Mechanisms of endothelial dysfunction in resistance arteries from patients with end-stage renal disease.
ABSTRACT The study focuses on the mechanisms of endothelial dysfunction in the uremic milieu. Subcutaneous resistance arteries from 35 end-stage renal disease (ESRD) patients and 28 matched controls were studied ex-vivo. Basal and receptor-dependent effects of endothelium-derived factors, expression of endothelial NO synthase (eNOS), prerequisites for myoendothelial gap junctions (MEGJ), and associations between endothelium-dependent responses and plasma levels of endothelial dysfunction markers were assessed. The contribution of endothelium-derived hyperpolarizing factor (EDHF) to endothelium-dependent relaxation was impaired in uremic arteries after stimulation with bradykinin, but not acetylcholine, reflecting the agonist-specific differences. Diminished vasodilator influences of the endothelium on basal tone and enhanced plasma levels of asymmetrical dimethyl L-arginine (ADMA) suggest impairment in NO-mediated regulation of uremic arteries. eNOS expression and contribution of MEGJs to EDHF type responses were unaltered. Plasma levels of ADMA were negatively associated with endothelium-dependent responses in uremic arteries. Preserved responses of smooth muscle to pinacidil and NO-donor indicate alterations within the endothelium and tolerance of vasodilator mechanisms to the uremic retention products at the level of smooth muscle. We conclude that both EDHF and NO pathways that control resistance artery tone are impaired in the uremic milieu. For the first time, we validate the alterations in EDHF type responses linked to kinin receptors in ESRD patients. The association between plasma ADMA concentrations and endothelial function in uremic resistance vasculature may have diagnostic and future therapeutic implications.
- Journal of the American Society of Nephrology 08/2003; 14(7):1927-39. · 8.99 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Accelerated cardiovascular disease is a frequent complication of renal disease. Chronic kidney disease promotes hypertension and dyslipidemia, which in turn can contribute to the progression of renal failure. Furthermore, diabetic nephropathy is the leading cause of renal failure in developed countries. Together, hypertension, dyslipidemia, and diabetes are major risk factors for the development of endothelial dysfunction and progression of atherosclerosis. Inflammatory mediators are often elevated and the renin-angiotensin system is frequently activated in chronic kidney disease, which likely contributes through enhanced production of reactive oxygen species to the accelerated atherosclerosis observed in chronic kidney disease. Promoters of calcification are increased and inhibitors of calcification are reduced, which favors metastatic vascular calcification, an important participant in vascular injury associated with end-stage renal disease. Accelerated atherosclerosis will then lead to increased prevalence of coronary artery disease, heart failure, stroke, and peripheral arterial disease. Consequently, subjects with chronic renal failure are exposed to increased morbidity and mortality as a result of cardiovascular events. Prevention and treatment of cardiovascular disease are major considerations in the management of individuals with chronic kidney disease.Circulation 08/2007; 116(1):85-97. · 15.20 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Endothelial dysfunction has been implicated as a key factor in the development of a wide range of cardiovascular diseases, but its definition and mechanisms vary greatly between different disease processes. This review combines evidence from cell-culture experiments, in vitro and in vivo animal models, and clinical studies to identify the variety of mechanisms involved in endothelial dysfunction in its broadest sense. Several prominent disease states, including hypertension, heart failure, and atherosclerosis, are used to illustrate the different manifestations of endothelial dysfunction and to establish its clinical implications in the context of the range of mechanisms involved in its development. The size of the literature relating to this subject precludes a comprehensive survey; this review aims to cover the key elements of endothelial dysfunction in cardiovascular disease and to highlight the importance of the process across many different conditions.Antioxidants & Redox Signaling 10/2008; 10(9):1631-74. · 8.20 Impact Factor
Mechanisms of Endothelial Dysfunction in Resistance
Arteries from Patients with End-Stage Renal Disease
Leanid Luksha1, Peter Stenvinkel2, Folke Hammarqvist3, Juan Jesu ´s Carrero2, Sandra T. Davidge4,
1Division of Obstetrics & Gynecology, Karolinska Institutet, Karolinska University Hospital, Department of Clinical Science, Intervention & Technology, Stockholm, Sweden,
2Division of Renal Medicine, Karolinska Institutet, Karolinska University Hospital, Department of Clinical Science, Intervention & Technology, Stockholm, Sweden,
3Division of Surgery, Karolinska Institutet, Karolinska University Hospital, Department of Clinical Science, Intervention & Technology, Stockholm, Sweden, 4Department of
Obstetrics and Gynecology, University of Alberta, Edmonton, Alberta, Canada
The study focuses on the mechanisms of endothelial dysfunction in the uremic milieu. Subcutaneous resistance arteries
from 35 end-stage renal disease (ESRD) patients and 28 matched controls were studied ex-vivo. Basal and receptor-
dependent effects of endothelium-derived factors, expression of endothelial NO synthase (eNOS), prerequisites for
myoendothelial gap junctions (MEGJ), and associations between endothelium-dependent responses and plasma levels of
endothelial dysfunction markers were assessed. The contribution of endothelium-derived hyperpolarizing factor (EDHF) to
endothelium-dependent relaxation was impaired in uremic arteries after stimulation with bradykinin, but not acetylcholine,
reflecting the agonist-specific differences. Diminished vasodilator influences of the endothelium on basal tone and
enhanced plasma levels of asymmetrical dimethyl L-arginine (ADMA) suggest impairment in NO-mediated regulation of
uremic arteries. eNOS expression and contribution of MEGJs to EDHF type responses were unaltered. Plasma levels of ADMA
were negatively associated with endothelium-dependent responses in uremic arteries. Preserved responses of smooth
muscle to pinacidil and NO-donor indicate alterations within the endothelium and tolerance of vasodilator mechanisms to
the uremic retention products at the level of smooth muscle. We conclude that both EDHF and NO pathways that control
resistance artery tone are impaired in the uremic milieu. For the first time, we validate the alterations in EDHF type
responses linked to kinin receptors in ESRD patients. The association between plasma ADMA concentrations and endothelial
function in uremic resistance vasculature may have diagnostic and future therapeutic implications.
