Plasma concentrations of extracellular matrix protein fibulin-1 are related to cardiovascular risk markers in chronic kidney disease and diabetes.

0.22 MB·English

by  Scholze, Alexandra

Scholzeetal.CardiovascularDiabetology2013,12:6 CARDIO
http://www.cardiab.com/content/12/1/6 VASCULAR
DIABETOLOGY
ORIGINAL INVESTIGATION Open Access
Plasma concentrations of extracellular matrix
protein fibulin-1 are related to cardiovascular risk
markers in chronic kidney disease and diabetes
Alexandra Scholze1*, Else-Marie Bladbjerg2, Johannes J Sidelmann2, Axel CP Diederichsen3, Hans Mickley3,
Mads Nybo5, W Scott Argraves4, Peter Marckmann1 and Lars M Rasmussen5
Abstract
Background: Fibulin-1 is one of a few extracellular matrix proteins present inblood inhighconcentrations. We
aimed to define therelationshipbetween plasma fibulin-1 levels and risk markersof cardiovascular disease.
Methods: Plasmafibulin-1was determined in subjects with chronic kidney disease (n=32; median age 62.5,
inter-quartile range 51 – 73 years) and 60 age-matched control subjects. Among kidney disease patients serological
biomarkers related to cardiovascular disease (fibrinogen, interleukin 6, C-reactive protein) were measured. Arterial
applanation tonometrywas used to determine central hemodynamic and arterial stiffness indices.
Results: We observed a positive correlationof fibulin-1 levels withage (r=0.38; p=0.033), glycated hemoglobin
(r=0.80; p=0.003), creatinine (r=0.35; p=0.045), and fibrinogen(r=0.39;p=0.027).Glomerular filtrationrate and
fibulin-1 were inversely correlated (r=−0.57; p=0.022). There was a positive correlation between fibulin-1 and
central pulse pressure (r=0.44; p=0.011)and central augmentation pressure (r=0.55;p=0.001). In a multivariable
regression model, diabetes, creatinine,fibrinogen and central augmentation pressure were independent predictors
ofplasma fibulin-1.
Conclusion: Increased plasma fibulin-1levels were associated with diabetes and impaired kidney function.
Furthermore, fibulin-1 levels were associated withhemodynamic cardiovascular risk markers. Fibulin-1 is a candidate
inthe pathogenesis of cardiovascular disease observed inchronic kidney disease and diabetes.
Keywords: Fibulin-1, Arterial stiffness, Cardiovascular disease,Kidney disease,Diabetes
Background isdeposited in the medial layers surroundingVSMCs and
Theextracellularmatrixproteinfibulin-1isemergingasa in association with elastic laminae and in the adventitial
new factor in cardiovascular disease. The plasma concen- ECM[1,4].
tration of fibulin-1 is a predictor of all cause and cardio- Fibulin-1 mRNA expression is increased in a model of
vascularmortalityinpatientswithdiabetesmellitus[1]. cardiomyopathy [5], and fibulin-1 plasma levels posi-
Fibulin-1 belongs to a family of extracellular matrix tivelycorrelatewith plasmaN-terminalpro-B-typenatri-
(ECM) proteins with functional associations with elastic uretic peptide, a cardiac marker of pressure or volume
fibers and basement membranes. Fibulin-1 interacts with overload[1,6].
a number of ECM molecules, but whether it serves a Several studies link fibulin-1 to atherosclerosis and
structural role has not yet been established [2]. Fibulin-1 thrombosis, including studies showing that itbinds fibri-
is expressed in embryonic and adult vascular smooth nogen, mediates platelet adhesion, is a component of
musclecells(VSMC)[3,4].Inadultbloodvesselsfibulin-1 newly formed fibrin-containing thrombi and that it
accumulatesinhuman atheroscleroticlesions [7-10].
Based on these findings fibulin-1 may be involved in
*Correspondence:[email protected]
1ClinicalResearchUnit,DepartmentofNephrology,OdenseUniversity the development or progression of cardiovascular disease.
