Coenzyme Q(10) improves endothelial dysfunction of the brachial artery in Type II diabetes mellitus

Abstract

We assessed whether dietary supplementation with coenzyme Q(10) improves endothelial function of the brachial artery in patients with Type II (non-insulin-dependent) diabetes mellitus and dyslipidaemia.
A total of 40 patients with Type II diabetes and dyslipidaemia were randomized to receive 200 mg of coenzyme Q(10) or placebo orally for 12 weeks. Endothelium-dependent and independent function of the brachial artery was measured as flow-mediated dilatation and glyceryl-trinitrate-mediated dilatation, respectively. A computerized system was used to quantitate vessel diameter changes before and after intervention. Arterial function was compared with 18 non-diabetic subjects. Oxidative stress was assessed by measuring plasma F(2)-isoprostane concentrations, and plasma antioxidant status by oxygen radical absorbance capacity.
The diabetic patients had impaired flow-mediated dilation [3.8 % (SEM 0.5) vs 6.4 % (SEM 1.0), p = 0.016], but preserved glyceryl-trinitrate-mediated dilation, of the brachial artery compared with non-diabetic subjects. Flow-mediated dilation of the brachial artery increased by 1.6 % (SEM 0.3) with coenzyme Q(10) and decreased by -0.4 % (SEM 0.5) with placebo (p = 0.005); there were no group differences in the changes in pre-stimulatory arterial diameter, post-ischaemic hyperaemia or glyceryl-trinitrate-mediated dilation response. Coenzyme Q(10) treatment resulted in a threefold increase in plasma coenzyme Q(10) (p < 0.001) but did not alter plasma F(2)-isoprostanes, oxygen radical absorbance capacity, lipid concentrations, glycaemic control or blood pressure.
Coenzyme Q(10) supplementation improves endothelial function of conduit arteries of the peripheral circulation in dyslipidaemic patients with Type II diabetes. The mechanism could involve increased endothelial release and/or activity of nitric oxide due to improvement in vascular oxidative stress, an effect that might not be reflected by changes in plasma F(2)-isoprostane concentrations.

Full text

In Type II (non-insulin-dependent) diabetes mellitus
the pathogenesis of vascular disease, its most com-
mon complication, remains unclear [1]. Endothelial
dysfunction reflects the disordered physiology of sev-
eral endothelium-derived vasoactive factors, in par-
ticular nitric oxide [2]. Endothelial dysfunction oc-
curs commonly in diabetes and is an early feature of
vasculopathy [3, 4]. Increased oxidative stress due to
the effects of hyperglycaemia and its sequelae is a
recognized feature of diabetes [5]. It might cause en-
dothelial dysfunction through the inactivation and
Diabetologia (2002) 45: 420±426
Coenzyme Q
10
improves endothelial dysfunction of the brachial
artery in Type II diabetes mellitus
G. F.Watts, D.A. Playford, K. D. Croft, N. C. Ward, T. A. Mori, V. Burke
Department of Medicine, University of Western Australia, Royal Perth Hospital, Perth, Australia
Ó Springer-Verlag 2002
Abstract
Aim/hypothesis. We assessed whether dietary supple-
mentation with coenzyme Q
10
improves endothelial
function of the brachial artery in patients with Type
II (non-insulin-dependent) diabetes mellitus and dys-
lipidaemia.
Methods. A total of 40 patients with Type II diabetes
and dyslipidaemia were randomized to receive
200 mg of coenzyme Q
10
or placebo orally for
12 weeks. Endothelium-dependent and independent
function of the brachial artery was measured as flow-
mediated dilatation and glyceryl-trinitrate-mediated
dilatation, respectively. A computerized system was
used to quantitate vessel diameter changes before
and after intervention. Arterial function was com-
pared with 18 non-diabetic subjects. Oxidative stress
was assessed by measuring plasma F
2
-isoprostane
concentrations, and plasma antioxidant status by oxy-
gen radical absorbance capacity.
Results. The diabetic patients had impaired flow-me-
diated dilation [3.8 % (SEM 0.5) vs 6.4% (SEM 1.0),
p = 0.016], but preserved glyceryl-trinitrate-mediated
dilation, of the brachial artery compared with non-di-
abetic subjects. Flow-mediated dilation of the brachi-
al artery increased by 1.6% (SEM 0.3) with coen-
zyme Q
10
and decreased by ±0.4% (SEM 0.5) with
placebo (p = 0.005); there were no group differences
in the changes in pre-stimulatory arterial diameter,
post-ischaemic hyperaemia or glyceryl-trinitrate-me-
diated dilation response. Coenzyme Q
10
treatment re-
sulted in a threefold increase in plasma coenzyme Q
10
(p < 0.001) but did not alter plasma F
2
-isoprostanes,
oxygen radical absorbance capacity, lipid concentra-
tions, glycaemic control or blood pressure.