Citation: Luksha L, Stenvinkel P, Hammarqvist F, Carrero JJ, Davidge ST, et al. (2012) Mechanisms of Endothelial Dysfunction in Resistance Arteries from Patients
with End-Stage Renal Disease. PLoS ONE 7(4): e36056. doi:10.1371/journal.pone.0036056
Editor: Marcelo G. Bonini, University of Illinois at Chicago, United States of America
Received September 27, 2011; Accepted March 29, 2012; Published April 26, 2012
Copyright: ? 2012 Luksha et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Financial support was provided by grants from the Swedish Society of Medicine, Swedish Heart and Lung Foundation, Center for Gender Medicine at
Karolinska Institutet, Loo and Hans Ostermans Foundation, the Swedish Kidney Association and the Karolinska Institute Research Funds. L. Luksha is supported by
PostDoc research grants from AFA insurance and Department of Clinical Science, Intervention & Technology (CLINTEC) at Karolinska Institutet, Sweden. S. Davidge
is supported by Canadian Institute for Health Research. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of
Competing Interests: The authors have read the journal’s policy and have the following conflicts: The study was partly funded by AFA Insurance. This does not
alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials, as detailed online in the guide for authors.
* E-mail: email@example.com
Adverse cardiovascular events are common complications of
end-stage renal disease (ESRD) and these patients are more likely
to die from cardiovascular disease (CVD) than from kidney failure
. Although the underlying mechanisms that predispose ESRD
patients to higher risk of CVD are incompletely understood,
morphological and functional abnormalities of the endothelium
may play an important role . Endothelial dysfunction is
considered an early marker of CVD , which facilitates the
progress of atherosclerosis  and contributes to the development
of hypertension through the enhancement of vascular resistance
. Thus, studies aimed to at investigating the mechanisms of
endothelial dysfunction in ESRD are of importance and may
provide a means to ameliorate cardiovascular complications and
introduce novel treatment strategies.
Our current knowledge of endothelial dysfunction in ESRD is
mainly based on findings from animal models , circulating
plasma markers  and in-vivo assessments in the human forearm
. Although few attempts have been made to estimate
endothelial function in resistance arteries of ESRD patients [9–
11], the mechanisms of endothelial dysfunction need further
clarification. NO deficiency has been considered as a principal
event leading to endothelial dysfunction in the uremic milieu .
However, the contribution of NO to endothelium-dependent
control of vascular tone is inversely associated with caliber of
arteries. Another vasodilator, known as endothelium-derived
hyperpolarizing factor (EDHF), seems to act as a predominant
mediator of endothelium-dependent dilatation in resistance-size
arteries. EDHF type responses are characterized by endothelium-
dependent hyperpolarization that occurs due to direct electrical
coupling via myoendothelial gap junctions (MEGJs) and/or the
release of different mediators . The role of EDHF in
endothelial maintenance has been introduced as a back-up
mechanism during NO deficiency. However when deprivation of
EDHF occurs, this may further aggravate endothelial dysfunction
PLoS ONE | www.plosone.org1April 2012 | Volume 7 | Issue 4 | e36056
leading to enhanced blood pressure and impaired blood flow to
target organs .
Studies concerning the detailed mechanisms of endothelial
dysfunction in resistance arteries with a focus on the relative
contribution of NO and EDHF in ESRD are scarce. Current data
on the contribution of EDHF to endothelium-dependent relaxa-
tion of resistance arteries in kidney failure are mainly based on
animal studies and characterized by explicit heterogeneity [6,14–
18]. To the best of our knowledge, the only study that has
investigated the relative contribution of EDHF vs. NO to
acetylcholine (ACh)-induced-relaxation in ESRD patients has
reported an impairment in NO-mediated responses but an
unchanged, or even increased, role of EDHF as assessed by
forearm blood flow .
In this study, we hypothesized that endothelial dysfunction in
resistance arteries of incident dialysis patients is not only restricted
to impairment in production and/or bioavailability of NO, but
EDHF type responses may also be affected by uremic milieu. To
test this hypothesis we isolated arteries from subcutaneous fat to
segregate pharmacologically the relative impairments in NO and
EDHF type responses that confer endothelial dysfunction in
ESRD. Targeted pathways of endothelial dysfunction were
assessed using basal and receptor-dependent stimulation of
endothelium-derived vasodilators, expression of endothelial NO
synthase (eNOS), prerequisites for MEGJ, and associations
between endothelium-dependent responses and plasma levels of
endothelial dysfunction surrogate markers.
Age, gender, and smoking status were similar between the
groups. The body mass index was lower in ESRD patients vs.
controls. Plasma levels of asymmetrical dimethyl L-arginine
(ADMA), soluble vascular cell adhesion molecule-1 (sVCAM-1),
interleukin-6, pentraxin-3, high sensitivity C-reactive protein
(hsCRP) and lipoprotein(a) and triglycerides were elevated in
ESRD. No differences in blood pressure or total cholesterol were
observed between the groups (Table 1).
In total, 84 subcutaneous arteries with internal diameter of
20966 mm were dissected from 35 ESRD patients and 71 arteries
with internal diameter of 22467 mm were dissected from 28
controls (P.0.05). There was no difference in the magnitude of
pre-constriction 3 mmol/L norepinephrine between the groups
(2.560.2 mN/mm2ESRD vs. 2.760.2 mN/mm2controls).
In controls, ACh and bradykinin (BK) caused relaxation of
arteries with similar magnitude (Figure 1). However, arteries were
more sensitive to BK vs. ACh (pEC50, here and in the following
text: BK 7.960.1 vs. ACh 7.760.1, P=0.01). NOS/cyclooxygen-
ase (COX) inhibition reduced endothelium-dependent relaxation
(Figure 1) and the sensitivity to agonists became similar (ACh:
7.260.1 vs. BK: 7.360.1, P=0.5).
Relaxation and sensitivity to both agonists was attenuated in
ESRD vs. controls (Figure 1). In contrast to the controls, the
sensitivities of ESRD arteries in PSS were similar between the
agonists (ACh: 7.360.1 vs. BK: 7.460.1, P=0.2).
