Matthew R.

Nangle
T. Michael Gibson
Effects of Eugenol on Nerve and Vascular Dysfunction Mary A. Cotter
in Streptozotocin-Diabetic Rats Norman E. Cameron
Original Paper

Abstract was 44 % reduced by diabetes; eugenol corrected this deficit by

69 %. For renal artery rings, maximum endothelium-dependent
Hyperglycaemia in diabetes mellitus results in oxidative stress relaxation to acetylcholine was 51 % reduced by diabetes; euge-
and pro-inflammatory changes which contribute to vascular nol corrected this deficit by 60 %, with improvements in both
complications including endothelial dysfunction and peripheral nitric oxide and endothelium-derived hyperpolarising factor

Downloaded by: Karolinska Institutet. Copyrighted material.

neuropathy. The aim of this study was to examine whether treat-
ment with the dominant ingredient of clove oil, eugenol, which
has antioxidant and anti-inflammatory properties, could im-
prove diabetic vascular and nerve function in streptozotocin-in-
duced diabetic rats. Intervention treatment was given for 2
weeks following 6 weeks of untreated diabetes. Dose-ranging
studies on diabetic deficits in sciatic nerve motor and saphenous
nerve sensory nerve conduction velocities gave ED50 values of
28 mg/kg and 9 mg/kg, respectively, conduction velocity being
within the non-diabetic range at a dose of 200 mg/kg. Sciatic
494 nerve endoneurial blood flow was 49 % reduced by diabetes and
this was completely corrected by 200 mg/kg eugenol treatment.
Gastric fundus maximum nitrergic nerve-mediated relaxation
(EDHF)-mediated vasorelaxation components. Diabetes in-
creased renal artery sensitivity to phenylephrine-mediated con-
traction, however, this was unaffected by eugenol treatment.
Thus, aspects of both vascular and neural complications in ex-
perimental diabetes are improved by eugenol, which could have
potential therapeutic implications for diabetic neuropathy and
vasculopathy.
Key words
Eugenol ´ Syzygium aromaticum ´ diabetes ´ nerve conduction ´
pain ´ blood flow ´ gastric fundus ´ neuropathy ´ vascular endothe-
lium ´ nitric oxide ´ EDHF ´ rat

Introduction autonomic and somatic system dysfunction. The latter includes

Hyperglycaemia and dyslipidaemia in diabetes mellitus give rise
to the major vascular complications of nephropathy, neuropathy
and retinopathy. Vascular endothelium and the nitric oxide (NO)
vasodilator system have been identified as vulnerable targets.
The mechanisms involved in both nerve and vascular dysfunc-
tion include elevated oxidative stress and stimulation of pro-in-
flammatory changes [1]. Diabetic neuropathy arises from degen-
erative changes affecting all nerve fibre populations, which cause
reduced nerve conduction velocity (NCV), which is reversible in
the early stages of the disease. The innervation to the major or-
gan systems is disrupted, causing adverse effects on cardiac, uri-
nogenital and gastrointestinal systems [2], [3]. The autonomic
non-adrenergic non-cholinergic (NANC) innervation to corpus
cavernosum and gastrointestinal system, which uses NO as the
major neurotransmitter, shows an early vulnerability in experi-
mental models such as the streptozotocin-induced diabetic rat
[4], [5], [6], [7]. The adverse vascular changes in diabetes contrib-

Affiliation
School of Medical Sciences, University of Aberdeen, Aberdeen, Scotland, UK

Correspondence
Prof. Dr. Norman E. Cameron ´ School of Medical Sciences ´ Institute of Medical Sciences ´ University of
Aberdeen ´ Foresterhill ´ Aberdeen AB25 2ZD ´ Scotland ´ UK ´ Phone: +44-1224-555-713 ´
Fax: +44-1224-555-719 ´ E-mail: n.e.cameron@abdn.ac.uk