Conclusion/interpretation. Coenzyme Q
10
supple-
mentation improves endothelial function of conduit
arteries of the peripheral circulation in dyslipidaemic
patients with Type II diabetes. The mechanism could
involve increased endothelial release and/or activity
of nitric oxide due to improvement in vascular oxida-
tive stress, an effect that might not be reflected by
changes in plasma F
2
-isoprostane concentrations.
[Diabetologia (2002) 45: 420±426]
Keywords Coenzyme Q
10
, endothelial function, nitric
oxide, diabetes.
Received: 14 August 2001 and in revised form: 15 November
2001
Corresponding author: G. F. Watts, PhD, MD, Department of
Medicine, University of Western Australia, Royal Perth Hos-
pital, GPO Box X2213, Perth, WA 6847, Australia.
e-mail: gfwatts@cyllene.uwa.edu.au
Abbreviations: CoQ, Coenzyme Q
10
; FMD, flow-mediated di-
latation; NMD, glyceryl-trinitrate-mediated dilatation;
ORAC, oxygen radical absorbance capacity

decreased synthesis of nitric oxide by reactive oxygen
species [6] such as superoxide.
While epidemiology suggests that conventional
antioxidant vitamins can benefit vascular disease,
the evidence from controlled clinical trials is less se-
cure [7], especially in diabetes [8]. Vitamin E supple-
mentation has not consistently been shown to im-
prove endothelial-dependent vasodilator tone in dia-
betic patients [9, 11]. Moreover, with vitamin C, the
only study reported in Type II diabetes was not place-
bo-controlled and involved acute intra-arterial ad-
ministration of this antioxidant [12].
Coenzyme Q
10
(CoQ) is a critical intermediate of
the mitochondrial electron transport chain that regu-
lates cytoplasmic redox potential and it can inhibit
superoxide generation by endothelial cells [13, 14].
CoQ is a more powerful antioxidant than vitamin E,
inhibiting its pro-oxidant activity [15, 16]. Its deficien-
cy can occur in diabetes in relation to impaired mito-
chondrial substrate metabolism [17] and increased
oxidative stress [6]. Mitochondrial CoQ deficiency
could be involved in the pathogenesis of Type II dia-
betes by impairing beta-cell function [18]. Low serum
CoQ concentrations have been negatively correlated
with poor glycaemic control and diabetic complica-
tions [19, 20]. Accordingly, some clinical trials have
shown that CoQ could improve glycaemic control
and blood pressure in diabetes [14, 18, 21, 22]. Hence,
CoQ could have a potential role in the treatment of
diabetes and its complications [14, 18, 23]. However,
to date no studies have reported the effects of CoQ
on vascular dysfunction or cardiovascular disease in
diabetes.
We report a placebo-controlled observation of the
effect of CoQ supplementation on vascular function
of the peripheral circulation in Type II diabetic pati-
ents. Endothelial function was quantitated as postis-
chaemic flow-mediated dilatation of the brachial ar-
tery using a new edge-detection software system that
increases the precision of measurements [24]. We
also measured potential changes in oxidative stress
by measuring plasma F
2
-isoprostanes and measuring
the oxygen radical absorbance capacity of plasma.
Subjects and methods
Subjects. A total of 40 patients with Type II diabetes diagnosed
by standard criteria and with dyslipidaemia were recruited
from the community. Dyslipidaemia was defined as a fasting
serum triglyceride of greater than 1.8 mmol/l or HDL choles-
terol of less than 1.0 mmol/l with total cholesterol of less than
6.5 mmol/l and a total cholesterol-to-HDL cholesterol ratio of
more than 4. Patients were excluded based on the following
criteria: age older than 75 years, BMI greater than 40 kg/m
2
,
history of myocardial infarction or stroke, insulin therapy,
smoking, macroalbuminuria, serum creatinine greater than
150 mmol/l, liver abnormalities, use of antioxidants or lipid-
regulating therapy, uncontrolled hypertension ( > 160/
90 mmHg), and treatment with angiotensin-converting en-
zyme inhibitors, calcium antagonists or aspirin. Volunteers un-
derwent a clinical examination, urinalysis and a 12-lead ECG.
The vascular function in the diabetic patients was compared
with 18 healthy, non-diabetic normolipidaemic subjects of sim-
ilar age [mean age 54 years (SD 12.0); cholesterol 5 mmol/l
(SD 0.4), HDL cholesterol 1.6 (0.3), triglyceride 0.9 (95 % CI
0.8, 1.1)].
Study design. This study is part of a larger study examining the
effects of CoQ and lipid-regulating therapies on vascular func-
tion of peripheral arteries measured by several techniques. We
report on the effect of CoQ monotherapy on postischaemic di-
lation of the brachial artery. Eligible patients entered a run-in
period of 6 weeks during which they were instructed to con-
sume an isocaloric fat-modified diet of constant antioxidant
composition. They then underwent the brachial artery reactiv-
ity test described below, after which they were randomized
double-blind to treatment with either CoQ (Blackmores, Syd-
ney, NSW, Australia) or matching placebo in a trial of
12 weeks duration. Coenzyme Q
10
(200 mg) was taken as two
50 mg capsules orally twice a day. Volunteers were interviewed
every 2 weeks to assess their compliance with therapeutic units
and brachial artery reactivity was re-studied 12 weeks after
randomization. The Ethics Committee of the Royal Perth
Hospital approved the study and all volunteers gave their writ-
ten consent.