After NOS/COX inhibition relaxation and sensitivity to ACh
and BK were attenuated in ESRD (Figure 1). The concentration-
response curves after NOS/COX inhibition were shifted to the
right in ESRD vs. controls (Figure 1). However, the maximal
EDHF type relaxation was reduced in ESRD vs. controls in
response to BK but not to ACh (BK: P=0.003; ACh: P=0.09).
Moreover, the relative contribution of EDHF was reduced in
ESRD vs. controls in response to BK but not to ACh (Figure 2).
An inhibitor of gap junctions (18-a-glycyrrhetinic acid, 18-aGA)
markedly reduced EDHF type relaxation in response to both
agonists (Figure 1). There was no difference in residual relaxation
after incubation with 18-aGA along with NOS inhibitor, Nv-nitro-
L-arginine-methyl ester (L-NAME) and COX inhibitor, indo-
methacin (Indo) between ESRD vs. controls (Figure 1). The
relative contribution of MEGJs to EDHF type responses was
similar between ESRD and controls independently of the agonist
used (ACh, 1 mmol/l: 8664 ESRD (n=16) vs. 8066 controls
(n=12), P=0.7; BK, 1 mmol/l: 8565 ESRD (n=13) vs. 8763
controls (n=8), P=0.9).
In order to eliminate the possible interference of co-morbidities,
the responses to agonists before and after NOS/COX inhibition in
ESRD patients without diabetes mellitus (DM) and CVD were
compared with those of controls. In response to BK we observed
similar results as above (Figure 1B). In contrast to the whole ESRD
group, in arteries from ESRD without DM and CVD, ACh-
induced relaxation was reduced in PSS as compared to controls
but similar after NOS/COX inhibition (ACh after NOS/COX
inhibition: 760.1 ESRD without DM and CVD (n=18) vs.
7.260.1 controls (n=23), P=0.1).
Table 1. Baseline characteristics of ESRD patients and
Parameters ESRD (n=35)
Males n (%)24 (69)21 (75)
Body mass index (kg/m2)24.163.3*27.663.7
Systolic blood pressure (mmHg)144621 138617
Diastolic blood pressure (mmHg) 8661184611
Total cholesterol (mmol/L)4.661.2* 5.161.0
Triglycerides (mmol/L) 1.6* (0.8–5.7)1.3 (0.7–19)
Lipoprotein(a) (mg/L) 409 (50–2572)*156 (50–802)
S-albumin (g/L) 34.963.4*38.463.4
S-Creatinine (mmol/L) 620 (249–1069)*78 (55–100)
Interleukin-6 (pg/ml) 5.0 (1.9–16.8)*1.5 (0.4–17.1)
hsC-reactive protein (mg/L)1.9 (0.2–24.9)* 1.4 (0.4–13.9)
Pentraxin-3 (ng/ml)1.2 (0.5–8.3)*0.6 (0.1–2.3)
Fibrinogen (g/L) 4.861.3* 3.261.3
Glomerular filtration rate (ml/min)1263* 8963
Asymmetric Dimethylarginine (mmol/L) 0.660.1* 0.560.1
Soluble VCAM-1 (ng/ml)1295 (637–1980)*588 (368–830)
Soluble ICAM-1 (ng/ml)223 (138–404)231 (161–363)
Diabetes mellitus n, (%) 10 (29)0
CVD, n, (%)13 (37)0
Antihypertensive treatment, n, (%)33 (94)0
Statin treatment, n, (%) 15 (43)0
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org2April 2012 | Volume 7 | Issue 4 | e36056
Transmission electron microscopy (TEM)
Analysis of TEM images focused on morphological prerequisites
for the gap junctions between EC and smooth muscle cells (SMC)
in arteries from ESRD patients and controls (n=3). The main
criteria for identification of MEGJs was the presence of the
characteristic pentalaminar membrane structure at points of cell to
cell contact, where the central region had a higher electron opacity
than the inner parts and distance between the EC and SMC
plasma membranes was around 3.5 nm . TEM images showed
the presence of long protrusions (up to 4.5 mm) from both ECs
(Figure 3A) and SMCs (Figure 3B) penetrating the internal elastic
lamina and forming close contacts with each other (Figure 3C and
3D). Although the observed EC-SMC contacts did not fulfill all
criteria for the characteristic pentalaminar structures, they could
be considered as prerequisites for MEGJs.
Endothelium-independent relaxation to NO donor sodium
nitroprusside (SNP) was similar between the groups (pEC50:
6.060.1 ESRD (n=17) vs. 6.260.2 controls (n=13), P=0.6). In
ESRD patients pinacidil-induced responses were blunted as
compared to controls (6.160.1 ESRD vs 5.760.1 controls,
P=0.01, Figure 4). NOS/COX inhibition attenuated the response
in controls but not in ESRD (6.160.1 PSS vs 5.860.1 L-
Figure 1. Concentration response curves to acetylcholine (ACh, A) and bradykinin (BK, B). Responses in physiological salt solution (PSS)
and after incubation with Nv-nitro-L-arginine methyl ester plus indomethacin alone (L-NAME+Indo) or together with 18a-glycyrrhetinic acid (L-
NAME+Indo+18a-GA) in arteries from ESRD patients (n=32 for ACh and n=22 for BK) and controls (n=23 for ACh and n=17 for BK). * ESRD vs.
Figure 2. The relative contribution of endothelium-derived
hyperpolarizing factor (EDHF). Contribution of EDHF in arteries
from ESRD patients and controls in response to acetylcholine (ACh) and
bradykinin (BK). * ESRD vs. controls, P,0.05.
Figure 3. Transmission electron images of arteries from ESRD
patients. The lower magnification pictures (A,B) show an overview of
the vascular wall with endothelium (End) and smooth muscle (SM)
being separated by the internal elastic lamina (IEL). The areas denoted
by the boxes are magnified and show the sites of intercellular contacts
that could be considered as prerequisites of myoendothelial gap
junctions (C, D). The width of the gap is ,20 nm (C, arrow), ,11 nm (D,
arrow). Bar: (A) 2 mm; (B) 3 mm; (C) 0.1 mm; (D) 0,2 mm.
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org3April 2012 | Volume 7 | Issue 4 | e36056
Name+Indo, P=0.03 controls; 5.760.1 vs 5.660.1, P=0.3
ESRD, respectively, Figure 4). There was no difference between
ESRD and controls in their responses to pinacidil after NOS/
COX inhibition (Figure 4).