Received August 20, 2005 ´ Accepted November 17, 2005

Bibliography
Planta Med 2006; 72: 494±500  Georg Thieme Verlag KG Stuttgart ´ New York
DOI 10.1055/s-2005-916262 ´ Published online February 10, 2006
ISSN 0032-0943
ute to nerve dysfunction because of impaired perfusion, which
leads to chronic endoneurial hypoxia [2].
Clove (Syzygium aromaticum) oil has antioxidant properties and is
composed primarily (91 %) of eugenol (2-allyl-4-methoxyphenol)
[8]. Eugenol also has anti-inflammatory actions, suppressing tu-
mour necrosis factor alpha signalling and cyclooxygenase 2 ex-
pression [9], [10]. At high concentrations, eugenol can cause vaso-
dilation [11]. The prediction is that all of these properties, which
may or may not be mechanistically linked, could potentially be
useful in the treatment of diabetic neuropathy [1]. Therefore, the
aim of this work was to assess the effectiveness of eugenol inter-
vention treatment on diabetic somatic and autonomic nerve and
vascular defects in streptozotocin-induced diabetic rats.
Materials and Methods
Animals, diabetes induction and treatment
Adult (19 weeks old) male Sprague-Dawley rats from the Univer-
sity of Aberdeen breeding colony were used. All animals received
standard laboratory chow and had access to water ad libitum. Ex-
periments were performed in accordance with regulations speci-
fied by the United Kingdom ªAnimal Procedures Act, 1986º and
the National Institutes of Health ªPrinciples of Laboratory Ani-
mal Care, 1985 revised versionº.
Diabetes was induced by approximately 42.5 mg/kg i. p. strepto-
zotocin freshly made up in sterile saline; duration was 8 weeks.
Groups comprised age-matched non-diabetic and diabetic con-
trol and diabetic rats receiving eugenol (Sigma-Aldrich, Poole,
Dorset, UK; purity 98 %) treatment for 2 weeks, after 6 weeks of
untreated diabetes, as a dietary supplement added to the rat
chow (1.5 g/kg chow). Diabetic rats eat 40 ± 50 g chow per day
and weigh 350 ± 400 g (Table 1), thus they received a dose of ap-
proximately 200 mg/kg/day. There were no indications of ad-
verse side effects over the treatment period for this dose of euge-
nol. Initial dose-ranging studies showed that this was at the top
of the dose-response curve for correction of motor and sensory
NCV deficits in diabetic rats (Fig. 1). Separate groups of rats
were used for in vivo studies on NCV and nerve blood flow, and
in vitro studies on gastric fundus and renal artery function.
Nerve conduction and blood flow
Rats were anaesthetised with thiobutabarbital (Astra-Zeneca,
Macclesfield, Cheshire, UK; 50 ± 100 mg/kg), by intraperitoneal
injection. The trachea was cannulated for artificial ventilation. A
cannula in the right carotid artery was used to monitor mean
systemic blood pressure. Core temperature was monitored and
regulated at 37 ± 38 8C, using a rectal probe and radiant heat.
The sciatic nerve was exposed between the sciatic notch and
knee and motor NCV was measured as previously described [12]
in the nerve branch to tibialis anterior muscle. Saphenous nerve
sensory NCV was measured between the groin and calf.
Blood flow was estimated by hydrogen clearance microelectrode
polarography as previously described [12]. The sciatic nerve was
exposed between the sciatic notch and knee and the skin around
the incision was used to form a pool that was filled with paraffin
oil maintained at 35 ± 37 8C by radiant heat during measure-
ments. A glass-insulated hydrogen-sensitive platinum micro-
electrode was inserted into the sciatic endoneurium. 10 % H2
was added to the inspired gas, the proportions of O2 and N2 being
adjusted to 20 % and 70 %, respectively. When the H2 current re-
corded by the electrode had stabilised, the H2 supply was shut off
and N2 delivery was increased appropriately. The H2 clearance
curve was recorded until baseline; this procedure was then re-
peated at another site. Clearance curves were fitted off-line by
computer using non-linear regression, the Marquardt algorithm
and the least squares method for optimising goodness-of-fit. The
slow exponent was taken to reflect nutritive flow.
Gastric fundus experiments
Rats were anaesthetised (5 % halothane in air) and the stomach
Original Paper
was removed. The animals were then exsanguinated before re-
moval of their left renal artery, and plasma was stored for subse-
quent glucose analysis (Ascensia Esprit 2 glucose meter, Bayer
Diagnostics, Dublin, Ireland).
The fundus was dissected free, pinned flat and three longitudinal
muscle strips were prepared as previously described [7]. Each
was mounted in a 10-mL water-jacketed organ bath under a rest-
Downloaded by: Karolinska Institutet. Copyrighted material.
ing tension of 2 g in modified Krebs-Ringer solution (144 NaCl, 5
KCl, 1.1 MgSO4, 25 NaHCO3, 1.1 NaH2PO4, 1.25 CaCl and 5.5 glu-
cose; in mM) at 37 8C (pH 7.35), which was gassed with 95 %
O2 : 5 % CO2. Intramural nerves were electrically stimulated (20
V, 4 ms pulse width, 0.5 ± 16 Hz) via platinum wire electrodes
placed either side of the fundus strip. Stimulus duration was
30 s except for the time-course measurements when a longer
(120 s) period was used. The Krebs-Ringer solution in the baths
contained guanethidine (5 mM) and atropine (3 mM) throughout
the duration of the experiment to block adrenergic and choliner-
gic responses to electrical stimulation. Under these conditions, 495
after precontraction with 10 mM 5-hydroxytryptamine, NANC fi-
bre-mediated relaxation responses were isolated. Further re-
sponses were examined at an optimal frequency (8 Hz) in the
presence of the NO synthase inhibitor NG-nitro-L-arginine (L-
NNA, 100 mM) without or with the peptidase a-chymotrypsin (1
U/mL) to abolish vasoactive intestinal polypeptide (VIP)-medi-
ated responses.
Renal artery experiments
The left renal artery was cleared of connective tissue and two 1-
mm sections were mounted as ring preparations using 40-mm
tungsten wire, into 5-mL organ baths in a small-vessel myograph
for measurement of isometric tension. The vessels were bathed
in modified Krebs-Ringer solution as for gastric fundus prepara-
tion. Resting tension was maintained at 0.3 g. Tissues were left to
equilibrate for 1 h, with frequent changing of the bathing solu-
tion. Tissue viability was assessed with a priming 3 mM (diabetic
groups) or 10 mM (non-diabetic group) phenylephrine and 1 mM
acetylcholine contraction/relaxation cycle. Following a 30 ± 45
min washout and recovery, cumulative concentration-response
curves to phenylephrine, and acetylcholine and sodium nitro-
prusside (against an approximate 80 % maximal phenylephrine
precontraction), were determined. Acetylcholine-mediated re-
laxation curves were repeated following a 30-min incubation
period with 10 mM of the cyclooxygenase inhibitor flurbiprofen
and 10 mM L-NNA to assess endothelium-dependent hyperpolar-
ising factor (EDHF)-mediated responses.
Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500
 Statistical analysis
Data are expressed as means  S.E.M. They were subjected to Bart-
lett's test for homogeneity of variances before one-way analysis
of variance. Where significance was reached (P < 0.05), between
groups differences were established using the Newman-Keuls or
Bonferroni multiple comparison tests. Otherwise, data were ana-
lysed by Kruskal-Wallis non-parametric one-way analysis of var-
iance and Dunn's multiple comparison test. Concentration-re-
sponse curves were fitted by sigmoid curves using the least
squares method to calculate EC50. Whole curve comparisons
were made using two-way ANOVA. Within-group serial compar-
isons were made using paired Student's t-tests. All calculations
used a standard statistics software package (Prism4, Graphpad,
San Diego, CA, USA).
Original Paper