Hospital,Kloevervaenget6,Odense5000,Denmark We tested the hypothesis that plasma fibulin-1 levels are
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Scholzeetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Scholzeetal.CardiovascularDiabetology2013,12:6 Page2of7
http://www.cardiab.com/content/12/1/6
associated with cardiovascular risk markers in patients Table1Populationcharacteristicsofchronickidney
withchronickidneydiseaseanddiabetesmellitus. diseasestudyparticipants
Characteristics
Methods Age,years 62.5(51–73)
Studysubjects
Sex,men,n(%) 25(78)
Chronic kidney disease (CKD) patients were recruited
Bodymassindex,kg/m2 24.8(21.7–28.1)
from the outpatient population of the Department
Smoking,n(%) 18(56)
of Nephrology, Odense University Hospital, Odense,
Denmark. Written informed consent was obtained Underlyingkidneydisease,n(%)
from each patient. The protocol was in accordance Diabeticnephropathy 2(6)
withtheethical standardsoftheDeclaration ofHelsinki Nephrosclerosis 9(28)
and was approved by the regional ethics committee
Glomerulonephritis 4(13)
(reference number: S-20090061). Fifty-seven measure-
Others 17(53)
ments were performed in 32 patients; 16 patients with
CKD stage 1 – 5 without hemodialysis therapy and Diseaseprevalence,n(%)
16 patients with hemodialysis therapy. In 25 of the Diabetes 11(34)
study participants two measurements per patient were Hypertension 25(78)
performed with an interval of 2 month. Population Peripheralarterydisease 2(6)
characteristics are described in Table 1.
Coronaryarterydisease 9(28)
CKD was defined by structural or functional abnor-
Stroke 5(16)
malities of the kidney with or without decreased glome-
eGFR,mL/min/1.73m2,* 39.9(23.3–58.3)
rular filtration rate (GFR) or estimated glomerular
filtration rate (eGFR) less than 60 mL/min/1.73 m2 for Kt/V† 1.1(0.9–1.4)
more than 3 months. Estimated GFR was calculated Medication,n(%)
according to the Modification of Diet in Renal Disease Phosphatebinder 16(50)
formula[11].
Erythropoietinanalog 15(47)
Hemodialysis (HD) patients were routinely dialyzed
Plateletaggregationinhibitor 8(25)
for 4 to 5 hours, three times weekly. Dialyses were per-
Diuretic 7(22)
formed using standardized techniques with bicarbonate-
based dialysates and controlled ultrafiltration rate. In ACEinhibitor/ATreceptorantagonist 14(44)
HD patients, the dialysis dose was calculated as the Calciumantagonist 15(47)
amount of plasma cleared of urea divided by the urea ß-Blocker 12(38)
distribution volume (Kt/V). Diabetes was defined using
Valuesaremedians(25%-75%percentile)ornumbers(percentages),n=32.
criteria of the American Diabetes Association or by a Abbreviations:ACEangiotensinconvertingenzyme,ATangiotensin,
historyoforcurrent treatmentfordiabetesmellitus. eGFRestimatedglomerularfiltrationrate,Kt/V=dialysisdose.
*n=16,eGFRdeterminationonlyinpatientswithouthemodialysistherapy.
Control samples for fibulin-1 determination were †n=8.
obtained from 60age-andsex-matched subjectswithout
kidney disease and without diabetes from the DanRisk PlasmasampleswereobtainedfromEDTA-anticoagulated
study population for the sole purpose of fibulin-1 con- bloodandstoredat−80°C.AnELISAforthedetermination
centration comparison (age, median 60 years, range of fibulin-1 concentration was performed as described by
60 to 61; sex, 46 men, 77%; BMI (body mass index), ourgrouppreviously[1].Briefly,96-wellplateswerecoated
median 27.1 kg/m2, range 24.7 to 30.7; smoking, 21 par- with rabbit anti human fibulin-1 IgG (Rb2954 IgG;
ticipants, 35%). Control samples were analyzed in the Argraves et al., Cell, 1989) and blocked with bovine serum
same timeframeandwiththesame pre-analytical condi- albumin [13]. Plasma samples were used at a 1:1000 dilu-
tions as the study samples. The DanRisk study popula- tionand,afterbindingtotheimmobilizedantibodies,were
tion hasbeendescribedpreviously [12]. incubatedwithanti-humanfibulin-1IgG(mousemonoclo-
nal, 3A11; Array Genetics, Newtown, Connecticut, USA).
ClinicalandBiochemicalAssessment Europium-labeled rabbit anti-mouse IgG (AD0124, Perkin
Examinations and blood samplings were performed Elmer, Skovlunde, Denmark) was used as secondary anti-
between 7 and 10 am. Examinations and blood sampling body quantified by time-resolved fluorescence on DELFIA
in HD patients were performed before HD sessions. (Perkin Elmer). Final fibulin-1 concentrations were derived
Analyses were performed at the laboratories of the par- fromastandardcurve.Theinter-assayvariationwas≤10%.
ticipating departments according to current laboratory PlasmasamplesforC-reactiveprotein(CRP),fibrinogen,
standards. and interleukin-6 (IL6) were stored at −80°C. Fibrinogen
Scholzeetal.CardiovascularDiabetology2013,12:6 Page3of7
http://www.cardiab.com/content/12/1/6
and CRP were determined with a nephelometric method Categorical variables are reported as numbers and per-
(Siemens Healthcare Diagnostics Inc., Marburg, Germany). centages. Non-parametric bivariate correlation analysis
A commercial ELISA was used to measure concentrations (Spearman) was performed. Multiple linear regression
ofIL6(QuantikineHS,R&DSystems,Abingdon,UK).The analysis with backward selection was used to judge vari-
inter-assaycoefficientsof variationwere:<10%forfibrino- ables of significance for the prediction of plasma fibulin-
gen,<6%forCRP,<7%forIL6. 1concentration.