Influence of the endothelium-derived factors on basal
tone and expression of eNOS
NOS/COX inhibitors induced constriction of arteries from
both ESRD and control groups. This constriction was reduced in
ESRD vs. controls (Figure 5). Exclusion of the patients with DM
and CVD from ESRD group did not change the outcome
(Figure 5; 0.2860.1 mN/mm2ESRD without DM and CVD
(n=17) vs. 0.5960.1 mN/mm2controls (n=26), P=0.02). There
was no difference in eNOS expression in ESRD vs. controls
Associations between endothelium-dependent
responses and plasma markers of endothelial
The sensitivity of arteries to ACh and BK was negatively
associated with plasma levels of ADMA in ESRD (Figure 7A) but
not in controls (Figure 7B). There was no association between
ADMA and vascular sensitivity to the endothelium-dependent
agonists after NOS/COX inhibition in both groups. Similarly,
sensitivity to SNP was not associated with ADMA levels (data not
In contrast to the agonists-induced relaxation, constriction in
response to NOS/COX inhibition was negatively associated with
plasma levels of ADMA in both ESRD and controls (Figure 7C
No relation was found between resistance artery function and
other surrogate markers of endothelial dysfunction (soluble
intercellular adhesion molecule-1 (sICAM-1) and sVCAM-1) in
ESRD and controls (data not shown).
In this ex-vivo study of human uremic resistance arteries we
describe, for the first time, the relative role of endothelium-derived
factors, agonist-specific differences and associations between
endothelial function and surrogate plasma markers of endothelial
dysfunction. We show that reduced EDHF type responses
contribute markedly to endothelial dysfunction in ESRD.
Impaired EDHF type responses in ESRD were detected with
endothelium-dependent agonist BK but not ACh. Thus, we
suggest that changes in signal transduction from endothelial
receptors towards generation and/or transformation of hyperpo-
larization to the smooth muscle are differently affected by uremic
toxins with a predominant impact on those mediated by kinin
receptors. Diminished vasodilator influence of the endothelium on
basal tone of SMCs along with enhanced plasma levels of ADMA
indicates an impairment in NO-mediated control of arterial tone
in ESRD. While the eNOS expression and the contribution of
Figure 4. Concentration response curves to pinacidil. Responses
in arteries from ESRD patients (n=10) and controls (n=7) in
physiological salt solution (PSS) and after incubation with Nv-nitro-L-
arginine methyl ester plus indomethacin (L-NAME+Indo). * ESRD vs.
#before vs. after incubation with L-NAME+Indo,
Figure 5. Contractile response to NOS/COX inhibitors of
arteries from controls (n=26) vs. ESRD patients with (n=32).
*, P,0.05 ESRD vs. controls.
Figure 6. Endothelial nitric oxide synthase (eNOS) expression
in arteries from controls (n=6) and ESRD patients (n=10).
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org4April 2012 | Volume 7 | Issue 4 | e36056
MEGJs to EDHF type responses appeared to be unaltered in
uremic arteries, the upstream machinery of both endothelial
pathways (i.e. NO and EDHF) were impaired. Since relaxation in
response to NO donor or hyperpolarizing agent pinacidil (after
NOS/COX inhibition) were similar between the groups, we
confirm that the endothelium is the main target of uremic
environment, whereas functional capacity of the vascular smooth
muscle appeared to be rather tolerant. In accordance with our
previous study  we corroborate a central role of the uremic
milieu in the genesis of endothelial dysfunction. The present study
also shows that among measured plasma markers of endothelial
dysfunction only ADMA was strongly associated with the
magnitude of endothelial dysfunction in uremic resistance arteries.
Thus, our findings provide novel insights into the mechanisms of
endothelial dysfunction in resistance circulation of ESRD patients.
The pattern of impairment of EDHF type responses after BK
but not ACh stimulation in uremic arteries emphasizes the agonist-
specific mechanisms of endothelial dysfunction in this toxic milieu.
We speculate that conventional and/or disease-specific risk factors
may differently affect kinin and muscarinic receptors and/or their
regulatory pathways. For example, endothelial dysfunction in
atherosclerosis appears to be receptor-specific, involving the
muscarinic receptors with relative sparing of the kinin receptor
pathways. Abnormal reactivity of epicardial coronary arteries
during physiologic stress is better represented by BK and not by
ACh responses . Moreover, differences exist between BK- and
ACh- induced relaxation of the mesenteric arteries from
spontaneously hypertensive rats at different ages, suggesting a
more detrimental effect of increased blood pressure on BK-
induced vasorelaxation , while selective impairement of
endothelium-dependent relaxation to ACh but not BK is observed
in isolated small omental arteries from women with preeclampsia
. Thus, prior conclusions based only on the vascular effects of
one agonist (i.e. ACh) should be considered with caution. In
contrast to previous studies, in which only ACh was tested [9–10],
our more comprehensive analysis of pathways involved in
endothelial dysfunction of resistance arteries in ESRD patients
revealed an impairment of EDHF contribution coupled with
stimulation of kinin receptors.
Figure 7. Spearman rank correlation between plasma levels of asymmetrical dimethyl L-arginine (ADMA, mmol/L) and artery
sensitivity to endothelium-dependent vasodilators (pEC50, A, B) or vasoconstriction in response to NOS/COX inhibition (L-
NAME+ +Indo, C, D) in ESRD patients (A, C) and controls (B, D).
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org5April 2012 | Volume 7 | Issue 4 | e36056
In contrast to our findings on heterogeneity of mechanisms of
EDHF type responses in preeclampsia [24–25], the present data
revealed MEGJs as a common pathway of EDHF in both groups.