Laboratory methods. Venous blood was collected after a 12 h
fast at baseline and at 12 weeks. Serum total cholesterol, tri-
glyceride and HDL cholesterol were measured using enzymat-
ic, colorimetric methods (Boehringer Mannheim, Mannheim,
Germany) on a Hitachi 917 biochemical analyser (Hitachi, To-
kyo, Japan). High-density lipoprotein cholesterol was mea-
sured after precipitation of apolipoprotein B-100 (apoB) con-
taining lipoproteins with dextran sulphate. Low-density lipo-
protein cholesterol was estimated by the Friedewald formula
and by a direct assay when triglycerides were more than
3.5 mmol/l. The particle size of LDL was estimated by non-de-
naturing gel electrophoresis. Glycated haemoglobin (HbA
1c
)
was measured by high performance liquid-chromatography
(HPLC, BioRad Laboratories, Sydney, Australia). Plasma glu-
cose and insulin were assayed using an enzymatic method
(Boehringer) and an automated immuno-enzymometric assay
(Tosoh, Kyobashi, Tokyo, Japan), respectively. Serum and uri-
nary creatinine were measured by the modified Jaffe reaction.
Total serum CoQ concentration was assayed by reverse-phase
high performance liquid-chromatography using electrochemi-
cal detection [25]. Plasma F
2
-isoprostanes were analysed using
gas-chromotography mass-spectrometry with electron capture
negative chemical ionization [26]. Plasma oxygen radical ab-
sorbance capacity (ORAC) was measured using a fluorescent
assay using a Trolox standard [27].
Brachial artery ultrasonography. Brachial artery ultrasonogra-
phy was carried out [28]. Briefly, a 12-megahertz transducer
connected to an Acuson Aspen ultrasound (Acuson, Mountain
View, Calif, USA) and fixed in position by a stereotactic clamp
5 to 10 cm proximal to the ante-cubital crease was used to im-
age the brachial artery. Images were recorded before and after
vasodilatory stimuli and recorded on s-VHS videotape (Sony
MQSE 180). Continuous ECG monitoring was done in all
studies. Reactive hyperaemia of the brachial artery was in-
duced after release of a pneumatic tourniquet placed around
the left forearm and inflated to 50 mmHg above systolic blood
pressure for 5 min. Pulse wave Doppler flow velocities were
used to derive flow rate (ml/min) pre-reactive and post-reac-
G. F. Watts et al.: CoQ improves endothelial function in diabetes 421

tive hyperaemia. After the brachial artery diameter returned
to baseline 400 mg glyceryl trinitrate was administered sublin-
gually to assess endothelium-independent vasodilatory re-
sponse. All images obtained were assessed blindly by two inde-
pendent observers and only pairs of scans that were of consis-
tently acceptable quality were included in this analysis. Analy-
sis of flow-mediated dilatation (FMD) and glyceryl-trinitrate-
mediated dilatation (NMD) of the brachial artery was carried
out using a semi-automated edge detection software system
[24] and operated by an experienced observer, who was blind-
ed to the treatment group assignment. Responses were calcu-
lated as percentage change in brachial artery diameter from
baseline. The analytical (intra-observer) coefficient of varia-
tion of the computerized technique is 6.7 % compared with
32.5 % using ultrasonic callipers, a more conventional visual
estimation. The resolving power of the method tested on
`phantom arteries' is 8.3 mm.
Statistical methods. Data were analysed parametrically after
logarithmic transformation of variables where appropriate.
Discrete variables were compared by Chi-square test. Treat-
ment effects were analysed using general linear models with
adjustments for baseline values and resting brachial artery di-
ameter. Association between variables were examined by lin-
ear regression methods. Statistical significance was defined at
a p value of less than 0.05.
Results
The diabetic patients were mostly middle-aged men
and were on average, overweight, normotensive, in
good glycaemic control and had typical diabetic dys-
lipidaemia. Of the 40 patients randomized, 39 com-
pleted the study and one withdrew due to an inciden-
tal illness. High quality ultrasound images were ob-
tained in all patients randomized to CoQ. However,
satisfactory image quality was not obtained in ultra-
sound scans from four patients randomized to place-
bo. Thus, our analysis refers to the residual 35 diabet-
ic patients. Their characteristics did not differ from
those withdrawn from the study.
Plasma F
2
-isoprostanes were not different be-
tween the diabetic patients and the non-diabetic con-
trol subjects [1245 mmol/l (95 % CI 1075, 1442) vs
1310 (1085, 1583), p = 0.670]. However, plasma
ORAC was lower in the diabetics patients than in
the control subjects [3766 mmol/l (95% CI 3476,
4081) vs 4729 (4551, 4914) p = < 0.001].