Glycated hemoglobin (HbA ) was determined using Analyses were performed with GraphPad prism soft-
1C
high-performance liquid chromatography as fraction ware (version 5.0, GraphPad Software, San Diego, CA,
of total hemoglobin A0. Analyses were performed on USA)andSPSSsoftware(release17.0,SPSSInc.,Chicago,
a Tosoh G7 automatic analyzer (Medinor, Broendby, IL, USA). All statistical tests were two-sided and p-values
Denmark). less than 0.05 were considered to indicate statistical
Blood pressure (BP) was measured after at least significance.
10 minutes of supine resting in a quiet room prior to
blood sampling. Themeasurementswereperformed with Results
an automated device (Omron M6, Omron HealthCare Thefibulin-1studygroupcontained32patients:3patients
EuropeB.V.,Hoofddorp,Netherlands). with CKD 2 (eGFR more than 60 ml/min/1.73 m2),
For pulse wave analysis an applanation tonometer 7 patients with CKD stage 3 (eGFR 30–59 ml/min/
(Millar, SPT-301B, Houston, Texas, USA) was applied 1.73 m2), 5 patients with CKD stage 4 (eGFR 15–29 ml/
to the radial artery. The recorded radial pressure wave- min/1.73 m2), 1 patient with CKD stage 5 without
forms and a corresponding brachial BP can be used to hemodialysis therapy (eGFR less than 15 ml/min/
generate the ascending aortic pressure waveform. For 1.73 m2). The remaining 16 patients were on chronic
this purpose a validated transfer function is used by hemodialysis treatment; their median dialysis vintage
the SphygmoCorW system (version 7.0, Atcor Medical, was 42 (13–89) months. The clinical and biochemical
Sydney, Australia) [14]. We analyzed systolic aortic BP variables are given in Table 2.
(SBP ), diastolic aortic BP (DBP ), and aortic Figure1showsthedistributionofplasmafibulin-1con-
aortic aortic
pulse pressure (PP ). centrations in patients with chronic kidney disease in the
aortic
The effect of pulse wave reflection on the incident absenceand presenceof diabetes mellitus as comparedto
central pulse wave was evaluated using central augmen- age- and sex-matched control subjects without kidney
tation pressure (CAP), which was determined from the disease and without diabetes (age, median 60 years, range
difference between the first and the second aortic sys- 60 to 61; sex, 46 men, 77%; BMI, median 27.1 kg/m2,
tolic pressure wave peaks. Aortic augmentation index range 24.7 to 30.7; smoking, 21 participants, 35%). Com-
(AIx ) was derived from CAP expressed as percen- pared to control subjects, patients with chronic kidney
aortic
tage of the central pulse pressure. Since the AIx depends diseaseshowedhigherfibulin-1levels(p<0.001).Further-
on heart rate we also analyzed the AIx normalized to a more, we observed significantly increased fibulin-1 con-
heart rate of 75 beats per minute (AIx@75 ). Aortic centrations in patients with chronic kidney disease plus
aortic
pulse wave velocity (PWV ) was assessed as a direct diabetes mellitus (median, 78 μg/mL; 74 to 113 μg/mL)
aortic
measure of arterial stiffness and determined between a comparedtopatientswithchronickidneydiseasewithout
carotid artery and femoral artery recording site [15]. The diabetes mellitus (median, 69 μg/mL; 49 to 79 μg/mL;
distance between the two measurement sites was deter- p<0.050).
mined by the subtraction method, hence by subtracting No differences in plasma fibulin-1 concentrations were
the distance between the sternal notch to the carotid observedbetweenmalesandfemales(Mann–Whitneytest;
artery from the sternal notch to the femoral artery. p=0.964) or smokers and non-smokers (Mann–Whitney
Patients rested for 10 minutes in a supine position test;p=0.444).
before themeasurements. Nextweanalyzedbivariate correlationsbetweenplasma
fibulin-1 concentration and clinical and biochemical vari-
Statistics ables (Table 3). We observed a positive correlation of
In those patients were the measurements were per- fibulin-1 with age and HbA concentration. Fibulin-1
1C
formed twice the mean of the obtained values was used levels correlated with markers of kidney function and
for the analyses. Normal distribution of continuous vari- uremia (P-creatinine, P-urea, and eGFR) with fibulin-1
ables was tested by D’Agostino & Pearson omnibus nor- levels increasing with deteriorating kidney function/
mality test. Fibrinogen, IL6, CRP, creatinine, and PWV increasing uremia. In hemodialysis patients, fibulin-1
did not show a normal distribution. Non-parametric concentrations were not correlated with dialysis dose
tests were therefore applied where necessary. Continu- (Kt/V, r=−0.43; p=0.300) or dialysis vintage (r=0.17;
ous data are reported as median (25% - 75% percentile). p=0.537).