Hence, the compensatory response for preservation of endothelial
function via the flexibility of mechanisms behind EDHF type
responses seems to be lacking in ESRD. Since the contribution of
EDHF and MEGJ to ACh-induced relaxation was similar between
the two experimental groups, it is unlikely that impairment at the
level of MEGJs could be linked with kinin receptors. Therefore, it
can be speculated that alterations in BK-induced EDHF type
responses in uremic resistance arteries may reside in the signaling
sequence extending from B2-kinin receptors to the activation of
Ca2+-dependent K+-channels (Kca-channels) generating hyperpo-
larization of the endothelium with following transformation to the
smooth muscle via MEGJ. Further studies are warranted to clarify
the involvement of particular endothelial Kca-channels in
alterations of BK-induced EDHF type responses in this patient
Despite the fact that EDHF normally does not act through the
KATP-channels , pinacidil-induced responses allowed us to
assess the general mechanism of relaxation induced by hyperpo-
larization due to outward K+currents at the level of the smooth
muscle . Moreover, an animal study of renal failure suggested
that alterations in smooth muscle K+-channels could be involved
in reduced endothelium-dependent hyperpolarization . In our
study, relaxation to pinacidil was reduced in ESRD vs. controls but
this difference disappeared after NOS/COX inhibition, which
opposes the findings in the animal study . Most likely basal
NO had a potentiating effect on pinacidil-induced relaxation in
controls but not in ESRD patients. While in general relaxation
induced by KATP-channels openers has yet been considered
endothelium-independent , the potentiating effects of endo-
thelium-derived factors has been reported before [25,28].
Impaired endothelial influence on pinacidil-induced responses
may further support our data about reduced basal release of
endothelium-derived factors in ESRD. Indeed, NOS/COX
inhibitors induced smaller constriction in uremic vs. control
arteries. As basal vascular tone is to a large extend NO-dependent
, our data implies a reduction in basal production of NO in
ESRD. In contrast, a previous study reported increased basal NO
production in the forearm of hemodialysis patients . The
inconsistent results may be caused by different methodology, and
selection of patients. Recently, we demonstrated the lack of NO
contribution to shear stress responses in subcutaneous uremic
arteries . In the current study, differences in sensitivity
between BK and ACh, depending from NOS/COX inhibition
in controls but not in ESRD, indicated on distinct NO
contribution to agonist-induced relaxation between the two
groups. Moreover, the negative correlation between serum ADMA
levels and relaxation to ACh and BK in ESRD but not in controls
further supports the impaired contribution of NO to agonists-
induced responses in uremia.
Multiple mechanisms may lead to NO deficiency in renal failure
[12,31]. A decreased bioavailability of NO due to increased pro-
oxidative environment has been suggested . Reduced
expression of eNOS has been linked to a decreased NO
production in an animal model of kidney failure . Studies on
EC cultures have shown that erythrocytes  or sera fractions
enriched with advanced glycation end products  from uremic
patients may directly affect expression and activity of eNOS.
However, we failed to find any difference in eNOS protein
expression between uremic and control arteries. Moreover, an
elevated vascular expression but unchanged activity of eNOS was
demonstrated in radial arteries of ESRD patients . Neverthe-
less, such observations might indicate that unchanged or even
increased expression of eNOS in the vascular wall could not
guarantee a sufficient NO bioavailability in ESRD patients. We
therefore speculate that endothelial dysfunction along with
unchanged expression of eNOS, may represent a potential
outcome of a reduced ability of the enzyme to generate NO via
eNOS uncoupling with following decrease in NO bioavailability in
uremic resistance arteries.
ADMA may serve as a feasible candidate to link the uremic
environment with endothelial dysfunction in the resistance
vasculature. Indeed, ADMA, as an endogenous inhibitor of eNOS
that accumulates when renal function declines , it has been
shown to induce eNOS uncoupling either via substrate reduction
or via direct effect on eNOS catalysis and, as a consequence,
eNOS will generate superoxide radicals instead of NO . On
the other hand, we cannot exclude the possibility that the effect of
ADMA on endothelial superoxide generation may be due to the
activation of other enzymatic sources of superoxide radicals such
as NADPH oxidases. ADMA may also stimulate renin angiotensin
system, and ADMA induced impairement of NO-mediated
function due to increased superoxide production has been shown
to occur via activation of of Ang II-NADPH oxidase pathway in
isolated small vessels from rats . In animal models of
experimental diabetic nephropathy, AngII induced activation of
NADPH oxidase and eNOS uncoupling serves as the major source
of superoxide, and the blockade of AngII signaling ameliorates
eNOS uncoupling by increased tetrahydrobiopterin levels with
following restoration of NO bioavailability and improved
glomerular hemodynamics [39–40].
Although, an association between elevated ADMA levels and
endothelium-dependent dilatation in the forearm was previously
reported , we are the first to show an association between
ADMA and endothelium-dependent relaxation in uremic resis-
tance vasculature. The inverse correlation between ADMA and
changes in basal tone after NOS/COX inhibition in both
experimental groups support in-vivo results demonstrating that
ADMA increases vascular resistance in ESRD patients  and in
healthy humans . Taken together our data endorse the
proposal that elevated ADMA may act as a potential mechanism
behind the impaired NO-dependent control of uremic resistance
On the other hand, adhesion molecules sICAM-1 and sVCAM-
1, two purported biomarkers of endothelial dysfunction, did not
correlate with changes in basal tone nor with responses to
endothelium-dependent agonists. As ADMA is a potentially
modifiable risk factor, future interventional studies primarily
focusing on acute and long term L-arginine supplementation
 or regulation of dimethylarginine dimethylaminohydrolase
activity that confers the intracellular ADMA concentrations 
are of interest.
In summary, by studying the mechanisms of endothelial
dysfunction in uremic resistance arteries we were able to dissect
the impairment in basal and agonist-specific effects of two
endothelium-derived vasodilators - EDHF and NO. For the first
time, we provided evidence of impaired EDHF type responses
particularly linked to kinin receptors in ESRD. The current results
support our previous findings that NO plays a critical role in
uremic endothelial dysfunction in resistance circulation. The
observation that preserved responses of smooth muscle to pinacidil
and NO-donor indicated a decisive role of malfunctions within the
endothelium and tolerance of vasodilator mechanisms to the
uremic retention products at the level of smooth muscle. As this
study showed an association between circulating plasma ADMA
concentrations and endothelial dysfunction in uremic resistance
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org6April 2012 | Volume 7 | Issue 4 | e36056
vasculature, our findings may have diagnostic and future
Materials and Methods
The study was approved by the Ethical Committee at
Karolinska University Hospital and conducted according to the
principles expressed in the Declaration of Helsinki. All participants
involved in the research gave written informed consent prior to
Subcutaneous fat biopsies were obtained from 35 ESRD
patients at the time of peritoneal dialysis catheter insertion. Only
patients starting dialysis treatment were included. Exclusion
criteria were acute infection, vasculitis or liver disease at the time
of evaluation. Control tissue was obtained from 28 age-matched
volunteers without renal, mental or diabetic disease who
underwent hernia repair (n=17) or laparoscopic cholecystectomy
Baseline Laboratory and Clinical Assessments
Clinical history of CVD or DM was obtained from medical
records. CVD was defined as the presence of ischemic cardiac
disease, peripheral vascular disease and/or cerebrovascular
disease. Ongoing medication was collected from medical charts.