Compared with the control subjects, the diabetic
patients overall had lower post-ischaemic FMD of
the brachial artery [3.8 % (SEM 0.5) vs 6.4 (1.0),
p = 0.016] but similar NMD responses [16.5 % (SEM
0.9) vs 18.5 (1.7), p = 0.291].
No differences were shown in the characteristics
(Table 1) between patients randomized to placebo
and CoQ treatment (p > 0.05), except for higher
baseline plasma CoQ concentrations (p = 0.03) in
the CoQ group. Only four patients in the study took
metformin, one in the placebo and three in the CoQ
group. The plasma glucose immediately before ultra-
sonography did not differ (p = 0.20) between the
groups (Table 1).
The changes in baseline diameter, reactive hyper-
aemia, FMD and NMD of the brachial artery in the
CoQ were compared with the placebo group (Ta-
ble 2). There was an improvement in FMD of the bra-
chial artery in the CoQ compared with the placebo
group, without changes or differences in baseline ar-
terial diameter, reactive hyperaemia or NMD. In an
analysis with absolute FMD (mm) at 12 weeks as the
dependent variable, there was still a favourable treat-
ment effect of CoQ (p = 0.002) after adjusting for
baseline brachial artery diameter (mm) at 12 weeks
and pre-randomization percentage FMD response.
This analysis together with the lack of significant
change in resting brachial artery diameter (Table 2)
demonstrates that the favourable effect of CoQ on
percentage FMD response was independent of chan-
ges in basal brachial artery tone. There were no dif-
ferences between the groups in the plasma glucose
concentration immediately before ultrasonography
post-intervention [8.2 mmol/l (SEM 0.8) for CoQ vs
7.3 mmol/l (0.7) for placebo, p = 0.43].
Treatment with CoQ was associated with an in-
crease in plasma CoQ concentrations, from
G. F. Watts et al.: CoQ improves endothelial function in diabetes422
Table 1. Clinical, biochemical and vascular characteristics of
the patients in the placebo and CoQ groups at baseline
Characteristics Placebo group CoQ group
n (male/female) 13/2 18/2
Age (years) 54.1 (10.4) 52.7 (6.2)
BMI (kg/m
2
) 31.3 (5.4) 29.9 (3.3)
SBP (mmHg) 139.1 (15.1) 128.0 (18.4)
DBP (mmHg) 81.0 (5.8) 75.8 (9.3)
Glucose (mmol/l) 6.9 (2.1) 8.2 (2.7)
HbA
1c
(%) 6.2 (0.8) 6.9 (1.4)
Insulin (mU/l) 15.1 (7.4) 12.2 (5.4)
Cholesterol (mmol/l) 5.3 (0.6) 5.3 (0.9)
Triglyceride (mmol/l) 2.5 (2.1, 3.0) 2.0 (1.7, 2.5)
HDL-cholesterol (mmol/l) 1.00 (0.08) 0.95 (0.15)
LDL-cholesterol (mmol/l) 3.2 (0.7) 3.2 (0.9)
LDL size (nm) 25.3 (0.8) 25.0 (0.8)
Coenzyme Q
10
(mmol/l) 1.2 (0.3) 1.5 (0.3)*
ORAC activity (mmol/l) 3638 (3270, 4046) 3723 (3275, 4233)
Plasma F
2
-isoprostanes
(pmol/l) 1297 (1023, 1643) 1102 (892, 1361)
Brachial artery:
Baseline diameter (mm) 4.3 (0.6) 4.2 (0.5)
Resting blood flow (ml/min) 186.6 (72.8) 205.9 (113.7)
Reactive hyperaemia (%) 462.2 (239.8) 431.9 (254.7)
Flow-mediated dilatation
(%) 4.5 (3.0) 2.8 (3.0)
Nitrate-mediated dilatation
(%) 16.9 (6.4) 16.3 (5.2)
Values are means (SD) or geometric means and 95 % CI
*p = 0.03 vs placebo group

1.3 mmol/l (SEM 0.1) to 4.8 (0.4), p < 0.001; however,
there were no alterations (p > 0.05) in plasma F
2
-iso-
prostanes, plasma ORAC activity, glucose, HbA
1C
,
plasma lipids, blood pressure or other variables (Ta-
ble 1). Changes in plasma CoQ concentrations were
not correlated with the changes in plasma F
2
-isopros-
tanes. The FMD of the brachial artery increased with
CoQ alone from 2.8 % (SEM 0.7) to 4.4 % (0.5),
p < 0.001 but the post-treatment response tended to
stay lower (p = 0.07) than the values for the non-dia-
betic control group. There was no correlation be-
tween the change in FMD and change in other vari-
ables in the CoQ group. In data pooled from both pa-
tient groups, improvement in FMD was only corre-
lated with treatment group assignment to CoQ.