Results

Diabetes caused an approximate 3.8-fold increase (P < 0.001) in

plasma glucose concentrations (Table 1). Additionally, there was
an approximately 25 % body weight loss (P < 0.001). These dia-
betes-induced changes were not affected by eugenol treatment.

Downloaded by: Karolinska Institutet. Copyrighted material.

Diabetes reduced (P < 0.001) motor and sensory NCVs by 21.6 
1.4 % and 19.4  1.3 % respectively (Fig. 1). Eugenol treatment
caused dose-dependent correction, which was complete for the
highest dose used (200 mg/kg) for motor NCV, and for doses of
150 mg/kg and above for sensory NCV. ED50 values were approxi-
mately 28 mg/kg for motor NCV and 9 mg/kg for sensory NCV,
the latter being significantly (P < 0.001) lower than the former.

Sciatic nerve nutritive endoneurial blood flow (Table 2) was 49 %

496 reduced by diabetes, and this deficit was completely corrected by
high-dose (200 mg/kg) eugenol treatment. Mean systemic blood
pressure was diminished by diabetes, irrespective of eugenol
treatment. The sciatic nerve does not show appreciable flow au-
toregulation when blood pressure varies [13], therefore, the data
are also presented as vascular conductance to compensate for
between-group pressure differences. A 29 % diabetic deficit was
more than completely corrected by eugenol, conductance being
29 % supernormal.
Fig. 1 Dose-response curves for correction of sciatic nerve motor (A)
and saphenous nerve sensory (B) conduction velocity by eugenol treat-
ment. Data are mean  S.E.M. Diabetic rats were treated daily for 2
weeks with doses of eugenol of 5.3 (n = 6), 16 (n = 6), 50 (n = 6)
and 200 (n = 12) mg/kg. Upper and lower dashed lines are the envel-
ope of the mean  S.E.M. for groups of non-diabetic and diabetic con-
trols (n = 8). ** P < 0.01, *** P < 0.001 vs. non-diabetic control group;
² P < 0.05, ²²² P < 0.001, eugenol treatment effect vs. diabetic control
group.