Glomerular filtration rate was estimated by the mean of
creatinine- and urea clearances in ESRD patients, whereas
cystatin-C estimated glomerular filtration rate in the controls.
Fasting venous blood samples were taken. Plasma and serum were
stored at 270uC pending further analyses. Serum interleukin-6
was measured on an ImmuliteH analyzer (Siemens Medical
Solution Diagnostic, Los Angeles, CA, USA). Serum concentra-
tions of albumin, creatinine, lipids and hsCRP were measured
routinely. ADMA was assessed in serum by ELISA assays (DLD
Diagnostika GMBH, Germany). Concentrations of pentraxin-3,
sICAM-1 and sVCAM-1 were measured in serum (ELISA assays
from R&D systems, USA).
Arteries were isolated and mounted on two stainless steel wires
(25 mm in diameter) in the organ baths of a four-channel wire
myograph (model 610, Danish Myo Technology; Aarhus, Den-
mark) as described previously . Arteries collected from patients
with ESRD we refer as ‘‘uremic arteries’’.
Once a sustained, steady contraction to norepinephrine
(3 mmol/L) was attained, the concentration-response curves to
the endothelium-dependent vasodilators ACh and BK (1 nmol/L
to 3 mmol/L) or SNP (10 nmol/L to 100 mmol/L), NO-donor and
an opener of ATP-sensitive K+-channels (KATP-channels), pinaci-
dil (10 nmol/L to 100 mmol/L), were obtained. Arteries were then
incubated for 20 min with NOS inhibitor (300 mmol/L) and COX
inhibitor, Indo (10 mmol/L). Subsequently, arteries were pre-
constricted again and second concentration-response curve for
‘‘EDHF’’ used in this study refers to the L-NAME+Indo-
insensitive component of endothelium-dependent vasodilatation.
The level of increased resting tone of the arteries after incubation
with L-NAME+Indo was considered as an index of vasoactive
properties of the endothelium, reflecting a basal release of
endothelium-derived vasoactive factors. To evaluate the contribu-
tion of gap junctions in EDHF type responses, the concentration-
response curves to ACh and BK were constructed after 15 min co-
incubation with 18-aGA (100 mmol/L) in the presence of L-
Freshly isolated arteries where cryopreserved in optimal cutting
temperature compound on dry ice. Transverse 8 mm cryosections
were prepared and mounted onto slides, air-dried, and stored at
280uC. For immunostaining, cryosections were incubated for
1.5 hr at room temperature with the mouse polyclonal anti-eNOS
antibody (1:250, BD Biosciences 610296). Incubation with the
secondary goat anti-mouse antibody (Invitrogen, Alexa fluor 488
A11001) was done for 1 hr in the dark. Glass coverslips were
mounted with Vectashield H-1200 Mounting Kit (Vector
Laboratories). Stained sections were examined immediately under
fluorescence microscope. All images presented are in (6100)
Transmission electron microscopy
Artery segments were fixed as described previously [24–25].
Serial transverse, ultra-thin sections (approximately 50–80 nm)
were cut. The series consisted of 3–5 sections. For each artery such
series were repeated three times after an interval of 10 mm.
Sections were examined in a Tecnai 10 transmission electron
microscope at 80 kV and digital images were captured.
The composition of PSS was (in mmol/L): NaCl 119, KCl 4.7,
CaCl22.5, MgSO41.17, NaHCO325, KH2PO41.18, EDTA
0.026, and glucose 5.5. The chemicals were obtained from Sigma,
St. Louis. To prepare stock solution, the substances were dissolved
in distilled water. Indo and pinacidil were dissolved in ethanol and
18a-GA was dissolved in DMSO. Pilot studies showed that the
solvents used had no effect upon vascular responses at their final
In figures, results are expressed as mean 6 SEM. In tables,
normally distributed variables are expressed as mean 6 SD, and
non-normally distributed variables as medians and interquartile
ranges. Baseline characteristics of the patients and arteries used
and staining were analysed by conventional parametric and non-
parametric methods. The isometric force developed by artery
segment during application of vasoactive compounds was
calculated using Myodata (Danish Myo Technology, Denmark)
and expressed as mN/mm2. Relaxation was expressed as a
percentage of the pre-constriction. In order to visualize the relative
contribution of EDHF or MEGJs, a percentage of the relaxation
was calculated after pre-incubation with L-NAME+Indo or L-
NAME+Indo+18a-GA and related to the full response in PSS or
after pre-incubation with L-NAME+Indo. Negative log concen-
tration (in mol/l) required to achieve 50% of the maximum
response (pEC50) was calculated by nonlinear regression analysis
(BioDataFit 1.02). ANOVA was used to compare concentration-
response curves before and after incubation with different
inhibitors. Spearman’s rank correlation was used to determine
the associations between artery sensitivity (pEC50) to endothelium-
dependent vasodilators and plasma markers of endothelial
dysfunction. Significance was taken at the 5% level for all
comparisons. All statistical analyses were performed with STA-
TISTICA (v.10.0, StatSoft, Uppsala, Sweden).
We would like to thank the patients and personnel at Karolinska University
Hospital-Huddinge involved in the sample collection. Special consideration
to our biochemical analysis coordinator Bjo ¨rn Anderstam, and Monica
Eriksson and Ann-Christin Bragfors-Helin for biochemical analysis; to
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org7 April 2012 | Volume 7 | Issue 4 | e36056
John Sandberg and Olof Heimbu ¨rger for patients recruitment; to KBC
(Annika Nilsson, Anki Emmot and Ulrika Jensen) for sampling protocol
Conceived and designed the experiments: LL PS KK. Performed the
experiments: LL PS STD KK. Analyzed the data: LL STD KK.