Discussion
Our randomized, double-blind study shows a favour-
able effect of oral CoQ supplementation on endothe-
lial dysfunction of the peripheral circulation in pati-
ents with Type II diabetes. Coenzyme Q
10
improved
abnormal endothelium-dependent vasodilator tone
of the brachial artery without altering the vasodilato-
ry response to the endothelium-independent agonist
glyceryl trinitrate. This favourable effect of CoQ was
independent of changes in resting brachial artery di-
ameter. The improvement in endothelial function
also occurred in the presence of dyslipidaemia and
was not related to changes in plasma F
2
-isoprostanes,
glycated haemoglobin or blood pressure.
That oxidative stress is increased in diabetes is well
supported by experimental and clinical observations
[5, 6, 11, 29, 30]. Oxidative stress occurs in diabetes
as a consequence of several mechanisms related to
hyperglycaemia [5, 6]. These include accumulation
of AGE, activation of the polyol pathway and stimu-
lation of protein kinase C activity. Diabetes-induced
generation of reactive oxygen species, in particular
superoxide, decreases the expression of nitric oxide
synthase and inactivates nitric oxide [6, 31]. Vascular
oxidative stress could explain why our patients had
impaired FMD of the brachial artery with preserved
vasodilation to glyceryl trinitrate. Unexpectedly, our
patients did not show evidence of increased oxidative
stress, as measured by plasma concentrations of F
2
-
isoprostanes. Plasma F
2
-isoprostanes reflect non-en-
zymic, free radical induced lipid peroxidation and
might not be sensitive to increases in oxidative stress
at the vascular wall. Previously reported increases in
plasma F
2
-isoprostanes and their diminution in diabe-
tes with antioxidant vitamins referred to patients with
poorer glycaemic control than our study group [29,
30]. We found that plasma antioxidant capacity was
decreased in our patients in agreement with previous
reports showing that diabetic patients have decreased
plasma concentrations of antioxidant vitamins [32].
However, low plasma ORAC was not corrected by
CoQ supplementation, suggesting that CoQ might
not contribute to plasma antioxidants measured by
this assay. A dissociation between changes in endo-
thelial function and plasma F
2
-isoprostane concentra-
tions was found using antioxidant vitamin supple-
mentation in non-diabetic subjects [33]. Our patients
were dyslipidaemic and it is possible that increased
plasma concentrations of lipoprotein remnants,
small-dense LDL and low HDL cholesterol might
have contributed to endothelial dysfunction in the
absence of a systemic increase in oxidative stress [6].
There is evidence in Type II diabetes for and
against improvement in vasodilator function of fore-
arm resistance arteries in response to the muscarinic
agonist acetylcholine with vitamin E supplementa-
tion [10, 11]. One positive study using a vitamin E an-
alogue was not placebo controlled but did show re-
duction in plasma F
2
-isoprostanes in a small number
of patients [11]. Vitamin E supplementation was re-
ported to improve forearm microcirculatory function
in a larger sample of Type II diabetic patients in a
well-controlled trial [10]. Improvement in methacho-
line-mediated vasodilator function of forearm resist-
ance vessels was also reported in Type II diabetes fol-
lowing the intra-arterial administration of vitamin C
[12]. However, that study used a small sample size, in-
volved an acute intervention, and did not use a place-
bo arm. A recent report has shown that intra-arterial
administration of the powerful antioxidant a -lipoic
acid improved forearm blood flow responses to ace-
tylcholine to the same extent as ascorbic acid in pati-
ents with Type II diabetes [34]. The greatest benefit
was seen in patients with low-plasma concentration
of CoQ, supporting an important role of CoQ in vas-
cular endothelial dysfunction in Type II diabetes. We
G. F. Watts et al.: CoQ improves endothelial function in diabetes 423
Table 2. Changes in baseline diameter, reactive hyperaemia, flow-mediated dilatation (FMD) and nitrate-mediated dilatation
(NMD) of the brachial artery in the diabetic patients treated with placebo or CoQ supplementation for 12 weeks
Variable Placebo group CoQ group p value
Change in baseline artery diameter (mm) 0.05 (0.09) ±0.02 (0.08) 0.611
Change in resting blood flow (ml/min) ±43.3 (19.7) 20.6 (30.4) 0.136
Change in reactive hyperaemia (%) 85.0 (38.6) ±27.3 (84.5) 0.316
Change in FMD (%) ±0.4 (0.5) 1.6 (0.3) 0.005
Change in NMD (%) ±0.1 (1.4) 0.4 (1.2) 0.771
Values are means  SEM

have extended previous reports by investigating in-
tervention with the antioxidant CoQ in a larger sam-
ple of patients and in a peripheral conduit artery. In
contrast to forearm microcirculatory blood flow re-
sponses to acetylcholine, post-ischaemic FMD of the
brachial artery has been shown to be a surrogate for
the coronary circulation and to predict coronary
events in patients with angina [35, 36].