For the gastric fundus, the contractile response to 5-hydroxy-

tryptamine did not differ significantly between groups. Electrical
field stimulation (30 s duration) produced a frequency-depen- relaxation (16 Hz) was 44.1  5.7 % reduced (P < 0.001) and this
dent relaxation in gastric fundus strips (Fig. 2). However, with was 69.0  12.5 % corrected (P < 0.001) by eugenol treatment. Im-
diabetes, the relaxations were attenuated (1 ± 16 Hz). Maximal provements were also observed at 2 Hz and above, although re-

Table 1 Body weights and plasma glucose concentrations of the groups studied

Group n Body weight (g) Plasma glucose

Start End (mmol L±1)

Non-diabetic 20 489  6 8.5  1.0

Diabetic 22 490  4 367  9 41.7  1.5
Diabetic + eugenol 44 494  4 371  7 40.9  1.4

Data are mean  SEM.

For body weights, start refers to the time of induction of diabetes, and end refers to the time of final measurements.

Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500
Table 2 Sciatic nutritive endoneurial blood flow for the groups studied
Group n Blood flow
(mL/min/100 g)
Non-diabetic 8 16.90  0.76
Diabetic 8 8.63  0.61 ***
Diabetic + eugenol 7 17.26  1.22 ²²²
Data are mean  SEM. Statistics; * P < 0.05, ** P < 0.01, *** P < 0.001 vs. non-diabetic group; ²²²P < 0.001 effects of 200 mg/kg eugenol treatment vs. diabetic group.
laxation remained partially attenuated in the lower frequency
range compared to the non-diabetic group.
L-NNA (100 mM) preincubation attenuated relaxation responses
to 8 Hz electrical stimulation in all groups (Fig. 3; P < 0.01) and
a-chymotrypsin co-incubation tended to exacerbate this, al-
though any differences were not statistically significant. Particu-
larly with joint L-NNA/a-chymotrypsin incubation, there was a
tendency to reveal contractile responses in the diabetic groups
which were not apparent in non-diabetic tissues. Eugenol did
not significantly alter this response pattern.
Longer stimulus duration (120 s; Fig. 4) revealed that tissues
from diabetic rats did not maintain relaxation as well as those
from non-diabetic control rats (P < 0.001), being severely attenu-
ated after 30 s. Eugenol treatment partially reversed (P < 0.05)
this tendency, however, relaxations remained depressed com-
pared to those from the non-diabetic group (P > 0.001).
Maximum renal artery endothelium-dependent relaxation to
acetylcholine (Fig. 5A), following phenylephrine precontraction,
was 51 % reduced by diabetes compared to non-diabetic controls
(34.1  4.2 % vs. 69.3  5.5 %, P < 0.001). Treatment with eugenol
reversed the diabetic deficit by approximately 60 % (55.2  5.4 %,
Fig. 2 Frequency-response curves for electrical field stimulation of 5-
hydroxytryptamine-precontracted gastric fundus strips from non-dia-
betic and diabetic rats in the presence of atropine and guanethidine,
and the effects of intervention eugenol treatment. Groups: non-diabet-
ic control (*, n = 11); 8 week diabetic control (l, n = 10); diabetic
treated eugenol for 2 weeks following 6 weeks of untreated diabetes
(n, n = 14). Data are mean  S.E.M. ** P < 0.01, *** P < 0.001 vs.
non-diabetic control group; ²² P < 0.01, ²²² P < 0.001, eugenol treat-
ment effect vs. diabetic control group.
Mean systemic blood Vascular conductance
pressure (mm Hg) (mL/min/100 g/mm Hg)
138.1  5.1 0.123  0.006
101.7  7.1 ** 0.087  0.006 **
111.6  7.1 ** 0.159  0.015 * ²²²
P < 0.01) such that relaxations did not significantly differ from
the non-diabetic value for any individual data point, although
Original Paper
two-way ANOVA revealed a significant (P < 0.001) whole curve
difference compared to the non-diabetic control group. Diabetes
tended to reduce sensitivity to acetylcholine; however, statistical
significance was not reached. Thus, (-log) EC50 values were 6.64 
0.11 and 6.21  0.12 for non-diabetic and diabetic controls and
6.50  0.12 for eugenol-treated diabetic tissues, respectively.
After L-NNA and flurbiprofen preincubation, to isolate the EDHF
Downloaded by: Karolinska Institutet. Copyrighted material.
component, maximum relaxation to acetylcholine from non-dia-
betic controls (Fig. 5B) was reduced by approximately 58 % com-
pared to relaxations prior to preincubation (29.0  4.3 %, P <
0.001). A more marked reduction of approximately 89 % was ob-
served in the diabetic control group (3.8  1.9 %, P < 0.001) such
that there was an 87 % deficit (P < 0.001) compared to the non-
diabetic control value. With eugenol treatment, EDHF-depen-
dent relaxation was reduced approximately 73 % compared to
acetylcholine-mediated relaxation prior to flurbiprofen and L-
NNA (15.1  3.5 %, P < 0.001), corresponding to correction of the
diabetic deficit by approximately 45 % (P < 0.01). However, re- 497
Fig. 3 Response to electrical field stimulation at 8 Hz of 5-HT precon-
tracted gastric fundus longitudinal muscle strips from non-diabetic
and diabetic rats in the presence of atropine and guanethidine, and
the effects of eugenol treatment: modulation by L-NNA preincubation
with or without a-chymotrypsin. Groups: non-diabetic control (n =
11); 8 week diabetic control (n = 10); diabetic treated with 200 mg/
kg/day eugenol for 2 weeks following 6 weeks untreated diabetes
(n = 12). Data presented as mean  S.E.M. Statistics: ** P < 0.01, ***
P < 0.001 vs. the corresponding conditions for the non-diabetic control
group; ²² P < 0.01, ²²² P < 0.001, effects of L-NNA or L-NNA + a-chymo-
trypsin vs. response in their absence.
Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500
 fect on multiple indices of oxidative stress and antioxidant de-
fences in tissues of streptozotocin-diabetic rats [16]. It is plausi-
ble that other effects of eugenol contributed to the efficacy with
which NCV deficits were corrected. The demonstration that eu-
genol can inhibit the tumour necrosis factor (TNF) a-nuclear fac-
tor (NF)kB cascade, at least in cell culture [9], may be pertinent.
In diabetes, the accumulation of advanced glycation end pro-
ducts (AGEs) as a result of hyperglycaemia, stimulates receptors
for AGE (RAGE), which activates NFkB and increases TNFa secre-
tion. Inhibition of AGE formation improves neurovascular func-
tion in diabetic rats [17], [18], [19] and in RAGE knockout mice
diabetes causes less nerve dysfunction [20]. Treatment with N-
acetylcysteine reduces elevated circulating levels of TNF in dia-
betic rats, and improves motor NCV [21]; nerve blood flow and
Original Paper