Contributed reagents/materials/analysis tools: LL PS JJC FH STD KK.
Wrote the paper: LL PS KK. Collected biopsy material: FH.
1. Stenvinkel P, Pecoits-Filho R, Lindholm B (2003) Coronary Artery Disease in
End-Stage Renal Disease: No Longer a Simple Plumbing Problem. J Am Soc
Nephrol 14: 1927–1939.
Schiffrin EL, Lipman ML, Mann JF (2007) Chronic Kidney Disease: Effects on
the Cardiovascular System. Circulation 116: 85–97.
Brocq ML, Leslie SJ, Milliken P, Megson I (2008) Endothelial Dysfunction:
From Molecular Mechanisms to Measurement, Clinical Implications, and
Therapeutic Opportunities. Antioxid Redox Signal 10: 1631–1674.
Chhabra N (2009) Endothelial dysfunction – A predictor of atherosclerosis.
Internet J Med Update 4: 33–41.
Fe ´le ´tou M, Ko ¨hler R, Vanhoutte P (2010) Endothelium-derived Vasoactive
Factors and Hypertension: Possible Roles in Pathogenesis and as Treatment
Targets. Curr Hypertens Rep 12: 267–275.
Vettoretti S, Ochodnicky P, Buikema H, Henning RH, Kluppel CA, et al. (2006)
Altered myogenic constriction and endothelium-derived hyperpolarizing factor-
mediated relaxation in small mesenteric arteries of hypertensive subtotally
nephrectomized rats. J Hypertens 24: 2215–2223.
Malyszko JS, Malyszko J, Hryszko T, Kozminski P, Pawlak K, et al. (2006)
Markers of endothelial damage in patients on hemodialysis and hemodiafiltra-
tion. J Nephrol 19: 150–154.
Yilmaz MI, Saglam M, Carrero JJ, Qureshi AR, Caglar K, et al. (2008) Serum
visfatin concentration and endothelial dysfunction in chronic kidney disease.
Nephrol Dial Transplant 23: 959–965.
Morris S, McMurray J, Spiers A, Jardine AG (2001) Impaired endothelial
function in isolated human uremic resistance arteries. Kidney Int 60:
10. Passauer J, Pistrosch F, La ¨ssig G, Herbrig K, Bu ¨ssemaker E, et al. (2005) Nitric
oxide- and EDHF-mediated arteriolar tone in uremia is unaffected by selective
inhibition of vascular cytochrome P450 2C9. Kidney Int 67: 1907–1912.
11. Luksha N, Luksha L, Carrero JJ, Hammarqvist F, Stenvinkel P, et al. (2011)
Impaired resistance artery function in patients with end-stage renal disease. Clin
Sci (Lond) 120: 525–536.
12. Baylis C (2008) Nitric oxide deficiency in chronic kidney disease. Am J Physiol
Renal Physiol 294: F1–9.
13. Luksha L, Agewall S, Kublickiene K (2009) Endothelium-derived hyperpolar-
izing factor in vascular physiology and cardiovascular disease. Atherosclerosis
14. Benchetrit S, Green J, Katz D, Bernheim J, Rathaus M (2003) Early endothelial
dysfunction following renal mass reduction in rats. Eur J Clin Invest 33: 26–33.
15. Kohler R, Eichler I, Schonfelder H, Grgic I, Heinau P, et al. (2005) Impaired
EDHF-mediated vasodilation and function of endothelial Ca2+-activated K+-
channels in uremic rats. Kidney Int 67: 2280–2287.
16. Kalliovalkama J, Jolma P, Tolvanen J-P, Ka ¨ho ¨nen M, Hutri-Ka ¨ho ¨nen N, et al.
(1999) Potassium channel-mediated vasorelaxation is impaired in experimental
renal failure. Am J Physiol Heart Circ Physiol 277: H1622–H1629.
17. Kimura K, Nishio I (1999) Impaired endothelium-dependent relaxation in
mesenteric arteries of reduced renal mass hypertensive rats. Scand J Clin Lab
Invest 59: 199–204.
18. Gschwend S, Haug MB, Nierhaus M, Schulz A, Vetter R, et al. (2009) Short-
term treatment with a beta-blocker with vasodilative capacities improves
intrarenal endothelial function in experimental renal failure. Life Sci 85:
19. Passauer J, Pistrosch F, Bu ¨ssemaker E, La ¨ssig G, Herbrig K, et al. (2005)
Reduced Agonist-Induced Endothelium-Dependent Vasodilation in Uremia Is
Attributable to an Impairment of Vascular Nitric Oxide. J Am Soc Nephrol 16:
20. Sokoya EM, Burns AR, Marrelli SP, Chen J (2007) Myoendothelial gap junction
frequency does not account for sex differences in EDHF responses in rat MCA.
Microvascular Research 74: 39–44.
21. Prasad A, Husain S, Schenke W, Mincemoyer R, Epstein N, et al. (2000)
Contribution of bradykinin receptor dysfunction to abnormal coronary
vasomotion in humans. J Am Coll Cardiol 36: 1467–1473.
22. Wirth KJ, Linz W, Wiemer G, Scho ¨lkens BA (1996) Differences in acetylcholine-
and bradykinin-induced vasorelaxation of the mesenteric vascular bed in
spontaneously hypertensive rats of different ages. Naunyn Schmiedebergs Arch
Pharmacol 354: 38–43.
23. Pascoal IF, Lindheimer MD, Nalbantian-Brandt C, Umans JG (1998)
Preeclampsia selectively impairs endothelium-dependent relaxation and leads
to oscillatory activity in small omental arteries. J Clin Invest 101: 464–470.
24. Luksha L, Luksha N, Kublickas M, Nisell H, Kublickiene K (2010) Diverse
Mechanisms of Endothelium-Derived Hyperpolarizing Factor-Mediated Dila-
tation in Small Myometrial Arteries in Normal Human Pregnancy and
Preeclampsia. Biol Reprod 83: 728–735.