The beneficial properties of CoQ could relate not
only to its antioxidant effect [15, 16], but also to im-
provements in glycaemic control [18] and blood pres-
sure [14]. Coenzyme Q
10
is a powerful antioxidant
that might decrease superoxide generation from en-
dothelial cells [13, 14]. Since our patients remained
dyslipidaemic, their increased plasma concentration
of small-dense LDL and low HDL would have con-
tributed to endothelial dysfunction [6]. By decreasing
vascular oxidative stress, CoQ could decrease the ox-
idative modification of LDL and HDL in the seques-
tered environment of the arterial wall, thereby in-
creasing the synthesis and/or action of endogenously
derived nitric oxide [6, 37, 38]. This effect might par-
ticularly extend to triglyceride-rich lipoproteins in
the post-prandial phase, when oxidative stress could
be maximal in diabetes [39]. Coenzyme Q
10
supple-
mentation has been reported to improve glycaemic
control [14, 18, 21, 22] and blood pressure [22] in pati-
ents with diabetes. However, along with other studies
[40, 41], we found no evidence to support this finding.
Other potential mechanisms whereby CoQ could
have improved endothelial function in the brachial
artery involve reduction in the cellular levels of asym-
metric dimethyl-arginine and AGEs [31, 42], as well
as an increase in the bioavailability of tetrahydrobi-
opterin [43] and glutathione [44]. Normalization of
mitochondrial superoxide production could be cen-
tral to these mechanisms and to an anti-inflammatory
effect of CoQ [45].
We did not rigorously study the mechanism of ac-
tion of CoQ with pharmacological agents, such as
acetylcholine with and without N
G
-monomethyl-l-
arginine. Post-ischaemic FMD of human conduit ar-
teries has predominately been shown to be mediated
by nitric oxide, although studies to date have mostly
referred to the radial artery [46]. We cannot infer
that the vascular benefit of CoQ extends to microcir-
culatory function, where the mediators of shear-stress
induced increase in blood flow could be different
from conduit vessels [4]. Because NMD of the brachi-
al artery was tested maximally with a high dose of
GTN, we cannot strictly exclude that our diabetic pa-
tients had non-endothelial vascular dysfunction, and
that this abnormality improved with CoQ. Acute hy-
perglycaemia has been shown to impair FMD of the
brachial artery [47] and we did not study patients at
isoglycaemia. However, at the time of ultrasound, pa-
tients were on average near-normoglycaemic and
there was no difference in blood glucose concentra-
tions before and after intervention. Metformin has
also been reported to improve brachial artery endo-
thelial function in diabetes [48], but the number of pa-
tients included in our study on this agent were small
and did not differ between intervention and control
groups. The clinical relevance of the small but signifi-
cant 1.6 % increase in FMD in our patients with CoQ
supplementation is not clear, given that FMD is only
weakly correlated with coronary responses and that
NMD did not change with treatment. We were able
to detect a small improvement in FMD, however, be-
cause of the enhanced precision of our computerized
method for assessing luminal diameter changes.
If our findings reflect the favourable effect of CoQ
on the bioavailability and action of nitric oxide, they
have implications for the prevention and reversal of
atherogenesis, procoagulopathy and myocardial dys-
function in diabetic patients [2, 4, 6]. That a benefit
in endothelial function was seen in the presence of di-
abetic dyslipidaemia without complete restoration of
FMD to normal, raises the possibility of further in-
vestigating the synergistic effects of CoQ with other
agents that could improve vascular function in diabe-
tes, such as angiotensin-converting enzyme inhibi-
tors, fish-oils and lipid regulators [4, 6].
Acknowledgements. This study was supported by research
grants from Diabetes Australia, The National Health and
Medical Research Council of Australia and The Medical Re-
search Foundation, Royal Perth Hospital. We thank Black-
mores (Australia) for providing the CoQ and matching place-
bo and Professor J. Best for estimating the particle size of
LDL. We are also grateful to the research nurses and technical
staff of the University Department of Medicine, in particular
Ms M-A Powell, for her excellent assistance in carrying out
the study. We acknowledge the assistance of Ms L. Rich and
Mr R. Woodman in assuring the quality of the analyses of the
ultrasound scans.