Fig. 4 Response to prolonged electrical field stimulation (120-s train) function are corrected by NFkB inhibition [22].
at 8 Hz of 5-hydroxytryptamine precontracted gastric fundus longitu-
dinal muscle strips from non-diabetic and diabetic rats in the presence
of atropine and guanethidine, and the effects of intervention eugenol
treatment. Groups: non-diabetic control (*, n = 8); 8 week diabetic
control (l, n = 9); diabetic treated with eugenol for 2 weeks following
6 weeks untreated diabetes (n, n = 12). Data are mean  S.E.M. Statis-
tics for whole curve: diabetic controls and eugenol treated (P < 0.001)
vs. non-diabetic controls; eugenol treated (P < 0.05) vs. diabetic con-
trols.

Downloaded by: Karolinska Institutet. Copyrighted material.

laxation remained significantly depressed compared to non-dia-
betic controls (P < 0.001).

In contrast, endothelium-independent relaxation to sodium ni-

troprusside was not altered by diabetes or treatment. Thus, max-
imum relaxations were 62.5  4.1 % (n = 10) and 63.2  7.5 %
(n = 11) for non-diabetic and diabetic control groups, and 64.1
498  5.5 % (n = 7) for the eugenol treated group, respectively: (±log)
EC50 values were 6.95  0.07 for non-diabetic, 6.99  0.15 for dia-
betic control, and 7.22  0.10 for eugenol-treated diabetic groups.

Diabetes caused a marked increase in sensitivity of renal artery

to phenylephrine-mediated contraction (Fig. 6) by greater than
0.6 log mol L±1 units compared to non-diabetic controls. This
was not significantly altered by eugenol treatment. Maximum
tensions were not altered by diabetes or treatment, being 13.75
 0.96, 16.28  1.33 and 13.61  2.37 mN for non-diabetic, diabet-
ic and eugenol-treated diabetic arteries, respectively.