Scholzeetal.CardiovascularDiabetology2013,12:6 Page4of7
http://www.cardiab.com/content/12/1/6
Table2Biochemical,hemodynamicandvascular Table3Bivariatecorrelationanalysisofplasmafibulin-1
variablesofchronickidneydiseasestudyparticipants inchronickidneydiseasestudyparticipantsinrelationto
Variable clinicalandbiochemicalvariables(n=32)
P-Fibulin-1,μg/mL 73.9(54.9–85.3) Variable r p
HbA* ,% 7.2(5.6–8.2) Age 0.38 0.033
1C
P-Fibrinogen,μmol/L 11.7(10.0–13.4) HbA1C* 0.80 0.003
P-IL6,pg/mL 5.03(3.50–8.36) P-Fibrinogen 0.39 0.027
P-CRP,mg/L 4.30(1.52–11.86) P-IL6 0.25 0.174
S-Albumin,g/l 41(39–44) P-CRP 0.24 0.195
P-Creatinine,μmol/L 380(171–701) S-Albumin −0.30 0.102
P-Urea,mmol/L 15.2(11.7–19.4) P-Creatinine 0.35 0.049
SBP ,mmHg 138(130–152) P-Urea 0.51 0.004
brachial
DBP ,mmHg 77(70–86) eGFR† −0.57 0.022
brachial
SBP ,mmHg 128(115–140) SBPaortic 0.19 0.302
aortic
DBP ,mmHg 79(71–87) DBPaortic −0.41 0.019
aortic
PP ,mmHg 49(34–59) PPaortic 0.44 0.011
aortic
Heartrate,beats/min 70(63–82) Heartrate −0.35 0.047
CAP,mmHg 13(8–21) CAP 0.55 0.001
AIx ,% 31(23–35) AIxaortic 0.47 0.007
aortic
AIx@75 ,% 26(21–32) AIx@75aortic 0.23 0.203
aortic
PWV ,m/s 10.1(7.9–13.3) PWVaortic 0.26 0.166
aortic
Abbreviations:HbA glycatedhemoglobin,IL6interleukin6,CRPC-reactive
Valuesaremedians(25%-75%percentile),n=32. 1C
protein,eGFRestimatedglomerularfiltrationrate,SBPsystolicbloodpressure,
Abbreviations:HbA glycatedhemoglobin,IL6interleukin6,CRPC-reactive
1C DBPdiastolicbloodpressure,PPpulsepressure,CAPcentralaugmentation
protein,SBPsystolicbloodpressure,DBPdiastolicbloodpressure,PPpulse
pressure,AIxaugmentationindex,AIx@75augmentationindexataheartrate
pressure,CAPcentralaugmentationpressure,AIxaugmentationindex,AIx@75
of75beats/min,PWVpulsewavevelocity.
augmentationindexataheartrateof75beats/min,PWVpulsewavevelocity.
*n=12. *n=12,
†n=16,eGFRdeterminationonlyinpatientswithouthemodialysistherapy.
As indicated in Table 3 we observed a positive corre-
lation between plasma levels of fibrinogen, a marker of
150 p < 0.001 coagulation and systemic inflammation, and plasma fibu-
lin-1, whereas CRP and IL6 levels were not significantly
) p < 0.001
L associated.
m
We also evaluated the degree to which plasma fibulin-1
/ 100
g
µ levels were associated with hemodynamic variables and
(
established markers of arterial stiffness. Several key vari-
n
uli ablesofcentralhemodynamicsshowedsignificantassocia-
b 50 tions with plasma fibulin-1 levels (Table 3). A positive
Fi
P- correlation was found with aortic pulse pressure (PPaortic)
andcentralaugmentationpressure(CAP),whileanegative
0 correlationexistedwithheartrate.Markersofarterialstiff-
Control CKD CKD
ness were also investigated. No association was observed
-Diab +Diab
between plasma fibulin-1 and central pulse wave velocity.