References
1. Semenkovich CF, Heinecke JW (1997) The mystery of dia-
betes and atherosclerosis. Time for a new plot. Diabetes 46:
327±334
2. Rubanyi GM (1993) The role of endothelium in cardiovas-
cular homeostasis and diseases. J Cardiovasc Pharmacol
22 [Suppl 4]: S1-S14
3. Cohen RA (1993) Dysfunction of vascular endothelium in
diabetes mellitus. Circulation 87 [Suppl V]:V67±V76
4. De Vriese AS, Verbeuren TJ, Van der Voorde J, Lamiere
NH, Vanhoutte PM (2000) Endothelial dysfunction in dia-
betes. Br J Pharmacol 130: 963±974
5. Giugliano D, Ceriello A, Paolisso G (1996) Oxidative
stress and diabetic vascular complications. Diabetes Care
19: 257±267
6. Watts GF, Playford D (1998) Dyslipoproteinaemia and hy-
peroxidative stress in the pathogenesis of endothelial dys-
function in NIDDM: an hypothesis. Atherosclerosis 141:
17±31
7. Stocker R (1999) Dietary and pharmacological antioxi-
dants in atherosclerosis. Curr Opin Lipidol 10: 589±597
G. F. Watts et al.: CoQ improves endothelial function in diabetes424

8. Yusuf S, Dagenais G, Pogne J, Bosch J, Sleight P (2000) Vi-
tamin E supplementation and cardiovascular events in
high-risk patients. The Heart Outcomes Prevention Evalu-
ation Study Investigators. N Eng J Med 342: 154±160
9. Skyrme-Jones RAP, O'Brien RC, Berry KL, Meredith IT
(2000) Vitamin E supplementation improves endothelial
function in type 1 diabetes mellitus: a randomized, place-
bo-controlled study. J Am Coll Cardiol 36: 94±102
10. Gazis A, White DJ, Page SR, Calcroft JR (1999) Effect of
oral vitamin E (a-tocopheral) supplementation in vascular
endothelial function in type 2 diabetes mellitus. Diabet
Med 16: 304±311
11. Chowienczyk PJ, Brett SE, Gopaul NK et al. (2000) Oral
treatment with an antioxidant (raxofelast) reduces oxida-
tive stress and improves endothelial function in men with
Type II diabetes. Diabetologia 43: 974±977
12. Ting HH, Timimi K, Boles KS, Creager SJ, Ganz P, Creagar
MA (1996) Vitamin C improves endothelium-dependent
vasodilation in patients with non-insulin dependent diabe-
tes mellitus. J Clin Invest 97: 22±28
13. Beyer RE (1990) The participation of coenzyme Q in free
radical production and anti-oxidation. Free Radic Biol
Med 8: 545±465
14. McCarty MF (1999) Coenzyme Q versus hypertension:
does CoQ decrease endothelial superoxide generation?
Med Hypotheses 53: 300±304
15. Stocker R, Bowry VW, Frei B (1991) Ubiquinol-10 protects
human low-density lipoprotein more efficiently against lip-
id peroxidation than does a-tocopherol. Proc Natl Acad Sci
USA 88: 1646±1650
16. Thomas SR, Neuzil J, Stocker R (1996) Cosupplementa-
tion with coenzymeQ prevents the prooxidant effect of a-
tocopherol and increase the resistance of LDL to transition
metal-dependent oxidation initiation. Arterioscler Thromb
Vasc Biol 16: 687±696
17. DeFronzo RA, Bonadonna RC, Ferrannini E (1992)
Pathogenesis of NIDDM: a balanced overview. Diabetes
Care 15: 318±368
18. McCarty MF (1999) Can correction of sub-optimal coen-
zymeQ staus improve b-cell function in type II diabetes?
Med Hypotheses 52: 397±400
19. Jameson S (1991) Coenzyme Q
10
alpha-tocopherol, and
free cholesterol levels in sera from diabetic patients. In:
Folkers K, Littarru G, Yamagami T (eds). Biomedical and
Clinical Aspects of Coenzyme Q. Elsevier Science, Am-
sterdam pp 151±158
20. Werbach MR (1993) Nutritional Influences on Illness 2
nd
edn Third Line Press, Tarzona pp 250±251
21. Shigeta Y, Izumei K, Abe H (1966) Effect of coenzyme Q
7
treatment on blood sugar and ketone bodies of diabetics. J
Vitaminol (Kyoto) 12: 293±298
22. Singh RB, Niaz MA, Rastogi SS, Shukla PK, Thakur AS
(1999) Effect of hydrosoluble coenzyme Q
10
on blood pres-
sures and insulin resistance in hypertensive patients with
coronary artery disease. J Hum Hypertens 13: 203±208
23. Thomas SR, Witting PK, Stocker R (1999) A role for re-
duced coenzymeQ in atherosclerosis. Biofactors 9: 207±224
24. Woodman RA, Playford DA, Watts GF et al. (2001)
Improved analysis of brachial artery ultrasound using a
novel edge detection software system. J Appl Physiol 91:
929±937
25. Long JK, Gohil K, Packer L (1986) Simultaneous determi-
nation of tocopherols, ubiquinols and ubiquinones in
blood, plasma, tissue homogenates and subcellular frac-
tions. Anal Biochem 157: 106±116
26. Mori TA, Croft KD, Puddey IB, Beilin LJ (1999) An im-
proved method for the measurement of urinary and plasma
F2-isoprostanes using Gas Chromatography Mass Spec-
trometry. Anal Biochem 268: 117±125
