovascular
system either by causing direct injury to the blood
vessel wall (8, 19, 54) or by modulating the synthesis
and/or release of vasoactive neurohumoral substances
(1, 49, 50) to promote vasospasm and platelet aggrega-
tion.
Of particular interest to us is the effect of nicotine on
skin circulation. With the use of skin flap models, we
have demonstrated that chronic nicotine treatment in
rats and pigs (13–15) mimicked the deleterious effect
of cigarette smoke in increasing the extent of skin
ischemic necrosis in skin flap surgery in the rat and
hamster (11, 23, 28, 42). This deleterious effect was
associated with an increase in skin flap content of
norepinephrine (NE) and a decrease in skin flap blood
flow (13–15). More recently, it has been reported by
other investigators that acute nicotine treatment selec-
tively potentiated NE-induced arteriole contraction in
the hamster cheek pouch. This potentiation effect was
attributed to the nicotine-induced impairment of endo-
thelial function in the arterioles (33). Most recently, it
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
R1097
R1098 NICOTINE AND ENDOTHELIAL FUNCTION
has been demonstrated that acute nicotine treatment
mimicked the cigarette smoking-induced endothelial
dysfunction in the human dorsal hand vein. Specifi-
cally, local or transdermal administration of nicotine at
a dose reproducing plasma concentration observed dur-
ing cigarette smoking impaired endothelium-depen-
dent vasodilation in human dorsal hand veins (9, 47).
All these recent findings may have important implica-
tions for the direct vascular effect of nicotine, which
contributes to the pathogenesis of vasospasm in skin
ischemic disease and in skin flap surgery. To our
knowledge, no studies have been conducted to examine
the effect of nicotine on skin vascular function, espe-
cially in human skin vasculature. Therefore, the objec-
tive of this project was to test the hypothesis that
nicotine may have deleterious effects on the endothe-
lial function in human skin vasculature. To this end,
we used the established isolated perfused human skin
flap model (25, 26, 32) to investigate the local effect of
nicotine on vasoconstrictor and vasodilator responses
in human skin vasculature.
MATERIALS AND METHODS
Source of Human Skin Specimens
The abdominal pannus excised from patients undergoing
dermolipectomy serves no purpose to the patient and is
normally disposed of by incineration. A clinical protocol was
approved for the design of skin flaps (⬃8 ⫻ 18 cm) from the
excised pannus for in vitro skin perfusion experiments. At
the end of each experiment, the skin specimen was returned
to the Department of Pathology for incineration in the nor-
mal manner.
The median age of the patients was 44 yr (range 26–65 yr),
and all these patients were female. The skin specimens
accepted for this project did not have scars, lesions, or infec-
tion, and the patients were not known to smoke cigarettes or
to have any systemic disease.
Skin Flap Model and Cannulation Technique
for In Vitro Skin Perfusion
The anatomy and design of the human skin flap model
used in the present studies were described by us previously
(32). This skin flap model resembled clinical skin free flaps
used for reconstructive surgery. This skin flap model has
been validated and studied with various kinds of vasodilator
and vasoconstrictor drugs (25, 26, 32).
The width and length of this skin flap model were ⬃8 ⫻ 18
cm, respectively. The vascular pedicle in the proximal end of
the skin flap consisted of a paired perforator artery and vein
(0.5- to 1.5-mm diameter), which were cannulated with 22- or
24-gauge angiocatheter, depending on vessel size. Perfusion
buffer containing 10 U/ml of heparin was gently instilled
through the arterial angiocather with a 3-ml syringe until
venous outflow was observed, thus confirming that the vas-
culature of the skin flap was satisfactory for an in vitro
perfusion experiment. All perfusion experiments in the
present studies were initiated within 2 h of excision of skin
specimens. It was previously demonstrated that the skin flap
was thereafter metabolically and physiologically stable for at
least 5 h of in vitro perfusion (26).
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
Skin Perfusion Technique
The skin flap with its cannulated arterial and venous
perforator was placed on an aluminum mesh stand and was
subsequently connected to the commercially available MX
Amber perfuser apparatus (model Two-Ten; MX Interna-
tional, Aurora, CO). The perfusate consisted of modified
Krebs-Henseleit buffer with the following composition (in
mM): 100 NaCl, 4.60 KCl, 1.10 NaH2PO4⫺, 1.20 MgSO42⫺, 2.25
CaCl2, 30 NaHCO3⫺, and 13 glucose. BSA (Cohen fraction V)
was added to the buffer for a final concentration of 6.5%. The
buffer was then stirred and filtered (Whatman no. 40) and
equilibrated within the reservoir chamber of the perfusion
apparatus with 95% O2 and 5% CO2 at 37 ⫾ 0.1°C, pH 7.40 ⫾
0.03 mmHg, and PO2 450 ⫾ 16 mmHg. The PO2 of the
perfusate collected at the venous outflow was 180 ⫾ 25
mmHg. This observation indicated adequate supply of oxy-
gen to the skin flap as there was minimal arteriovenous
shunt flow (⬍1%) in this skin flap model (27). An adjustable-
rate pump (model 7014; Cole-Parmer Instrument, Vernon
Hills, IL) was used to deliver perfusate, which passed
through a bubble trap in the reservoir to the arterial catheter
of the skin flap. A three-way connector linked the tubing from
the pump to the flap and allowed for parallel connection of a
pressure transducer (AB high-performance pressure trans-
ducer; Data Instruments, Lexington, MA). The transducer
output continuously displayed the perfusion pressure on a
digital monitor (Trandicator II 621A digital strain gauge;
Doric Scientific, San Diego, CA) and a chart recorder (Linea-
corder WR 3101; Graphtec).