Figure1Distributionofplasmafibulin-1concentrationsin Since the extent of pulse wave augmentation is also heart
relationtodiabetesoccurrenceandkidneyfunction.Distribution rate dependent we used aortic augmentation index and
ofplasmafibulin-1concentrationsinage-andsex-matchedcontrol
heart rate normalized AIx (AIx @75 ). We did not
subjectswithoutchronickidneydiseaseandwithoutdiabetes aortic
observe a significant association of plasma fibulin-1 con-
(Control,n=60)andpatientswithchronickidneydisease(CKD)
with(+Diab,n=11)orwithoutdiabetesmellitus(−Diab,n=21). centrationswithAIx@75aortic.
Theplasmafibulin-1valuesofthegroupsshowedasignificantly Multivariableregressionanalysiswasperformedtoassess
differentdistribution(Kruskal-Wallistestp<0.001).Thedifferences the independent contributions of different variables to
obtainedbyDunn’smultiplecomparisontestareindicatedinthe
plasmafibulin-1concentration.Accordingtotheresultsof
figure.Scholzeetal.,Figure1.
the bivariate correlation analysis age, presence of diabetes,
Scholzeetal.CardiovascularDiabetology2013,12:6 Page5of7
http://www.cardiab.com/content/12/1/6
creatinine,andfibrinogenasamarkerofinflammationand Our previous studies demonstrated elevated plasma
coagulation was taken into account. To avoid multicolli- concentrations of fibulin-1 in diabetic patients [1]. We
nearity between the hemodynamic variables, CAP was confirmed these results and again saw a correlation bet-
chosen for the analyses. The results are presented in ween higher plasma fibulin-1 concentrations and higher
Table 4. The presence of diabetes, creatinine, CAP, and HbA values. The presence of diabetes was an inde-
1C
plasmafibrinogenconcentrationtogetherexplained59%of pendentpredictorofelevatedplasma fibulin-1 levels.
thevariabilityofplasmafibulin-1concentrationinpatients The mechanism underlying the increase of plasma
withCKD. fibulin-1 is not known. In patients with diabetes we
showed that it exists also in those patients without
Discussion microalbuminuria hence without signs of initial kidney
In the present study we show a positive association bet- damage [1]. It should be noted, that diabetes and CKD
weenplasmafibulin-1concentrationandseveralcardiovas- share important features at the molecular level, e.g. an
cular risk markers in patients with CKD. Our findings are increase in oxidative stress and advanced protein glyca-
in accordance with findings from earlier experimental and tion [20,21]. One or several factors from this group
clinical studies which suggested links between fibulin-1, might be involved in the regulation of plasma fibulin-1
cardiovasculardiseaseanddiabetes[7,9,16,17]. concentration. This should be the subject of future ex-
Inthepresentstudy,plasmafibulin-1concentrationwas perimentalwork.
found to increase with age in the study population. This An important dimension of the present study was the
is in agreement with our data previously published [1]. analysisofcentralhemodynamicsandvascularstiffnessin
A multivariable analysis of the current data showed that relation to plasma fibulin-1 concentrations. We found a
the importance of age is not retained in the simultaneous correlation with aortic AIx and central augmentation
contextofkidneyfunction,presenceofdiabetes,CAPand pressure (CAP). CAP was retained as independent var-
fibrinogenstatus. iable in the multivariable analysis. When we performed
Correlations were also observed between markers of de- analyses of arterial stiffness markers we did not find a
teriorating kidney function and increasing plasma fibulin-1 correlation of plasma fibulin-1 concentrations to aortic
concentrations, with creatinine being an independent pre- PWVand heart rate normalized aortic AIx. Thus, plasma
dictorofplasmafibulin-1concentration.Thesefindingsare fibulin-1 seems to be unrelated to arterial changes that
in accordance with findings from a multiplex proteomic result in changes of aortic PWV. Also, increased arterial
study showing that plasma fibulin-1 could be a marker of stiffnessthatresultsinanincreasedheart ratenormalized
renalimpairment[18]. aortic AIx [22] was unrelated to plasma fibulin-1. By
We also found an independent association between contrast,arelationshipwasfoundtoexistbetweenplasma
fibulin-1 and fibrinogen, a positive acute phase protein fibulin-1 and the actual central augmentation the left
whose blood levels are increased in CKD. Fibrin clot ventricle faces in the late systole. The pressure effort of
characteristics in CKD differ from non-CKD patients in the ventricle to overcome this augmented pressure was
close correlation to inflammatory level rather than to termed“wasted”sinceitdoesnotcontributetobloodflow
the extent of uremia [19]. Not only has fibulin-1 been production[23].
shown to bind to fibrinogen and incorporate into fibrin Several potential new biomarkers for cardiovascular
thrombi, but fibulin-1 also qualitatively influences the diseases in diabetes have appeared during recent years.
polymerization of fibrin [7,9,10]. We regard these fin- These biomarkers are associated with dysfunctions in
dings as important in the context of increased plasma different disease pathways and consequently display
fibulin-1concentrations thatwereporthere. both, similarities and discrepancies in their clinical
appearance.