Discussion

The data show the deleterious effects of diabetes on nerve and

vascular function in diabetic rats, in agreement with numerous
previous investigations [2], [3]. Eugenol had marked beneficial
neurovascular effects in diabetes. Thus, motor and sensory NCV
deficits were corrected in a dose-dependent manner and high-
dose treatment completely reversed the defect in nerve blood
flow. It is possible that the antioxidant properties of eugenol are
sufficient to explain these effects. However, the ED50 values for
motor (28 mg/kg) and sensory NCV (9 mg/kg) were surprisingly
low. For comparison, treatment with the lipophilic antioxidant,
vitamin E, has an ED50 of 620 mg/kg for motor NCV in our model
[14]. In addition, isoeugenol, which is a more potent antioxidant
than eugenol [15], at a dose of 10 mg/kg had a rather modest ef-
Fig. 5 Cumulative concentration-response curves for relaxation to
acetylcholine before (A) and after (B) cyclooxygenase and nitric oxide
synthase inhibition, of phenylephrine preconstricted renal arteries
from non-diabetic and diabetic rats, and the effects of eugenol treat-
ment. Groups: non-diabetic control (*: A, n = 12; B, n = 10); 8 week
diabetic control (l: A, n = 14; B, n = 13); diabetic treated with
200 mg/kg/day eugenol for 2 weeks following 6 weeks untreated dia-
betes (n: A, n = 10; B, n = 8). Data are mean  S.E.M. ** P < 0.01, ***
P < 0.001 vs. non-diabetic control group; ² P < 0.05, ²² P < 0.01, ²²² P <
0.001, eugenol treatment effect vs. diabetic control group.

Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500

Fig. 6 Cumulative concentration-response curves and (±log) EC50 val-
ues (inset) for contraction to phenylephrine for renal arteries from non-
diabetic and diabetic rats, and the effects of eugenol treatment.
Groups: non-diabetic control (*, C, n = 14); 8 week diabetic control
(l, D, n = 13); diabetic treated with eugenol for 2 weeks following 6
weeks untreated diabetes (n, EUG, n = 6). Data are mean  S.E.M. Sta-
tistics for (-log) EC50, ** P < 0.01 vs. non-diabetic control group.
The vascular actions of eugenol are likely to have made a major
contribution to the beneficial effects on nerve function [1]. The
diabetes-induced reduction in NO and particularly EDHF-medi-
ated endothelium-dependent vasorelaxation for renal artery is
in agreement with previous findings on renal and other vascular
beds [23], [24], [25], [26]. This is the first report that short-term
eugenol treatment partially corrects both NO and EDHF-medi-
ated deficits in renal arteries from diabetic rats. Results for so-
dium nitroprusside revealed no diabetic or eugenol treatment ef-
fect on endothelium-independent relaxation, indicating that
vascular smooth muscle sensitivity to NO was unaltered. Pre-
vious studies on eugenol have examined acute vascular effects
in non-diabetic animals. From experiments on isolated vascular
preparations there is evidence of vasodilation by an endothe-
lium-dependent mechanism involving NO and an endothelium
independent effect directly on vascular smooth muscle [11],
[27]. In vivo, bolus i. v. injection of eugenol produced a rapidly de-
veloping hypotension [28]. It is not clear whether these acute ef-
fects contribute to the longer-term actions noted for endothe-
lium-dependent relaxation in renal artery of diabetic rats as the
tissues were not exposed to eugenol under the in vitro measure-
ment conditions. However, a vasodilator action, as noted in vivo,
could contribute to improved nerve blood flow and hence NCV.
Thus, treatment with several classes of conventional vasodilators
increases NCV in diabetic rats [1]. Eugenol treatment gave rise to
supernormal vasa nervorum vascular conductance values, em-
phasising the marked vascular effect of eugenol, which is com-
parable to a similar phenomenon noted for treatment with po-
tent vasodilators. This would not be expected to give rise to su-
pranormal or aberrant indices of nerve function as these meas-
ures are not normally flow-limited provided that the normal per-
fusion range is at least attained [1]. The possibility exists that in-
creased tissue blood flow resulting from the acute vasodilatory
actions of eugenol could improve NO and EDHF systems as a re-
sult of upregulation due to chronic flow-dependent endothelial
stimulation [29].
There was a marked increase in renal artery sensitivity to phenyl-
ephrine with diabetes that was not altered by chronic eugenol
treatment. In contrast, acute experiments on rat aorta showed
that eugenol depressed phenylephrine-induced contractile re-
sponses [11]. A possible explanation for this discrepancy is that
appreciable concentrations of eugenol were unlikely to be pres-
ent in the tissue bath during renal artery measurements, and
that this diabetic deficit was resistant to eugenol treatment.
Original Paper
However, a depression of adrenergic contractility in vivo could
have contributed to the beneficial actions on eugenol on nerve
perfusion and function.
In addition to the endothelial NO system, both diabetes and eu-
genol treatment had effects on neuronal NO. Thus, the data con-
firm previous findings that NO is a major NANC neurotransmitter
in rat fundus and that NANC-mediated relaxation is depressed by
Downloaded by: Karolinska Institutet. Copyrighted material.
diabetes [4], [7]. This is the first report of the partial correction of
impaired gastric fundus NANC relaxations by eugenol treatment.
Experiments using preincubation with L-NNA to block the NO
component showed that eugenol's effect was completely abol-
ished. For tissues from non-diabetic rats, L-NNA treatment re-
vealed the second smooth muscle relaxing component, based
on VIP. This was largely abolished by diabetes and the defect
was not altered by eugenol treatment. When the VIPergic com-
ponent was also removed by a-chymotrypsin, a third residual re-
laxation component of unknown origin was noted in non-diabet-
ic tissues [4], [7]. For diabetic fundus, a contractile component 499
was revealed [7], which was not significantly affected by eugenol
treatment. Longer electrical stimulation durations showed that
NANC nerve-mediated relaxation could not be maintained for
diabetic fundus. The VIP component becomes progressively
more important under these conditions; eugenol treatment had
a relatively modest effect, presumably because it did not im-
prove VIPergic effects. Thus, the benefits of eugenol treatment
appear restricted to nitrergic neurotransmission. Previous stud-
ies on the nitrergic innervation in diabetic rats have demonstrat-
ed the protective effects of antioxidant treatment [5], [7]. For
fundus, a-lipoic acid acted like eugenol in that the main effects
were on the NO system. Thus, the mechanism of eugenol's action
could have derived from its antioxidant properties. However, it is
also plausible that vasodilation by eugenol could have improved
perfusion of fundus nerves and autonomic ganglia [30], which
may contribute to partial correction of defective neurotransmis-
sion.
In conclusion, treatment of diabetic rats with a dietary supple-
ment of eugenol resulted in beneficial effects on nerve and vas-
cular endothelium function. This was possibly due to the antiox-
idant and anti-inflammatory properties of eugenol. The NO sys-
tem appears to be a particular target, and effects on neural tissue