The buffer flow rate was adjusted (2.0 ⫾ 0.4 ml/min) to
achieve a stable baseline perfusion pressure of 45–50 mmHg.
A baseline of ⬃50 mmHg was chosen because results from
previous studies with this human skin flap model revealed it
to provide good tissue perfusion with minimal leakage and
edema formation of less than 10% water retention (25, 32). In
addition, it has been estimated that a perfusion pressure of
50 mmHg, when Krebs buffer with an albumin concentration
of 65 g/l is used, is equivalent to the perfusion pressure of
⬃90 mmHg in a whole blood perfusion (4). A perfusion
pressure of 50 mmHg with a similar perfusate was also
successfully used in studies of the more aerobically active
isolated perfused cat hindlimb (53).
A 45-min stabilization period was allowed at the beginning
of each experiment. The surface temperature (⬃34°C) of the
skin flap was monitored using a thermistor probe (YS1 series
400; Yellow Springs Instruments, Yellow Springs, OH) con-
nected to a microcomputer thermometer (Series 084202;
Cole-Parmer Instrument). Drugs were infused intra-arteri-
ally through a sidearm port driven by a separate pump of
adjustable rates (model P-1; Pharmacia LKB). Under the
constant flow condition in which the pump rate remained
constant, a change in perfusion pressure in response to drug
administration is indicative of a change in vascular resis-
tance of the skin flap. The change in perfusion pressure
induced by drug treatment in each skin flap was standard-
ized by its baseline perfusion pressure and is expressed as a
percentage of baseline perfusion pressure.
Criteria for Rejection of Skin Flap
A skin flap was deemed to be vascularly reactive if it
demonstrated a concentration-dependent increase in perfu-
sion pressure to NE. At the end of the last dose of NE in each
experiment, ACh (10⫺5 M) was used to test endothelium-
dependent relaxation. Finally, 0.2 ml of fluorescein dye (20
mg) was infused into each flap to characterize the area of skin
281 • OCTOBER 2001 • www.ajpregu.org
 NICOTINE AND ENDOTHELIAL FUNCTION
perfusion. Any skin flap that did not respond to the concen-
tration-dependent vasoconstrictor effect of NE and vasodila-
tor effect of ACh and showed ⬍6-cm length in dye stain
(perfusion) was not included in this study.
Surface Dermofluorometry Technique for Assessment
of Skin Perfusion in Isolated Perfused Skin Flaps
The dermofluorometry technique for indirect assessment
of in vivo dermal perfusion has been validated against the
radioactive microsphere technique in the pig (51). Dermoflu-
orometry has also been applied to the present isolated per-
fused human skin flap model in vitro (24, 25, 32). In the
present study, dermofluorometry was used to corroborate the
observation of skin vasoconstriction assessed by measure-
ment of perfusion pressure. Specifically, confluent circles of
1-cm diameter were marked along the longitudinal midline of
the skin flap surface. After a 45-min stabilization period, the
background fluorescence in each circular skin area was mea-
sured (fluorescence unit) using a dermofluorometer (Fluore-
scan Unit; Santa Barbara Technology, Santa Barbara, CA).
Fluorescein dye with a final concentration of 3 ⫻ 10⫺5 M was
then infused into the skin flap for 4 min, and fluorescence in
each circular area was measured again. A washout period of
15 min was allowed, and the background fluorescence was
taken again. After the perfusion pressure had stabilized
subsequent to drug infusion for study of skin vascular reac-
tivity, fluorescein dye infusion was repeated, and skin fluo-
rescence was measured again. The difference in fluorescence
units for each circular skin area between the background and
postfluorescein dye infusion was defined as the net fluores-
cence units for that area. The total dye fluorescence is the
sum of all net fluorescence units measured from all circular
skin areas along the midline of the skin flap.
Biochemicals
All reagents and drugs were purchased from Sigma Chem-
ical (Oakville, Ontario) except the following: NE from SABEX
Pharmaceutical (Boucherville, Quebec) and fluorescein dye
(fluorescein sodium 10%) from Dioptic Laboratories
(Markham, Ontario).
Indomethacin, a cyclooxygenase inhibitor, was dissolved in
200 ␮l ethanol before mixing with buffer solution. This con-
centration of alcohol did not affect the baseline perfusion
pressure. Other drugs used in the present studies were
dissolved in buffer. Purified water (Milli-Q Water System,
Bedford, MA) was used for making perfusion buffer. Fresh
perfusion buffer and drug solutions were made on the day of
experiment. The drugs were stored at 4°C before use.
Experimental Protocols
Protocol 1: to investigate the effect of nicotine on NE-
induced skin vasoconstriction by assessment of perfusion
pressure in isolated perfused human skin. Cumulative con-
centration-dependent (10⫺7-10⫺5 M) vasoconstrictor effects
of NE on skin perfusion pressure were studied in the absence
and presence of 10⫺7 M nicotine in isolated human skin flaps
perfused with a constant flow rate of Krebs buffer. NE was
used in this study because it was reported that nicotine
selectively potentiated the vasoconstrictor effect of NE in
hamster cheek pouch arterioles (33).
Protocol 2: to investigate the effect of nicotine on NE-
induced skin vasoconstriction by assessment of skin perfusion
in isolated perfused human skin. Skin perfusion was assessed
by the dermofluorometry technique to corroborate the obser-
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
R1099
vation of skin vasoconstriction assessed by measurement of
perfusion pressure in the preceding study. Specifically, the
NE-induced (8 ⫻ 10⫺7 M) decrease in dermal perfusion in
isolated perfused human skin flaps was assessed in the
absence and presence of 10⫺7 M of nicotine, using the der-
mofluorometry technique. Total dye fluorescence was as-
sessed at the end of: the stabilization period (baseline); 15-
min infusion of vehicle; and 15-min infusion of NE in the
absence and presence of 10⫺7 M of nicotine.