Table4Multivariableregressionanalysisonplasma While it was shown that plasma fibulin-1 was able to
fibulin-1 predict all-cause mortality in patients with diabetes [1]
Independentvariable Adjustedr2 F β p the underlying mechanisms which regulate the protein
model 0.59 11.98 <0.001 amount in certain disease entities are still uncertain. A
similar dilemma applies for another interesting protein
Diabetes 0.40 0.005
that has extensively been studied in cardiovascular
Creatinine 0.43 0.002
disease – osteoprotegerin. Plasma osteoprotegerin con-
CAP 0.25 0.073
centrations are positively correlated to the presence of
Fibrinogenlog 0.32 0.011 coronary, carotid and peripheral artery disease in dia-
Age,presenceofdiabetes,creatinine,centralaugmentationpressure,and betic patients [24,25]. Elevated plasma osteoprotegerin
logarithmicallytransformedplasmafibrinogenconcentrationwereusedina
concentrations also predicted all-cause mortality in
stepwisebackwardselectionprocess(n=32).
Abbreviations:CAPcentralaugmentationpressure. diabetes [26]. Several interesting mechanisms for the
Scholzeetal.CardiovascularDiabetology2013,12:6 Page6of7
http://www.cardiab.com/content/12/1/6
involvement of osteoprotegerin in cardiovascular morbi- Competinginterests
dity, like an increased expression of adhesion molecules Theauthorsdeclarethattheyhavenocompetinginterests.
byendothelialcellsoraregulativeeffectonvascularcalci-
Authors’contributions
fication, were described (for review see [27]). Although
PMandLMRdesignedresearch;AS,PM,EMB,JJS,MN,ACPD,HM,WSA,
the function of fibulin-1 in the cardiovascular system is andLMRconductedresearch;AS,PM,JJS,MN,ACPD,HM,WSA,andLMR
not well defined it could be speculated that it is different wrotethepaper;AShadprimaryresponsibilityforfinalcontent.Allauthors
read,criticallyrevisedandapprovedthefinalmanuscript.
from osteoprotegerin; more related to changes in the
structure of the extracellular matrix. Such functional Authordetails
differences between biomarkers are likely to give rise to a 1ClinicalResearchUnit,DepartmentofNephrology,OdenseUniversity
Hospital,Kloevervaenget6,Odense5000,Denmark.2UnitforThrombosis
different clinical meaning. Nevertheless, the proof of a
Research,InstituteofPublicHealth,DepartmentofClinicalBiochemistry,
cause-effect relation between cardiovascular disease in HospitalofSouthwestDenmark,UniversityofSouthernDenmark,Esbjerg,
humans and osteoprotegerin or fibulin-1 is still lacking Denmark.3DepartmentofCardiology,OdenseUniversityHospital,Odense,
Denmark.4DepartmentofRegenerativeMedicineandCellBiology,Medical
and the role of both proteins for diagnosis or assessment
UniversityofSouthCarolina,Charleston,SC,USA.5DepartmentofClinical
of progression of cardiovascular disease awaits further Biochemistry,OdenseUniversityHospital,Odense,Denmark.
study.
Received:12December2012Accepted:3January2013
The exact role of fibulin-1 in cardiovascular disease
Published:7January2013
may be complex since it is associated with pathologies
in both, arterial and heart tissue. In diabetes an accumu- References
lation of fibulin-1 in the arterial wall and in plasma has 1. CangemiC,SkovV,PoulsenMK,FunderJ,TwalWO,GallMA,HjortdalV,
JespersenML,KruseTA,AagardJ,ParvingHH,KnudsenS,Høilund-Carlsen
been described. Plasma fibulin-1 correlated significantly
PF,RossingP,HenriksenJE,ArgravesWS,RasmussenLM:Fibulin-1isa
with left atrial volume index and plasma N-terminal markerforarterialextracellularmatrixalterationsintype2diabetes.
pro-B-type natriuretic peptide [1]. On the other hand, a ClinChem2011,57:1556–1565.