27. Cao G, Alessio HM, Cutler RG (1993) Oxygen-radical ab-
sorbance capacity assay for antioxidants. Free Radic Biol
Med 14: 303±311
28. Dogra G, Rich L, Stanton K, Watts GF (2001) Endotheli-
um-dependent and independent vasodilation studied at
normoglycaemia in Type I diabetes mellitus with and with-
out microalbuminuria. Diabetologia 44: 593±601
29. Gopaul NK, Annggard EE, Mallet AI, Betteridge DJ,
Wolfe SP, Nourouz-Zadeh J (1995) Plamsa 8-epi-PGF
2a
levels are elevated in individuals with non-insulin depen-
dent diabetes mellitus. FEBS Lett 368: 225±229
30. Davi G, Ciabattoni G, Consoli A et al. (1999) In vivo for-
mation of 8-Iso-prostaglandin
F2a
and platelet activation in
diabetes mellitus: effects of improved metabolic control
and vitamin E supplementation. Circulation 99: 224±229
31. Pieper GM (1998) Review of alterations in endothelial ni-
tric oxide production in diabetes: protective effect of argin-
ine on endothelial dysfunction. Hypertension 31:
1047±1060
32. Sundaram RK, Bhaskar A, Vijayalingam S, Viswanathan
M, Mohan R, Shanmugasundarum KR (1996) Antioxidant
status and lipid peroxidation in type 2 diabetes mellitus
with and without complications. Clin Sci (Colch) 90:
255±260
33. Simons LA, von Konigsmark M, Simons J, Stocker R, Cel-
ermajer DS Vitamin E (1999) ingestion does not improve
arterial endothelial dysfunction in older adults. Atheroscle-
rosis 143: 1193±1199
34. Heitzer T, Finckh B, Albers S, Krohn K, Kohlschutter A,
Meinertz T (2001) Beneficial effects of a-lipoic acid and
ascorbic acid on endothelium-dependent, nitric oxide-me-
diated vasodilation in diabetic patients: relation to parame-
ters of oxidative stress. Free Radic Biol Med 31: 53±61
35. Anderson TJ, Uehata A, Gerhard MD et al. (1995) Close
relation of endothelial function in the human coronary
and peripheral circulations. J Am Coll Cardiol 26:
1235±1241
36. Neunteufl T, Heher S, Katzenschlager R et al. (2000)
Late prognostic value of flow-mediated dilation in the bra-
chial artery of patients with chest pain. Am J Cardiol 86:
207±210
37. Liao JK, Shin WS, Lee WY, Clark SL (1995) Oxidized low-
density lipoprotein decreases the expression of endothelial
nitric oxide synthase. J Biol Chem 270: 319±324
38. McNeill KL, Fontana L, Russell-Jones DL, Rajmau I, Rit-
ter JM, Chowienczyk PJ (2000) Inhibitory effects of low-
density lipoproteins from men with type 2 diabetes or en-
dothelium-dependent relaxation. J Am Cardiol 35:
1622±1627
39. Anderson RA, Evans ML, Ellis GR et al. (2001) The rela-
tionship between post-prandial lipaemia, endothelial func-
tion and oxidative stress in healthy individuals and patients
with type 2 diabetes. Atherosclerosis 154: 475±483
40. Eriksson JG, Forsen TJ, Mortensen SA, Rooohde M
(1999) The effect of coenzyme Q
10
administration on meta-
bolic control in patients with type 2 diabetes mellitus. Bio-
factors 9: 315±318
41. Henricksen JE, Andersen CB, Hother-Nielsen O, Vaag A,
Mortensen SA, Beck-Nielsen H. Impact of ubiquinone (co-
enzyme Q
10
) treatment on glycaemic control, insulin re-
quirements and well-being in patients with Type 1 diabetes
mellitus. Diabetic Medicine 199; 16: 312±318
42. Miyazaki H, Matsuoka H, Cooke JP et al. (1999) Endoge-
nous nitric oxide synthase inhibitor: A novel marker of ath-
erosclerosis. Circulation 99: 1141±1146
G. F. Watts et al.: CoQ improves endothelial function in diabetes 425

43. Huang A, Vita JA, Venema RC, Kearney JF (2000) Ascor-
bic acid enhances endothelial nitric oxide activity by in-
creasing intracellular tetrahydrobioterin. J Biol Chem 275:
17399±17406
44. Prasad A, Andrews NP, Padder FA, Hussain M, Quyyumi
AA (1999) Glutathione reverses endothelial dysfunction
and improves nitric oxide bioavailability. J Am Coll Car-
diol 34: 507±514
45. Nishikawa T, Edelstein D, Du XL et al. (2000) Normalizing
mitochondrial superoxide production blocks three path-
ways of hyperglycaemic damage. Nature 404: 787±790
46. Mullen MJ, Kharbanda RK, Cross J et al. (2001) Heteroge-
nous nature of flow-mediated dilatation in human conduit
arteries in vivo. Circ Res 88: 145±151
47. Kawano H, Motoyama T, Hirashima O et al. (1999) Hyper-
glycaemia rapidly suppresses flow-mediated endothelium-
dependent vasodilation of brachial artery. J Am Coll Car-
diol 34: 146±154
48. Mather KJ, Verma S, Anderson TJ (2001) Improved endo-
thelial function with metformin in type 2 diabetes mellitus.
J Am Coll Cardiol 37: 1344±1350
G. F. Watts et al.: CoQ improves endothelial function in diabetes426