blood flow are likely to make a major contribution to improved
nerve function.
Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500
 References
1
Cameron NE, Eaton SEM, Cotter MA, Tesfaye S. Vascular factors and
metabolic interactions in the pathogenesis of diabetic neuropathy.
Diabetologia 2001; 44: 1973 ± 88
2
Low PA. Diabetic autonomic neuropathy. Semin Neurol 1996; 16:
143 ± 51
3
Vinik AI, Maser RE, Mitchell BD, Freeman R. Diabetic autonomic
neuropathy. Diabetes Care 2003; 26: 1553 ± 79
4
Jenkinson KM, Reid JJ. Effects of diabetes on relaxations to non-adre-
nergic non-cholinergic nerve stimulation in longitudinal muscle of
the rat gastric fundus. Br J Pharmacol 1995; 116: 1551 ± 6
5
Keegan A, Cotter MA, Cameron NE. Effect of diabetes and treatment
with the antioxidant, a-lipoic acid, on endothelial and neurogenic re-
sponses on corpus cavernosum in rats. Diabetologia 1999; 42: 343 ±
50
6
Cellek S, Foxwell NA, Moncada S. Two phases of nitrergic neuropathy
Original Paper
in streptozotocin-induced diabetic rats. Diabetes 2003; 52: 2353 ± 62
7
Gibson TM, Cotter MA, Cameron NE. Effects of a-lipoic acid on im-
paired gastric fundus innervation in diabetic rats. Free Radic Biol
Med 2003; 35: 160 ± 8
8
Fujisawa S, Atsumi T, Kadoma Y, Sakagami H. Antioxidant and prooxi-
dant action of eugenol-related compounds and their cytotoxicity. Tox-
icology 2002; 177: 39 ± 54
9
Chainy GB, Manna SK, Chaturvedi MM, Aggarwal BB. Anethole blocks
both early and late cellular responses transduced by tumor necrosis
factor: effect on NF-kappaB, AP-1, JNK, MAPKK and apoptosis. Onco-
gene 2000; 19: 2943 ± 50
10
Kim SS, Oh OJ, Min HY, Park EJ, Kim Y, Park HJ et al. Eugenol suppresses
cyclooxygenase-2 expression in lipopolysaccharide-stimulated mouse
macrophage RAW264.7 cells. Life Sci 2003; 73: 337 ± 48
11
Damiani CEN, Rossoni LV, Vassallo DV. Vasorelaxant effects of eugenol
on rat thoracic aorta. Vascul Pharmacol 2003; 40: 59 ± 66
12
Cameron NE, Cotter MA, Hohman TC. Interactions between essential
fatty acid, prostanoid, polyol pathway and nitric oxide mechanisms
in the neurovascular deficit of diabetic rats. Diabetologia 1996; 39:
172 ± 82
13
Low PA, Lagerlund TD, McManis PG. Nerve blood flow and oxygen de-
livery in normal, diabetic and ischemic neuropathy. Int Rev Neurobiol
1989; 31: 355 ± 438
500 14
Cotter MA, Love A, Watt MJ, Cameron NE, Dines KC. Effects of natural
free radical scavengers on peripheral nerve and neurovascular func-
tion in diabetic rats. Diabetologia 1995; 38: 1285 ± 94
15
Priyadarsini KI, Guha SN, Rao MNA. Physico-chemical properties and
antioxidant activities of methoxy phenols. Free Radic Biol Med 1998;
24: 933 ± 41
16
Rauscher FM, Sanders RA, Watkins JB. Effects of isoeugenol on oxida-
tive stress pathways in normal and streptozotocin-induced diabetic
rats. J Biochem Mol Toxicol 2001; 15: 159 ± 64
17
Kihara M, Schmelzer JD, Poduslo JF, Curran GL, Nickander KK, Low PA.
Aminoguanidine effect on nerve blood flow, vascular permeability,
Nangle MR et al. Effects of Eugenol ¼ Planta Med 2006; 72: 494 ± 500
electrophysiology, and oxygen free radicals. Proc Natl Acad Sci USA
1991; 88: 6107 ± 11
18
Cameron NE, Cotter MA. Rapid reversal by aminoguanidine of the neu-
rovascular effects of diabetes in rats: modulation by nitric oxide syn-
thase inhibition. Metabolism 1996; 45: 1147 ± 52
19
Archibald V, Cotter MA, Keegan A, Cameron NE. Contraction and re-
laxation of aortas from diabetic rats: effects of chronic anti-oxidant
and aminoguanidine treatments. Naunyn Schmeidebergs Arch Phar-
macol 1996; 353: 584 ± 91
20
Bierhaus A, Haslbeck KM, Humpert PM, Liliensiek B, Dehmner T, Mor-
cos M et al. Loss of pain perception in diabetes is dependent on a re-
ceptor of the immunoglobulin superfamily. J Clin Invest 2004; 114:
1741 ± 51
21
Sagara M, Satoh J, Wada R, Yagihashi S, Takahashi K, Fukuzawa M et al.
Inhibition of development of peripheral neuropathy in streptozotocin-
induced diabetic rats with N-acetylcysteine. Diabetologia 1996; 39:
263 ± 9
22
Cotter MA, Gibson TM, Cameron NE. Nuclear factor kappa B inhibition
improves nerve function in diabetic rats. Diabetologia 2003; 46 Suppl
2: A70
23
Kamata K, Hosokawa M. Endothelial dysfunction in the perfused kid-
ney from the streptozotocin-induced diabetic rat. Res Commun Mol
Pathol Pharmacol 1997; 96: 57 ± 70
24
De Vriese AS, van de Voorde J, Blom HJ, Vanhoutte PM, Verbeke M, La-
meire NH. The impaired renal vasodilator response attributed to en-
dothelium-derived hyperpolarizing factor in streptozotocin-induced
diabetic rats is restored by 5-methyltetrahydrofolate. Diabetologia
2000; 43: 1116 ± 25
Downloaded by: Karolinska Institutet. Copyrighted material.
25
Keegan A, Jack AM, Cameron NE, Cotter MA. Effects of aldose reduc-
tase inhibition on responses of the corpus cavernosum and mesenteric
vascular bed of diabetic rats. J Cardiovasc Pharmacol 2000; 35: 606 ±
13
26
Nangle MR, Cotter MA, Cameron NE. Effects of rosuvastatin on nitric
oxide-dependent function in aorta and corpus cavernosum of diabetic
mice: relationship to cholesterol biosynthesis pathway inhibition and
lipid lowering. Diabetes 2003; 52: 2396 ± 402
27
Criddle DN, Madeira SVF, de Moura RS. Endothelium-dependent and
-independent vasodilator effects of eugenol in the rat mesenteric vas-
cular bed. J Pharm Pharmacol 2003; 55: 359 ± 65
28
Lahlou S, Interaminense LFL, Magalhaes PJC, Leal-Cardoso JH, Duarte
GP. Cardiovascular effects of eugenol, a phenolic compound present
in many plant essential oils, in normotensive rats. J Cardiovasc Phar-
macol 2004; 43: 250 ± 7
29
Fujii K, Heistad DD, Faraci FM. Effects of diabetes mellitus on flow
mediated and endothelium-dependent dilation of the rat basilar ar-
tery. Stroke 1992; 23: 1494 ± 8
30
Cameron NE, Cotter MA. Diabetes causes an early reduction in auto-
nomic ganglion blood flow in rats. J Diabet Complications 2001; 15:
198 ± 202