Protocol 3: to investigate the role of nicotine receptors and
endogenous cyclooxygenase products in the effect of nicotine
on NE-induced skin vasoconstriction in isolated perfused
human skin. The effect of nicotine (10⫺7 M) on the cumula-
tive concentration-dependent vasoconstrictor effect of NE on
skin perfusion pressure was studied in the absence and
presence of the nicotine-receptor antagonist hexamethonium
(10⫺6 M). Hexamethonium infusion was started 45 min be-
fore nicotine infusion and followed by NE infusion 45 min
later.
In a separate study, all skin flaps were pretreated with
10⫺5 M of indomethacin. The vasoconstrictor effect of NE
(8 ⫻ 10⫺7 M) on skin perfusion pressure was studied in the
absence and presence of 10⫺7 M of nicotine. Infusion of
indomethacin and nicotine was started 45 min before NE
infusion.
Protocol 4: to investigate the effect of nicotine on ACh- or
nitroglycerin-induced vasorelaxation in isolated perfused hu-
man skin. The effect of nicotine (10⫺7 M) on cumulative
concentration-dependent vasorelaxation induced by ACh or
nitroglycerin (NTG) was studied in isolated perfused human
skin flaps preconstricted with 8 ⫻ 10⫺7 M of NE. ACh and
NTG were used in this study because it is well known that
ACh elicits endothelium-dependent vasorelaxation and NTG
elicits endothelium-independent vasorelaxation (35–37).
Statistics
All values are expressed as means ⫾ SE. The number of
observations and the specific statistical test used in each
study are indicated in the legend of each figure. The level of
significance for all tests was set at Pⱕ 0.05.
RESULTS
Effect of Nicotine on NE-Induced Skin
Vasoconstriction Assessed by Perfusion Pressure in
Isolated Perfused Human Skin Flaps
Intra-arterial infusion of nicotine (10⫺7 M) was
started 45 min before infusion of NE in the isolated
perfused human skin flap. Infusion of nicotine alone
did not have any effect on the baseline perfusion pres-
sure. The baseline perfusion pressures before (47.6 ⫾
1.7 mmHg) and after (48.3 ⫾ 1.0 mmHg) nicotine
infusion were not significantly different.
NE elicited an increase in perfusion pressure in
isolated perfused human skin flaps in a concentration-
dependent (10⫺8–10⫺5 M) manner (Fig. 1). This in-
crease in skin perfusion pressure induced by NE was
significantly higher (P ⬍ 0.05) in the presence of 10⫺7
M of nicotine compared with the control, thus indicat-
ing that nicotine amplified the skin vasoconstrictor
effect of NE.
281 • OCTOBER 2001 • www.ajpregu.org
R1100 NICOTINE AND ENDOTHELIAL FUNCTION
Fig. 1. Effect of nicotine (10⫺7 M) on norepinephrine (NE)-induced
skin vasoconstriction. Values are means ⫾ SE, n ⫽ 4. The cumula-
tive concentration-dependent increase in perfusion pressure elicited
by NE in isolated perfused human skin flaps was significantly higher
in the presence of nicotine compared with the control; analysis of
variance with repeated measures, P ⬍ 0.05.
Effect of Nicotine on NE-Induced Skin
Vasoconstriction Assessed by Dermofluorometry
Technique in Isolated Perfused Human Skin Flaps
The amplification effect of nicotine on NE-induced
vasoconstriction was further assessed by using the
dermofluorometry technique. The total dye fluores-
cence in the vehicle-treated skin flaps was similar to
the baseline perfusion (100 ⫾ 4%). Infusion of 8 ⫻ 10⫺7
M of NE significantly (P ⬍ 0.05) reduced the total dye
fluorescence to 37 ⫾ 4% (P ⬍ 0.05) of the control
(vehicle). In the presence of 10⫺7 M of nicotine, this
same concentration of NE further reduced the total
fluorescence to 20 ⫾ 5% (P ⬍ 0.05) of the control (Fig.
2). Thus the present results from the dermofluorom-
etry study supported the observation from the preced-
ing study that nicotine amplified the skin vasoconstric-
tor effect of NE by assessment of perfusion pressure.
Fig. 2. Effect of nicotine (10⫺7 M) on NE-induced (8 ⫻ 10⫺7 M)
decrease in dermal perfusion. Dermal perfusion was assessed by
measuring the total dye fluorescence in the skin flap using a der-
mofluorometry technique. Total dye fluorescence is expressed as a
percentage of the control baseline before drug treatment. Values are
means ⫾ SE, n ⫽ 4. Means without a common letter are significantly
different (a ⬎ b ⬎ c); 1-way analysis of variance followed by Duncan’s
multiple range test for comparison of means, P ⬍ 0.05.
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
Fig. 3. Role of nicotine receptors in the effect of nicotine (10⫺7 M) on
NE-induced skin vasoconstriction. Values are means ⫾ SE, n ⫽ 4.
The cumulative concentration-dependent increase in perfusion elic-
ited by NE and nicotine was not significantly different in the pres-
ence or absence of the nicotine-receptor antagonist hexamethonium
(10⫺6 M); analysis of variance with repeated measures.