2. ArgravesWS,GreeneLM,CooleyMA,GallagherWM:Fibulins:
down-regulation of fibulin-1 has been described in atrial physiologicalanddiseaseperspectives.EMBORep2003,4:1127–1131.
fibrillation[28]. 3. HungerfordJE,HoefflerJP,BowersCW,DahmLM,FalchettoR,
With respect to the above described results from our ShabanowitzJ,HuntDF,LittleCD:Identificationofanovelmarkerfor
primordialsmoothmuscleanditsdifferentialexpressionpatternin
study and from previously presented investigations we contractilevsnoncontractilecells.JCellBiol1997,137:925–937.
find it reasonable to speculate that in clinical conditions 4. RoarkEF,KeeneDR,HaudenschildCC,GodynaS,LittleCD,ArgravesWS:
with increased plasma fibulin-1 such as diabetes or Theassociationofhumanfibulin-1withelasticfibers:an
immunohistological,ultrastructural,andRNAstudy.JHistochemCytochem
chronic kidney disease plasma fibulin-1 may be involved 1995,43:401–411.
in disease pathways which contribute to thrombotic and 5. RedfernCH,DegtyarevMY,KwaAT,SalomonisN,CotteN,NaneviczT,
cardiovascular complications. Furthermore, plasma FidelmanN,DesaiK,VranizanK,LeeEK,CowardP,ShahN,WarringtonJA,
FishmanGI,BernsteinD,BakerAJ,ConklinBR:Conditionalexpressionofa
fibulin-1displays apotential ascardiovascularbiomarker
G-coupledreceptorcausesventricularconductiondelayandalethal
i
which awaitsfurtherstudies. cardiomyopathy.PNAS2000,97:4826–4831.
6. KrugerR,SchutteR,HuismanHW,ArgravesWS,RasmussenLM,OlsenMH,
SchutteAE:NT-proBNPisassociatedwithfibulin-1inAfricans:
Conclusions TheSAfrEICstudy.Atherosclerosis2012,222:216–221.
In summary, we demonstrate an increase in plasma 7. GodynaS,Diaz-RicartM,ArgravesWS:Fibulin-1mediatesplatelet
adhesionviaabridgeoffibrinogen.Blood1996,88:2569–2577.
fibulin-1 concentration in association with diabetes and 8. HolvoetP,CollenD:Thrombosisandatherosclerosis.CurrOpinLipidol
impaired kidney function. A positive association exists 1997,8:320–328.
between plasma fibulin-1 and fibrinogen levels. Also, an 9. TranH,TanakaA,LitvinovichSV,MedvedLV,HaudenschildCC,
ArgravesWS:Theinteractionoffibulin-1withfibrinogen.Apotentialrole
increase of central augmentation pressure and augmen- inhemostasisandthrombosis.JBiolChem1995,270:19458–19464.
tation index resulting in higher left ventricular systolic 10. ArgravesWS,TanakaA,SmithEP,TwalWO,ArgravesKM,FanD,
loading is associated with higher fibulin-1 plasma con- HaudenschildCC:Fibulin-1andfibrinogeninhumanatherosclerotic
lesions.HistochemCellBiol2009,132:559–565.
centrations. These associations of plasma fibulin-1 are 11. LeveyAS,BoschJP,LewisJB,GreeneT,RogersN,RothD:Amoreaccurate
compatible with the idea that fibulin-1 may have a role methodtoestimateglomerularfiltrationratefromserumcreatinine:
inthe complexpathogenesisofcardiovascular disease. anewpredictionequation.Modificationofdietinrenaldiseasestudy
group.AnnInternMed1999,130:461–470.
12. DiederichsenAC,SandNP,NørgaardB,LambrechtsenJ,JensenJM,
Abbreviations MunkholmH,AzizA,GerkeO,EgstrupK,LarsenML,PetersenH,
ACE:Angiotensinconvertingenzyme;AIx:Augmentationindex; Høilund-CarlsenPF,MickleyH:Discrepancybetweencoronaryartery
AIx@75:AIxnormalizedtoaheartrateof75beatsperminute; calciumscoreandHeartScoreinmiddle-agedDanes:theDanRiskstudy.
AT:Angiotensin;BMI:Bodymassindex;BP:Bloodpressure; EurJPrevCardiol2012,19:558–564.
CAP:Centralaugmentationpressure;CKD:Chronickidneydisease; 13. ArgravesWS,DickersonK,BurgessWH,RuoslahtiE:Fibulin,anovelprotein
CRP:C-reactiveprotein;DBP:Diastolicbloodpressure; thatinteractswiththefibronectinreceptorβsubunitcytoplasmic
ECM:Extracellularmatrix;GFR/eGFR:Glomerularfiltrationrate/estimated domain.Cell1989,58:623–629.
glomerularfiltrationrate;HbA :Glycatedhemoglobin;HD:Hemodialysis; 14. PaucaAL,O’RourkeMF,KonND:Prospectiveevaluationofamethodfor
1C
IL6:Interleukin-6;Kt/V:Dialysisdose;PP:Pulsepressure;SBP:Systolicblood estimatingascendingaorticpressurefromtheradialarterypressure
pressure;VSMC:Vascularsmoothmusclecells. waveform.Hypertension2001,38:932–937.