25. Luksha L, Nisell H, Luksha N, Kublickas M, Hultenby K, et al. (2008)
Endothelium-derived hyperpolarizing factor in preeclampsia: heterogeneous
contribution, mechanisms, and morphological prerequisites. Am J Physiol Regul
Integr Comp Physiol 294: R510–R519.
26. Ku ¨hberger E, Kukovetz WR, Groschner K (1993) Cromakalim Inhibits
Multiple Mechanisms of Smooth Muscle Activation with Similar Stereoselec-
tivity. J Cardiovasc Pharmacol 21: 947–954.
27. Stojnic N, Gojkovic-Bukarica L, Peric M, Grbovic L, Lesic A, et al. (2007)
Potassium channel opener pinacidil induces relaxation of the isolated human
radial artery. J Pharmacol Sci 104: 122–129.
28. Deka DK, Raviprakash V, Mishra SK (1998) Basal nitric oxide release
differentially modulates vasodilations by pinacidil and levcromakalim in goat
coronary artery. Eur J Pharmacol 348: 11–23.
29. Rees DD, Palmer RM, Moncada S (1989) Role of endothelium-derived nitric
oxide in the regulation of blood pressure. Proc Natl Acad Sci USA 86:
30. Passauer J, Bussemaker E, Range U, Plug M, Gross P (2000) Evidence In Vivo
Showing Increase of Baseline Nitric Oxide Generation and Impairment of
Endothelium-Dependent Vasodilation in Normotensive Patients on Chronic
Hemodialysis. J Am Soc Nephrol 11: 1726–1734.
31. Kao M, Ang D, Pall A, Struthers AD (2009) Oxidative stress in renal
dysfunction: mechanisms, clinical sequelae and therapeutic options. J Hum
Hypertens 24: 1–8.
32. Kim SW, Lee JU, Paek YW, Kang DG, Choi KC (2000) Decreased nitric oxide
synthesis in rats with chronic renal failure. J Korean Med Sci 15: 425–430.
33. Bonomini M, Pandolfi A, Pietro ND, Sirolli V, Giardinelli A, et al. (2005)
Adherence of uremic erythrocytes to vascular endothelium decreases endothelial
nitric oxide synthase expression. Kidney Int 67: 1899–1906.
34. Linden E, Cai W, He JC, Xue C, Li Z, et al. (2008) Endothelial Dysfunction in
Patients with Chronic Kidney Disease Results from Advanced Glycation End
Products-Mediated Inhibition of Endothelial Nitric Oxide Synthase through
RAGE Activation. Clin J Am Soc Nephrol 3: 691–698.
35. Go ´mez-Ferna ´ndez P, Pe ´rez-Requena J, Sa ´nchez-Margalet V, Esteban J,
Murillo-Carretero M, et al. (2005) Vascular damage in chronic renal failure.
The increase of vascular nitrotyrosine and cytochines accumulation is
accompanied by an increase of endothelial nitric oxide synthase expression.
Nefrologia 25: 155–162.
36. Billecke SS, D’Alecy LG, Platel R, Whitesall SE, Jamerson KA, et al. (2009)
Blood content of asymmetric dimethylarginine: new insights into its dysregu-
lation in renal disease. Nephrol Dial Transplant 24: 489–496.
37. Antoniades C, Shirodaria C, Leeson P, Antonopoulos A, Warrick N, et al. (2009)
Association of plasma asymmetrical dimethylarginine (ADMA) with elevated
vascular superoxide production and endothelial nitric oxide synthase uncou-
pling: implications for endothelial function in human atherosclerosis. European
Heart Journal 30: 1142–1150.
38. Veresh Z, Racz A, Lotz G, Koller A (2008) ADMA Impairs Nitric Oxide-
Mediated Arteriolar Function Due to Increased Superoxide Production by
Angiotensin II-NAD(P)H Oxidase Pathway. Hypertension 52: 960–966.
39. Satoh M, Fujimoto S, Arakawa S, Yada T, Namikoshi T, et al. (2008)
Angiotensin II type 1 receptor blocker ameliorates uncoupled endothelial nitric
oxide synthase in rats with experimental diabetic nephropathy. Nephrology
Dialysis Transplantation 23: 3806–3813.
40. Satoh M, Fujimoto S, Haruna Y, Arakawa S, Horike H, et al. (2005) NAD(P)H
oxidase and uncoupled nitric oxide synthase are major sources of glomerular
superoxide in rats with experimental diabetic nephropathy. American Journal of
Physiology - Renal Physiology 288: F1144–F1152.
41. Yilmaz MI, Saglam M, Caglar K, Cakir E, Ozgurtas T, et al. (2005) Endothelial
Functions Improve with Decrease in Asymmetric Dimethylarginine Levels after
Renal Transplantation. Transplantation 80: 1660–1666.
42. Mittermayer F, Schaller G, Pleiner J, Vychytil A, Sunder-Plassmann G, et al.
(2005) Asymmetrical Dimethylarginine Plasma Concentrations Are Related to
Basal Nitric Oxide Release but Not Endothelium-Dependent Vasodilation of
Resistance Arteries in Peritoneal Dialysis Patients. J Am Soc Nephrol 16:
43. Achan V, Broadhead M, Malaki M, Whitley G, Leiper J, et al. (2003)
Asymmetric Dimethylarginine Causes Hypertension and Cardiac Dysfunction in
Humans and Is Actively Metabolized by Dimethylarginine Dimethylaminohy-
drolase. Arterioscler Thromb Vasc Biol 23: 1455–1459.
44. Albrecht E, van Goor H, Smit-van Oosten A, Stegeman C (2003) Long-term
dietary -arginine supplementation attenuates proteinuria and focal glomerulo-
sclerosis in experimental chronic renal transplant failure. Nitric Oxide 8: 53–58.
45. Wadham C, Mangoni AA (2009) Dimethylarginine dimethylaminohydrolase
regulation: a novel therapeutic target in cardiovascular disease. Expert Opin
Drug Metab Toxicol 5: 303–319.
Resistance Arteries in ESRD
PLoS ONE | www.plosone.org8 April 2012 | Volume 7 | Issue 4 | e36056