Role of Nicotine Receptors and Endogenous
Cyclooxygenase Products in Effect of Nicotine on NE-
Induced Vasoconstriction in Isolated Perfused Human
Skin Flaps
In the preceding study, it was observed that the
NE-induced increase in skin perfusion pressure in hu-
man skin was significantly (P ⬍ 0.05) higher in the
presence of 10⫺7 M of nicotine compared with the
control (Fig. 1). Here, it was observed that the increase
in skin perfusion pressure induced by combined NE
and nicotine was similar in skin flaps with or without
pretreatment with the nicotine-receptor antagonist
hexamethonium (10⫺6 M) (Fig. 3). Therefore, nicotine
receptors blocked by hexamethonium were not in-
volved in the amplification effect of nicotine in NE-
induced increase in perfusion pressure.
In a separate study, all skin flaps were pretreated
with indomethacin (10⫺5 M). It was observed that the
NE-induced increase in perfusion pressure was still
significantly (P ⬍ 0.05) higher in the presence of 10⫺7
M of nicotine compared with NE alone (Fig. 4). This
observation is interpreted to indicate that the endoge-
nous cyclooxygenase products were not involved in the
Fig. 4. Role of cyclooxygenase products in the effect of nicotine (10⫺7
M) on NE-induced skin vasoconstriction. Values are means ⫾ SE,
n ⫽ 4. All isolated perfused skin flaps were pretreated with the
cyclooxygenase inhibitor indomethacin (10⫺5 M). The increase in
perfusion pressure induced by NE and nicotine was significantly
higher than NE alone; Mann-Whitney U test, P ⬍ 0.05.
281 • OCTOBER 2001 • www.ajpregu.org
 NICOTINE AND ENDOTHELIAL FUNCTION R1101

amplification effect of nicotine on NE-induced increase

in perfusion pressure.

Effect of Nicotine on ACh- and NTG-Induced

Vasorelaxation in Isolated Perfused Human Skin
Flaps Preconstricted with NE
ACh elicited a concentration-dependent relaxation in
isolated perfused human skin flaps preconstricted with
8 ⫻ 10⫺7 M of NE (Fig. 5). The concentration-depen-
dent vasorelaxation induced by ACh in isolated per-
fused human skin flaps preconstricted with NE was
significantly (P ⬍ 0.05) reduced in the presence of 10⫺7
M of nicotine compared with the control (Fig. 5). Fig. 6. Effect of nicotine (10⫺7 M) on cumulative concentration-
NTG also elicited a concentration-dependent vasore- dependent vasorelaxation of nitroglycerin in isolateral perfused hu-
laxation in isolated perfused human skin flaps precon- man skin flap preconstricted with 8 ⫻ 10⫺7 M of NE. Vasorelaxation
stricted with 8 ⫻ 10⫺7 M of NE. However, the concen- is expressed as a percentage of contraction to 8 ⫻ 10⫺7 M of NE.
Values are means ⫾ SE, n ⫽ 5. The vasorelaxation induced by
tration-dependent vasorelaxation induced by NTG in nitroglycerin was not significantly different in the presence or ab-
human skin flaps preconstricted with 8 ⫻ 10⫺7 M of NE sence of nicotine; analysis of variance with repeated measures.
was not significantly different in the absence or pres-
ence of 10⫺7 M of nicotine (Fig. 6).
impair the endothelial vasorelaxation function in the
DISCUSSION
vasculature.
Major Findings in the Present Studies
Levels of Nicotine and NE Used for the Study
We used the isolated perfused human skin flap of Endothelial Function
model to investigate the acute local effect of nicotine on
Studies on nicotine pharmacokinetics in cigarette
vasoconstrictor and vasodilator responses in human
smokers have shown that peak levels of plasma nico-
skin vasculature. We observed that 1) nicotine ampli-
tine range from ⬃30 to 50 ng/ml (7, 12). The half-life of
fied the concentration-dependent skin vasoconstrictor
plasma nicotine due to metabolism is ⬃2.2 h (12). It
effect of NE; 2) nicotine attenuated endothelium-de-
has also been reported that most cigarette smokers
pendent skin vasorelaxation induced by ACh in NE-
tend to modify their smoking behavior (e.g., depth of
preconstricted skin vasculature; 3) nicotine did not
puffing or inhalation, and frequency of smoking) to
affect the endothelium-independent vasorelaxation in-
maintain fairly constant plasma levels of nicotine (12).
duced by NTG in NE-preconstricted skin vasculature;
The average peak plasma levels of nicotine in smoke-
and 4) cyclooxygenase products and nicotine receptors
less tobacco users have also been reported to range
blocked by hexamethonium were not involved in the
from ⬃22 to 30 ng/ml (17). The level of nicotine in the
amplification of NE-induced skin vasoconstriction by
perfusion buffer used in the present studies was 10⫺7
nicotine. These observations led us to speculate that
M (16.2 ng/ml), which is well within the range of
exposure of human skin vasculature to nicotine may
plasma levels of nicotine in cigarette smokers and
users of smokeless tobacco.
We previously observed that NE elicited concentra-
tion-dependent skin vasoconstriction over a range of
10⫺7–10⫺5 M in isolated perfused pig skin flaps (44).
The mean concentration of NE required to produce
half-maximal increase in perfusion pressure (EC50)
was 1.1 ⫻ 10⫺6 M. Similar results were reported by
other investigators for isolated perfused pig skin flaps
(46). In addition, the skin contents of NE in rat and pig
skin flaps were also in the range of 10⫺7–10⫺6 M (15,
22). Therefore, 10⫺7- to 10⫺5-M concentrations of NE
were used for the present studies.

Experimental Model for the Study of Nicotine

Fig. 5. Effect of nicotine (10⫺7 M) on cumulative concentration-
dependent vasorelaxation induced by ACh in isolated perfused hu-
man skin flaps preconstricted with 8 ⫻ 10⫺7 M of NE. Vasorelaxation
is expressed as a percentage of contraction to 8 ⫻ 10⫺7 M of NE.
Values are means ⫾ SE, n ⫽ 5. The ACh-induced vasorelaxation was
significantly reduced in the presence of nicotine compared with the
control; analysis of variance with repeated measures, P ⬍ 0.05.
AJP-Regulatory Integrative Comp Physiol • VOL
on Vasoconstrictor and Vasodilator Responses
in Human Skin Vasculature
Many of the in vivo cardiovascular effects observed
after cigarette smoking, such as increases in blood
pressure and heart rate, result from activation of the
sympathoadrenal system by nicotine (5, 6). Thus the
281 • OCTOBER 2001 • www.ajpregu.org

Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.

R1102 NICOTINE AND ENDOTHELIAL FUNCTION
net effect of nicotine on vascular tone is determined by
the balance between peripheral and central sites of
action of nicotine, and the local vascular effect of nico-
tine cannot be distinguished. Recently, the human
dorsal hand vein model was used to study the acute
local effect of nicotine on vascular endothelial function
with minimal systemic effect. Specifically, physiologi-
cal saline with or without nicotine and/or vasoactive
drugs was infused continuously into a dorsal hand vein
to study the local vasoconstriction and vasorelaxation
effects of these drugs by measuring the changes in
diameters of the dorsal hand vein (9). So far, there is no
study on the local effect of nicotine on human skin
vasculature relevant to vasospasm in skin ischemic
disease and in skin flap surgery. Our isolated perfused
human skin flap model permitted local intra-arterial
infusion of nicotine and drugs into the skin vasculature
to study local vasoconstrictor and vasodilator re-
sponses (26, 32). Krebs buffer instead of whole blood
was used for perfusate in the present studies so that
the local effect of nicotine on endothelial function could
be investigated in the absence of blood cells and circu-
lating neurohumoral vasoactive substances.
Local Effect of Nicotine on Vasoconstrictor and
Vasodilator Responses in Human Skin Vasculature
In the present studies, we have obtained evidence to
indicate that nicotine amplified skin vasoconstriction
induced by NE. Specifically, NE-induced skin vasocon-
striction assessed by perfusion pressure or dermal per-
fusion was significantly increased in the presence of
10⫺7 M of nicotine compared with NE alone (Figs. 1
and 2).
There is also evidence from the present studies to
indicate that nicotine amplified the NE-induced skin
vasoconstriction independent of nicotine receptors
blocked by hexamethonium and endogenous cyclooxy-
genase products. This interpretation is based on the
following observations. The nicotine-receptor antago-
nist hexamethonium did not attenuate the concentra-
tion-dependent increase in perfusion pressure induced
by combined NE and nicotine in isolated perfused hu-
man skin flaps (Fig. 3). Moreover, the cyclooxygenase
inhibitor indomethacin did not block the amplification
effect of nicotine in NE-induced increase in perfusion
pressure in isolated perfused human skin flaps (Fig. 4).
Furthermore, we have also obtained evidence to in-
dicate that nicotine attenuated endothelium-depen-
dent but not endothelium-independent vasorelaxation
in human skin. Specifically, the concentration-depen-
dent vasorelaxation effect of ACh in isolated perfused
human skin flaps preconstricted with NE was signifi-
cantly reduced in the presence of 10⫺7 M of nicotine
compared with ACh alone (Fig. 5). However, the same
nicotine treatment did not affect the concentration-
dependent vasorelaxation induced by NTG in isolated
perfused human skin flaps preconstricted with NE
(Fig. 6).
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
Mechanism of Action of Nicotine in Vasoconstrictor
and Vasodilator Responses in Human Skin
Vasculature
We previously observed in isolated perfused pig skin
flaps that NE-induced concentration-dependent vaso-
constriction was enhanced in the presence of the nitric
oxide synthase (NOS) inhibitor NG-nitro-L-arginine
(L-NNA), and this skin vasoconstriction was further
enhanced in the presence of both L-NNA and the cyclo-
oxygenase inhibitor indomethacin (15). We also previ-
ously observed that L-NNA attenuated the vasorelax-
ation effect of ACh in isolated perfused human skin
flaps preconstricted with NE (25). These observations
were interpreted to indicate that the endothelium of
skin vasculature could be stimulated to synthesize
and/or release vasorelaxing factors such as NO and
prostacyclin (PGI2) to counteract the vasoconstrictor
effect of NE or to mediate the vasorelaxation effect of
ACh. Experimental evidence from other laboratories is
also available to indicate that contractions of isolated
blood vessels caused by NE or stimulation of adrener-
gic nerves were associated with simultaneous release
of endothelium-derived NO and PGI2 (10, 48). The
mechanisms were not investigated in these studies.
However, it is known that NE activates ␣1-adrenocep-
tors in smooth muscle cells to cause vasoconstriction.
NE also activates ␣2-adrenoceptors in the endothelium
to stimulate synthesis/release of NO and PGI2 (52).
ACh is also known to activate muscarinic receptors in
the endothelium to stimulate synthesis/release of NO
(52). These previous observations provide insight into
the mechanism of nicotine in amplification of NE-in-
duced skin vasoconstriction (Fig. 1) and in attenuation
of ACh-induced vasorelaxation (Fig. 5) observed in
isolated perfused human skin flaps in the present stud-
ies. Specifically, we speculate that nicotine might have
inhibited the NOS activity in the endothelium to cause
a reduction in synthesis/release of NO in the endothe-
lium. Therefore, there was less NO to counteract NE-
induced skin vasoconstriction or to mediate ACh-in-
duced vasorelaxation. We further speculate that the
mechanism of action of nicotine did not involve hexa-
methonium-sensitive receptors or cyclooxygenase
products, because the amplification effect of nicotine on
NE-induced skin vasoconstriction was not attenuated
by hexamethonium or indomethacin (Figs. 3 and 4).
This line of reasoning may also explain why nicotine
did not affect the baseline perfusion pressure of iso-
lated perfused human skin flaps. Specifically, the de-
crease in NO synthesis due to inhibition of NOS activ-
ity in the endothelium by nicotine could have been
compensated by an increase in PGI2 synthesis at base-
line level; thus the baseline perfusion pressure re-
mained unchanged. However, this small increase in
PGI2 synthesis could not compensate for the decrease
in NO synthesis in the endothelium caused by nicotine
in NE-induced skin vasoconstriction or in ACh-induced
vasorelaxation. There is in vivo evidence to support
this speculation as well. Specifically, it has been re-
ported that a similar level of nicotine (14.0 ⫾ 1.6 ng/ml)
281 • OCTOBER 2001 • www.ajpregu.org
 NICOTINE AND ENDOTHELIAL FUNCTION
did not affect the baseline diameter of the hamster
cheek pouch arterioles, but this same level of nicotine
potentiated the NE-induced vasoconstriction in the
hamster cheek pouch in vivo (33, 34).
Potential Limitations
It is important to point out that we only studied the
acute effect of nicotine on vasoconstrictor and vasodi-
lator responses in NE-preconstricted human skin vas-
culature. It is not known if these responses will also be
seen in chronic nicotine treatment and/or with vaso-
constrictors other than NE. There is evidence to indi-
cate that nicotine impaired ACh-induced endothelium-
dependent vasorelaxation in both acute and chronic
nicotine treatment in hamster cheek pouch arterioles
(35–37); thus there is evidence for us to speculate that
the acute and chronic deleterious effect of nicotine on
endothelial function is similar. However, the amplifi-
cation effect of nicotine on vasoconstriction in hamster
pouch arterioles seemed to be selective to NE (33).
So far, we have demonstrated that nicotine amplified
the vasoconstrictor effect of NE and impaired the ACh-
induced endothelium-dependent vasorelaxation in hu-
man skin vasculature in vitro. These findings are in
line with the in vivo observations reported by other
investigators that local administration of nicotine po-
tentiated the vasoconstrictor effect of NE (33) and
impaired the endothelium-dependent vasorelaxation
in hamster cheek pouch arterioles (35–37) and in hu-
man dorsal hand veins (9, 47). However, it is unclear
why these vascular effects of nicotine were not ob-
served in tail artery ring segments and isolated per-
fused mesenteric vasculature obtained from rats with
chronic nicotine treatment or in coronary artery seg-
ments obtained from dogs with chronic nicotine treat-
ment (29, 39). The differences in these results could
reflect differences in vasculature, experimental design,
and levels of nicotine used.
Perspectives
The effect of nicotine on amplification of NE-induced
vasoconstriction and impairment of endothelium-de-
pendent vasorelaxation observed in human skin in the
present studies may have important clinical implica-
tions in the pathogenesis of skin vasospasm in skin flap
surgery and skin ischemic disease. For example, there
is the clinical impression that cigarette smoking in-
creases the incidence of skin ischemic necrosis in skin
flap surgery (45). We have previously demonstrated
that chronic nicotine treatment in rats and pigs (13–
15) mimicked the detrimental effect of cigarette smok-
ing in increasing the extent of skin flap ischemic ne-
crosis, and this detrimental effect was associated with
an increase in skin content of NE, presumably released
by nerve endings in the skin (15). Here, we demon-
strated for the first time in human skin that nicotine
may also amplify the skin vasoconstrictor effect of NE,
probably by inhibition of NOS activity in the endothe-
lium to reduce NO synthesis. Together, these observa-
tions provide insight into the pathogenesis of intensi-
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
R1103
fied skin vasospasm induced by cigarette smoking in
skin flap surgery. However, we have also previously
observed in rats that the detrimental effect of chronic
nicotine treatment on skin flap blood flow and viability
was reversible and was not encountered when nicotine
was withheld 2 wk before skin flap surgery (14). This
information may also be relevant to certain states of
ischemic skin disease as it has been reported that
resolution of symptoms such as skin vasospasm and
pain in Buerger’s disease (thromboangiitis obliterans)
induced by use of tobacco could be achieved in some
patients by abstinence from tobacco use and a regimen
of a vasodilating drug (20, 30, 40, 43).
In summary, we demonstrated for the first time that
acute nicotine treatment amplified the skin vasocon-
strictor effect of NE and impaired the endothelium-
dependent vasorelaxation in isolated perfused human
skin flaps preconstricted with NE. Cyclooxygenase
products and nicotine receptors blocked by hexametho-
nium were not involved in the amplification of NE-
induced skin vasoconstriction by nicotine. These obser-
vations lent support to our hypothesis that nicotine
may have deleterious effects on endothelial function in
skin vasculature, but the mechanism has yet to be
investigated.
The authors thank D. Whitwell and D. McIntyre for typing this
manuscript.
This research project was supported by an operating grant from
The Smokeless Tobacco Research Council (Grant 0784) and the
Canadian Institute of Health Research (Grant MOP-8048) to C. Y.
Pang.
REFERENCES
1. Armitage AK. Effects of nicotine and tobacco smoke on blood
pressure and release of catecholamines from the adrenal glands.
Br J Pharmacol 25: 515–526, 1965.
2. Aronow WS and Swansen AJ. The effect of low-nicotine ciga-
rettes on angina pectoris. Ann Intern Med 71: 599–601, 1969.
3. Auerback O, Carter HW, Garfinkle L, and Hammond EC.
Cigarette smoking and coronary artery disease. Chest 70: 697–
705, 1976.
4. Avolio AP, Spain JAE, and Liard JD. Plasma protein concen-
tration and control of coronary vascular resistance in isolated rat
heart. Am J Physiol Heart Circ Physiol 238: H471–H480, 1980.
5. Benowitz NL. The role of nicotine in smoking-related cardio-
vascular diseases. Prev Med 26: 412–417, 1997.
6. Benowitz NL and Gourlay SG. Cardiovascular toxicity of
nicotine: implications for nicotine replacement therapy. J Am
Coll Cardiol 29: 1422–1431, 1997.
7. Benowitz NL, Kuyt F, and Jacob P III. Circadian blood
nicotine concentrations during cigarette smoking. Clin Pharma-
col Ther 32: 758–764, 1982.
8. Booyse FM, Ouikowicz G, and Quarfoot AJ. Effects of
chronic oral consumption of nicotine on the rabbit aortic endo-
thelium. Am J Pathol 102: 229–238, 1981.
9. Chalon S, Moreno H, Benowitz NL, Hoffman BB, and
Blaschke TF. Nicotine impairs endothelium-dependent dilation
in human veins in vivo. Clin Pharmacol Ther 67: 391–397, 2000.
10. Cohen R. Role of the endothelium in vascular adrenergic neu-
rotransmission. In: Endothelial Regulation of Vascular Tone,
edited by Ryan US and Rubanyi GM. New York: Dekker, 1992, p.
155–169.
11. Craig S and Rees TD. The effects of smoking on experimental
skin flaps in hamsters. Plast Reconstr Surg 75: 842–846, 1985.
12. Feyerabend C, Ings RMJ, and Russell MAH. Nicotine phar-
macokinetics and its application to intake from smoking. Br J
Clin Pharmacol 19: 239–247, 1985.
281 • OCTOBER 2001 • www.ajpregu.org
R1104 NICOTINE AND ENDOTHELIAL FUNCTION
13. Forrest CR, Pang CY, and Lindsay WK. Dose and time
effects of nicotine treatment on the capillary blood flow and
viability of random pattern skin flaps in the rat. Br J Plast Surg
40: 295–299, 1987.
14. Forrest CR, Pang CY, and Lindsay WK. Pathogenesis of
ischemic necrosis in random-pattern skin flaps induced by long-
term low-dose nicotine treatment. Plast Reconstr Surg 87: 518–
528, 1991.
15. Forrest CR, Xu N, and Pang CY. Evidence for nicotine-in-
duced skin flap ischemic necrosis in the pig. Can J Physiol
Pharmacol 72: 30–38, 1994.
16. Gifford RW and Hines EA. Complete clinical remission in
thromboangiitis obliterans during abstinence from tobacco: re-
port of case. Proc Mayo Clinic 26: 241–245, 1951.
17. Gritz ER, Baer-Weiss V, Benowitz NL, Van Vunakis H, and
Jarvick ME. Plasma nicotine and cotinine concentration in
habitual smokeless tobacco users. Clin Pharmacol Ther 30: 201–
209, 1981.
18. Hill P and Wynder EL. Smoking and cardiovascular disease:
effect of nicotine on the serum epinephrine and corticoids. Am
Heart J 87: 491–496, 1974.
19. Hladove CJ. Endothelial injury by nicotine and its prevention.
Experientia 35: 1585–1586, 1978.
20. Joyce JW. Buerger’s disease (thromboangiitis obliterans).
Rheum Dis Clin North Am 16: 463–470, 1990.
21. Juergens JL, Barker NW, and Hines EA. Arteriosclerosis
obliterans: review of 520 cases with special reference to patho-
genic and prognostic factors. Circulation 21: 188–195, 1960.
22. Jurell G, Hjemdahl P, and Fredholm BB. On the mechanism
by which antiadrenergic drugs increase survival of critical skin
flaps. Plast Reconstr Surg 72: 518–525, 1983.
23. Kaufman T, Eihenlaub EH, Levine M, Hurwitz DJ, and
Klain M. Tobacco smoking: importance of experimental flap
survival. Ann Plast Surg 13: 468–472, 1984.
24. Kreidstein ML, Levine RH, Knowlton RJ, and Pang CY.
Serial fluorometric assessments of skin perfusion in isolated
perfused human skin flaps. Br J Plast Surg 48: 288–293, 1995.
25. Kreidstein MD, Pang CY, Carlsen LN, and Xu N. Evidence
for endothelium-dependent and endothelium-independent vaso-
dilation in human skin flaps. Can J Physiol Pharmacol 70:
1208–1216, 1992.
26. Kreidstein ML, Pang CY, Levine RH, and Knowlton RJ.
The isolated perfused human skin flap: design, perfusion tech-
nique, metabolism and vascular reactivity. Plast Reconstr Surg
87: 741–749, 1991.
27. Kreidstein ML, Zacks SL, and Pang CY. Study of the micro-
circulation in the isolated perfused human skin flap model. Surg
Forum 41: 664–666, 1990.
28. Lawrence WT, Murphy CR, Robson MC, and Hegger JP.
The detrimental effect of cigarette smoking on flap survival: an
experimental study in the rat. Br J Plast Surg 37: 216–219,
1984.
29. Li Z, Barrios V, Buchholz JN, Glenn TG, and Duckles SP.
Chronic nicotine administration does not affect peripheral vas-
cular reactivity in the rat. J Pharmacol Exp Ther 271: 1135–
1142, 1994.
30. Lie JT. Thromboangiitis obliteran (Buerger’s disease) revisited.
Pathol Annu 23: 257–291, 1988.
31. Lie JT. Thromboangiitis obliteran (Buerger’s disease) and
smokeless tobacco. Arthritis Rheum 31: 812–813, 1988.
32. Lipa JE, Neligan PC, Perreault TM, Baribeau J, Levine
RH, Knowlton RJ, and Pang CY. Vasoconstrictor effect of
endothelin-1 in human skin: role of ETA and ET␤ receptors. Am J
Physiol Heart Circ Physiol 276: H359–H367, 1999.
33. Mayhan WG. Acute infusion of nicotine potentiates norepineph-
rine-induced vasoconstriction in the hamster cheek pouch. J Lab
Clin Med 133: 48–54, 1999.
AJP-Regulatory Integrative Comp Physiol • VOL
Downloaded from journals.physiology.org/journal/ajpregu (046.097.169.022) on June 2, 2021.
34. Mayhan WG and Patel KP. Effect of nicotine on endothelium-
dependent arteriolar dilation in vivo. Am J Physiol Heart Circ
Physiol 272: H2337–H2342, 1997.
35. Mayhan WG and Sharpe GM. Superoxide dismutase restores
endothelium-dependent arteriolar dilation during acute infusion
of nicotine. J Appl Physiol 85: 1292–1298, 1998.
36. Mayhan WG and Sharpe GM. Chronic exposure to nicotine
alters endothelium-dependent arteriolar dilatation: effect of su-
peroxide dismutase. J Appl Physiol 86: 1126–1134, 1999.
37. Mayhan WG, Sharpe GM, and Anding P. Agonist-induced
release of nitric oxide during acute exposure nicotine. Life Sci 65:
1829–1857, 1999.
38. McKusick VA, Harris WS, Otteson OE, Goodman RM, Shel-
ley WM, and Bloodwell RO. Burger’s disease: a distinct clin-
ical and pathologic entity. JAMA 181: 5–12, 1962.
39. Miller VM, Clouse WD, Tonnessen BH, Boston US, Sever-
son SR, Bonde S, Rud KS, and Hurt RD. Time and dose effect
of transdermal nicotine on endothelial function. Am J Physiol
Heart Circ Physiol 279: H1913–H1921, 2000.
40. Mills JL and Proten JM. Buerger’s disease: a review and
update. Semin Vasc Surg 6: 14–23, 1993.
41. Moreno H Jr, Chalon S, Urae A, Tangphao O, Abiose AK,
Hoffman BB, and Blaschke TF. Endothelial dysfunction in
human hand vein is rapidly reversible after smoking cessation.
Am J Physiol Heart Circ Physiol 275: H1040–H1045, 1998.
42. Nolan J, Jenkins RA, Kurihara K, and Schultz R. The acute
effects of cigarette smoke exposure on experimental skin flaps.
Plast Reconstr Surg 75: 544–549, 1985.
43. O’Dell JR, Linder J, Markin RS, and Moore GF. Thrombo-
angiitis obliterans (Buerger’s disease) and smokeless tobacco.
Arthritis Rheum 30: 1054–1056, 1987.
44. Pang CY, Yang RZ, Neligan P, Xu N, Chin C, Zhong A, and
Forrest CR. Vascular effects and mechanisms of action of en-
dothelium-1 in isolated perfused pig skin. J Appl Physiol 79:
2106–2113, 1995.
45. Rees TD, Liverett DM, and Guy CI. The effect of cigarette
smoking on skin flap survival in the face-lift patient. Plast
Reconstr Surg 73: 911–915, 1984.
46. Rogers RA and Riviere JE. Pharmacologic modulation of the
cutaneous vasculature in the isolated perfused porcine skin flap.
J Pharm Sci 83: 1682–1689, 1994.
47. Sabha M, Tanus-Santos JE, Toledo JCY, Cittadino M,
Rocha JC, and Moreno H. Transdermal nicotine mimics the
smoking-induced endothelial disfunction. Clin Pharmacol Ther
68: 167–174, 2000.
48. Segarra G, Medina P, Revert F, Masia S, Vila JM, Sucha L,
and Aldasoro M. Modulation of adrenergic contraction of dog
pulmonary arteries by nitric oxide and prostacyclin. Gen Phar-
macol 32: 583–589, 1999.
49. Seidler FJ and Slotkin TA. Effects of chronic nicotine admin-
istration on the denervated rat adrenal medulla. Br J Pharmacol
56: 201–208, 1976.
50. Su C and Bevan JA. Blockade of the nicotine-induced norepi-
nephrine release by cocaine, phenoxybenzamine and desipri-
mine. J Pharmacol Exp Ther 175: 533–540, 1970.
51. Thomson JG and Kerrigan CL. Dermofluorometry: thresh-
olds for prediction flap survival. Plast Reconstr Surg 83: 859–
864, 1989.
52. Vanhoutte PM. Endothelium dysfunction and vascular disease.
In: Endothelium, Nitric Oxide and Atherosclerosis: From Basic
Mechanisms to Clinical Implications, edited by Panza JA and
Cannon RO III. Armonk, NY: Futura, 1999, p. 79–96.
53. Watson PD. Colloid osmotic pressure changes in isolated iso-
gravimetric cat hindlimb. Am J Physiol Heart Circ Physiol 242:
H512–H519, 1982.
54. Zimmerman M and McGeachie J. The effect of nicotine on
aortic endothelium: a quantitative ultrastructural study. Athero-
sclerosis 63: 33–41, 1987.
281 • OCTOBER 2001 • www.ajpregu.org