Scholzeetal.CardiovascularDiabetology2013,12:6 Page7of7
http://www.cardiab.com/content/12/1/6
15. Frimodt-MøllerM,NielsenAH,KamperAL,StrandgaardS:Reproducibilityof
pulse-waveanalysisandpulse-wavevelocitydeterminationinchronic
kidneydisease.NephrolDialTransplant2008,23:594–600.
16. ChanMY,LinM,LucasJ,MoseleyA,ThompsonJW,CyrD,UedaH,
KajikawaM,OrtelTL,BeckerRC:Plasmaproteomicsofpatientswith
non-valvularatrialfibrillationonchronicanti-coagulationwithwarfarin
oradirectfactorXainhibitor.ThrombHaemost2012,108:1180–1191.
17. KaurP,RizkNM,IbrahimS,YounesN,UppalA,DennisK,KarveT,
BlakesleeK,KwagyanJ,ZirieM,RessomHW,CheemaAK:iTRAQ-based
quantitativeproteinexpressionprofilingandMRMverificationof
markersintype2diabetes.JProteomRes2012,11:5527–5539.
18. NeimanM,HedbergJJ,DönnesPR,Schuppe-KoistinenI,HanschkeS,
SchindlerR,UhlénM,SchwenkJM,NilssonP:Plasmaprofilingreveals
humanfibulin-1ascandidatemarkerforrenalimpairment.
JProteomRes2011,10:4925–4934.
19. SjølandJA,SidelmannJJ,BrabandM,PedersenRS,EsbensenK,
StandevenKF,AriënsRA,GramJ:Fibrinclotstructureinpatientswith
end-stagerenaldisease.ThrombHaemost2007,98:339–345.
20. KoyamaH,NishizawaY:AGEs/RAGEinCKD:irreversiblemetabolic
memoryroadtowardCVD?EurJClinInvest2010,40:623–635.
21. ThornalleyPJ,RabbaniN:Proteindamageindiabetesand
uremia-identifyinghotspotsofproteomedamagewhereminimal
modificationisamplifiedtomarkedpathophysiologicaleffect.
FreeRadicRes2011,45:89–100.
22. O’RourkeMF,HashimotoJ:Arterialstiffness–Amodifiablecardiovascular
riskfactor?JCardiopulmRehabilPrev2008,28:225–237.
23. DenardoSJ,NandyalaR,FreemanGL,PierceGL,NicholsWW:
Pulsewaveanalysisoftheaorticpressurewaveforminsevereleft
ventricularsystolicdysfunction.CircHeartFail2010,3:149–156.
24. ReinhardH,NyboM,HansenPR,WiinbergN,KjærA,PetersenCL,
WintherK,ParvingHH,RasmussenLM,RossingP,JacobsenPK:
Osteoprotegerinandcoronaryarterydiseaseintype2diabeticpatients
withmicroalbuminuria.CardiovascDiabetol2011,10:70.
25. PoulsenMK,NyboM,DahlJ,HosbondS,PoulsenTS,JohansenA,
Høilund-CarlsenPF,Beck-NielsenH,RasmussenLM,HenriksenJE:
Plasmaosteoprotegrinisrelatedtocarotidandperipheralarterial
disease,butnottomyocardialischemiaintype2diabetesmellitus.
CardiovascDiabetol2011,10:76.
26. ReinhardH,LajerM,GallMA,TarnowL,ParvingHH,RasmussenLM,
RossingP:Osteoprotegerinandmortalityintype2Diabeticpatients.
DiabetesCare2010,33:2561–2566.
27. VenurajuSM,YerramasuA,CorderR,LahiriA:Osteoprotegerinasa
predictorofcoronaryarterydiseaseandcardiovascularmortalityand
morbidity.JAmCollCardiol2010,55:2049–2061.
28. GarciaA,EirasS,ParguinaAF,AlonsoJ,RosaI,Salgado-SomozaA,RicoTY,
Teijeira-FernandezE,Gonzales-JuanateyJR:High-resolution
two-dimensionalgelelectrophoresisanalysisofatrialtissueproteome
revealsdown-regulationoffibulin-1inatrialfibrillation.IntJCardiol2011,
150:283–290.
doi:10.1186/1475-2840-12-6
Citethisarticleas:Scholzeetal.:Plasmaconcentrationsofextracellular
matrixproteinfibulin-1arerelatedtocardiovascularriskmarkersin
chronickidneydiseaseanddiabetes.CardiovascularDiabetology2013
12:6.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit