Atherogenic Vascular and Lipid Phenotypes in Young Type 1 Diabetics are associated with

Diabetes High Risk HLA genotype

Brief title: Diabetes High Risk HLA and early atherosclerosis

Authors:

1Michal Odermarsky, MD, 2Anita Nilsson, MSc, 2Åke Lernmark, MD, PhD, 3Sture Sjöblad, MD, PhD,

1Petru Liuba P, MD, PhD

Affiliation:

Pediatric 1Cardiology and 3Endocrinology, Lund University Hospital, Lund, and 2Department of Clinical

Sciences, University Hospital MAS, Lund University, Malmö, Sweden

Corresponding author:

Petru Liuba, PhD, MD

Division of Cardiology, Department of Paediatrics

Lund University Hospital

SE-22185 Lund, SWEDEN

Tel: +46 46 17 82 67 / Fax: + 46 415 51606 / E-mail: petru.liuba@med.lu.se

Page 1 of 21

Copyright Information

Articles in PresS. Am J Physiol Heart Circ Physiol (September 28, 2007). doi:10.1152/ajpheart.00795.2007

Copyright © 2007 by the American Physiological Society.

2

ABSTRACT

Background: Expression of human leukocyte antigen (HLA) class II molecules on islet endothelial

cells (EC) is a central vascular event in the pathogenesis of type 1 diabetes. Previous studies

demonstrated the ability of other vascular EC to express HLA and thereby to process islet autoantigens

on their surface. We investigated whether the HLA-DQ2/8 genotype, which confers the highest risk

for type 1 diabetes, is associated with early atherosclerosis in youths with this disease.

Methods: Brachial artery endothelium-dependent, flow-mediated dilation (BA-FMD) and carotid

artery intima-media thickness (CA-IMT), as well as markers of systemic inflammation (C-reactive

protein, fibrinogen, and orosomucoid), HbA1C, LDL, HDL and total cholesterol (TC), were assessed

in 86 children and adolescents with type 1 diabetes (mean age and diabetes duration: 15 and 7 years,

respectively) between 2004 and 2006. HLA genotypes were determined in dried blood spots by an

oligoblot hybridization method.

Results: HLA-DQ2/8 was detected in 34 patients (“DQ2/8”). When this group was compared to the

remaining patients (“non-DQ2/8“, n=52), there were no differences in age, diabetes duration, HbA1C,

body mass index, inflammatory markers, IMT, and TC (p>0.4). In the DQ2/8 group, LDL-to-HDL

ratio was elevated as compared to the non-DQ2/8 group (1.8 vs 1.3, respectively; p=0.001), while

FMD did not significantly differ between the groups (5.3 vs 6.7%, respectively; p=0.08). When

patients were further categorized in relation to CRP (cut-off value 1 mg/l), BA-FMD was significantly

lower (3%, p<0.01), whereas LDL-to-HDL ratio increased further (2.2, p<0.001) in the subgroup of

“DQ2/8”&“CRP>1” patients compared to the remaining three subgroups. These associations remain

significant after adjustment for age, diabetes duration, and HbA1C by ANCOVA. The brachial artery

responses to nitroglycerine were similar in all subgroups.

Conclusion: In conclusion, the diabetes-predisposing HLA-DQ2/8 genotype in children and

adolescents with type 1 diabetes interferes with endothelial and lipid-related mechanisms of early

atherosclerosis, possibly in part through inflammatory pathways.

Key-words: type 1 diabetes; genetics; lipids; vascular function

Page 2 of 21

Copyright Information

3

CONDENSED ABSTRACT

We investigated in 86 children and adolescents with type 1 diabetes whether the HLA-DQ2/8

genotype could influence atherogenic vascular and lipid phenotypes. In the DQ2/8 group, LDL-to-

HDL ratio was significantly elevated (1.8 vs 1.3, p<0.01), while brachial artery`s flow-mediated

dilation (BA-FMD) was only slightly decreased (5.3 vs 6.7%, p=0.08) as compared to the non-DQ2/8

group. When patients were further categorized in relation to CRP (cut-off value 1 mg/l), BA-FMD was

significantly lower (3%, p<0.01) and LDL-to-HDL ratio increased further (3.2, p<0.001) in the

subgroup of “DQ2/8”&”CRP>1” patients compared to the remaining three subgroups.

Page 3 of 21

Copyright Information

4

INTRODUCTION Type 1 diabetes mellitus occurs most frequently in children and adolescents, and is of growing

concern particularly in industrialized countries, including Sweden and the United States, as its

incidence continues to rise (3).

A genetic susceptibility involving class II human leukocyte antigen (HLA) genes

is recognized in more than 80 % of young patients with type 1 diabetes (13). The primary loci of

genetic susceptibility to type 1 diabetes have been mapped to the HLA-DQ region, which is

located on chromosome 6 (23). Two HLA haplotypes DQB1*0302-A1*0301 (DQ8), and

DQB1*0201-A1*0501 (DQ2) appear to confer the highest risk for developing type 1 diabetes

(27), especially when both are present in the genotype (ie, HLA-DQ2/8).

Previous studies suggested that expression of HLA-DQ molecules on the surface of

pancreatic microvascular endothelial cells could be an important vascular pathway in the

autoimmune lymphocyte-mediated reaction that leads to β-cell destruction, the underlying

pathology of type 1 diabetes (26, 17). This process is seemingly not confined to the pancreas

since other organs such as gut and thyroid may be affected as well (1). Recent observation that

aortic vascular endothelial cells also owe the ability to express HLA (6) lends support to this

hypothesis.

An important question is whether the HLA-mediated endotheliopathy could also

occur at the site of the conductance arteries, where atherosclerosis incepts. Type 1 diabetics are at

significant risk for cardiovascular disease in the adult life (2). Early and accelerated

atherosclerosis appears to play an important role in the excess cardiovascular morbidity (2, 5).

In light of the afore-mentioned findings, and given the importance of vascular

endothelial cells throughout the course of atherosclerosis (4) including in diabetes (22), we

investigated whether the diabetes-predisposing HLA genotype DQ2/8 could interfere with lipid

and vascular phenotypes of early atherosclerosis in children and adolescents with type 1 diabetes.

Page 4 of 21

Copyright Information

5

METHODS

A. Study population

Eighty-six children and adolescents aged between 7 – 22 years (mean age 15; 49 males/ 37

females) with type 1 diabetes for at least 6 months (mean duration 7 years) were randomly

recruited from the diabetes outpatient clinic at Children’s Hospital Lund. All patients were on

insulin treatment (Lantus® and Novorapid®). Four patients were on thyroid hormone replacement

therapy (Levaxin®). Exclusion criteria were: familiar hypercholesterolemia, active smoking,

history of premature coronary or cerebrovascular disease among first-degree relatives, and

systemic hypertension. Body weight, height, arterial blood pressure (systolic and diastolic) and

blood glucose were measured upon the ultrasound visit. Data on demographic information,

parental education and current occupation, family and personal history for major cardiovascular

risk factors (primary hypercholesterolemia, hypertension, premature coronary and

cerebrovascular disease) were assessed by questionnaire. Data regarding diabetes duration were

obtained from the registry of the outpatient diabetes clinic.

The study was approved by the ethical committee for human research at the Lund

University. Written consent was obtained from all participants aged 18 years and older or, if

under 18, from their guardians. All participants gave oral consent.

B. HLA typing, inflammatory, and lipid analyses

HLA typing was performed from dried blood spots by DELFIA® method. Briefly, DNA in the

blood was amplified and the presence of the particular alleles was determined by a hybridization

reaction using allele-specific, short oligonucleotides labelled with lanthanide chelates (14).

High-density lipoprotein (HDL), low-density lipoprotein (LDL) and total cholesterol (TC) were

analysed by an enzymatic method (Roche/Hitachi 912, Roche Diagnostics). Plasma high-

sensitivity C-reactive protein (CRP) was measured by enzyme-linked immunoassay using

Page 5 of 21

Copyright Information

6

polyclonal antibodies (DACO Diagnostics, Glostrup, Denmark). Plasma orosomucoid and

fibrinogen were assessed by specific immunoassays.

C. Assessment of endothelium-dependent dilation of the brachial artery

The dilatory responses to hyperaemia (endothelium-dependent agonist) and glycerol trinitrate

(GTN, endothelium-independent) were obtained in 70 patients. Briefly, longitudinal scans of the

brachial artery (nondominant arm) were imaged several centimeters above the antecubital fossa

via a 15-MHz linear ultrasound transducer of an Acuson Sequoia C256 (Siemens AG,

Germany). The ultrasound beam frequency was set at 8 MHz. Once the image was obtained, the

transducer was positioned throughout the ultrasound study with aid of a transducer arm holder

(Great Ormond Street Hospital, London, UK). ECG-gated end-diastolic scans of the artery were

recorded at baseline, and a pressure cuff tourniquet placed around the forearm was thereafter

inflated to 200 mmHg (minimum 50 mmHg over the systolic blood pressure) for 5 minutes. A

new series of frames were taken for 15 seconds before and 120 seconds after cuff deflation.

Arterial flow velocity was obtained prior to and during the first 15 seconds after cuff release by

pulsed Doppler signal at 70° to the vessel with the range gate in the center of the artery. Blood

flow volume was calculated by multiplying the velocity-time integral of Doppler signal by heart

rate and the vessel cross-sectional area. Reactive hyperemia was calculated as percent increase in

flow after cuff release as compared with baseline flow.Following a 10-minute recovery period,

additional frames were taken before and over a 4-minute period after sublingual administration of

400 µg GTN spray. Flow-mediated and GTN-induced dilation of the brachial artery were

expressed as maximum percent dilatation following cuff deflation and GTN administration,

respectively. All scans were taken in a “blind” fashion, with the sonographer being unaware of

the patients` inflammatory and HLA characteristics.

Page 6 of 21

Copyright Information

7

D. Carotid artery ultrasound protocol

The ultrasound system described above was used. The imaging protocol was described in details

previously (16). In short, longitudinal scans in bi-dimensional mode of the 1-cm-long distal end

of the left common carotid artery were imaged so that the lumen-intima and intima-media

interfaces were clearly distinguishable. All scans corresponded to the R-wave on ECG. Four to

six scans from each individual were recorded on videotape for offline analysis of the carotid

artery compliance (CAC), stiffness index (SI), and intima-media thickness (IMT). The mean

carotid IMT of 4 measurements along a 1-cm segment was calculated from each scan. Mean IMT

obtained from all scans from the same subject were averaged, and the resulted mean IMT was

used for statistical analyses.

Statistics

ANOVA was used to assess the differences between the DQ2/8 and non-DQ2/8 groups in age,

BMI, diabetes duration, HbA1C, lipid, inflammatory, and vascular indexes (BA-FMD and CA-

IMT). ANCOVA was used to control the associations of HLA-DQ2/8 genotype with the lipid and

vascular indexes for the possible confounding effects of age, diabetes duration, BMI, and HbA1C.

Simple and multiple regression was used for determining the association between inflammatory,

lipid and vascular indexes. CRP was log-transformed given its skewed distribution. Statistical

significance was set at p<0.05. Data are given as mean±standard errors unless otherwise

specified. StatView for Windows (SAS Institute Inc.®) was employed as statistical software.

RESULTS

Thirty-four patients (39 %, 18 male and 16 female) were positive for the HLA-DQ2/8 genotype.

Among the remaining 52 patients (31 male and 21 female), 15 had one of the two haplotypes

(HLA-DQ2 or DQ8). No differences in age, BMI, diabetes duration, and HbA1C were noted

Page 7 of 21

Copyright Information

8

between the DQ2/8 and non-DQ2/8 groups (p>0.4, Figure 1). Blood glucose was similar in both

groups (8.8 mmol/L in non-DQ2/8 group vs. 8.4 mmol/L in DQ2/8 group, p=0.7).

Lipid and inflammatory profiles

No significant differences in CRP (p>0.4), fibrinogen (p>0.8), and orosomucoid (p>0.7) were

observed between the DQ2/8 and non-DQ2/8 groups.

Lipid data were available in 83 patients. Two patients (one from each group) had HDL

cholesterol below the acceptable level (0.9 mmol/l). LDL cholesterol (acceptable upper limit: 3.3

mmol/l) was abnormally elevated in 5 patients (3 in the non-DQ2/8 group, and 2 in the DQ2/8

group).

When the groups were compared, LDL cholesterol was higher in the DQ2/8 group

(2.5±0.1 mmol/l) than in the non-DQ2/8 group ( 2.1±0.2 mmol/l, p<0.05), whereas HDL

cholesterol did not significantly differ between the DQ2/8 group (1.5±0.07 mmol/l) and the non-

DQ2/8 group (1.7±0.05 mmol/l, p=0.1). Total cholesterol showed a trend toward increased levels

in the DQ2/8 group compared to the non-DQ2/8 group (4.4±0.2 vs 4.1±0.1 mmol/l, p=0.08).

LDL-to-HDL ratio, an important atherogenic lipid index, was significantly higher in

the DQ2/8 group (1.8±0.2) than in the non-DQ2/8 group (1.3±0.2, p<0.01; Figure 2/Panel A).

The difference remained significant after adjustment for age, BMI, diabetes duration, HbA1C, and

CRP (p=0.02 by ANCOVA).

Overall, LDL-to-HDL ratio showed a modest correlation with CRP (r=0.4,

p<0.001) and BMI (r=0.4, p<0.01). When the patients were grouped based on the genotype, the

correlation between CRP and LDL-to-HLD ratio became significant among the DQ2/8 patients

(r=0.6, p<0.001;Figure 2/Panel B). In this group, the association between LDL-to-HDL ratio and

CRP remained significant after adjustment for age, BMI, diabetes duration and HbA1C (p<0.01).

Page 8 of 21

Copyright Information

9

Flow-mediated and GTN-induced dilations of the brachial artery, and intima-media

thickness of the carotid artery

There groups were similar in baseline brachial artery diameter, reactive hyperemia, GTN-induced

dilation and carotid IMT (Table 1).

No significant difference was noted between the groups with regard to BA-FMD

(5.3±0.7 in the DQ2/8 group vs 6.7±0.5% in the non-DQ2/8 group, p=0.08; Figure 3/Panel A).

However, in the DQ2/8 group, FMD inversely correlated with CRP (r=0.5, p=0.01), but no such

association was observed in the non-DQ2/8 group (r=0.1, p=0.8; Figure 3/Panel B). FMD did not

correlate with LDL-to-HDL ratio in either groups (p>0.5).

.

Low-grade inflammation, lipid, and vascular indexes in “DQ2/8+” patients

When the patients were further categorized according to their CRP levels (cut-off value 1 mg/l),

both LDL-to-HDL ratio (Figure 4/Panel A) and FMD (Figure 4/Panel B) showed a shift toward a

more atherogenic profile in the “DQ2/8”&”CRP>1” subgroup (2.2±0.2 and 3±1.1%,

respectively) as compared to the remaining subgroups. The GTN-induced dilatory responses of

the brachial artery and carotid IMT were similar in all subgroups (p>0.3).

DISCUSSION

To the best of our knowledge, this study is the first to suggest that type 1 diabetes children and

adolescents with genotype conferring the highest risk for this disease could be more susceptible

to atherogenic endothelial and lipid phenotypes. Our findings also suggest a possible relationship

between this genetic susceptibility and systemic inflammation.

It is nowadays agreed that type 1 diabetes increases the risk of cardiovascular

disease through inflammatory, oxidative, and glucose-related events that lead to vascular

endothelial damage (11), which is a key-mechanism in atherosclerosis through all of its stages

(4). Mounting evidence points out that a close-to-normal glycemic control in type 1 diabetes

Page 9 of 21

Copyright Information

10

patients is not sufficient to fully prevent the widespread vasculopathy (7), supporting thus the

view that other intrinsic and extrinsic factors are important as well. Although the concept of

heritability of cardiovascular disease has gained increasing attention during the past decade (19),

no study has so far investigated whether genetic factors predisposing for type 1 diabetes could

interfere with atherogenic events in young patients with type 1 diabetes.

Approximately 60% of the familial genetic risk for type 1 diabetes is attributable to

the HLA region (4). HLA DQA1*0301-B1*0302 and DQA1*0501-B1*0201 haplotypes,

especially when both are present in the genotype, have been found to be most strongly associated

with the onset of type 1 diabetes, while other haplotypes, e.g., HLA-DQA1*0102- B1*0602,

confer an age-dependent negative association (27). The precise mechanisms whereby certain

HLA increase diabetes susceptibility are not yet thoroughly clarified, but there is a general

consensus that the HLA DQ molecules exert their effects in part via presentation of peptides from

islet antigens to T cells, which contribute to the destruction of insulin-producing cells (28).

Previous studies (26, 17) have shown that expression of HLA molecules on the endothelial cells

of islet microcirculation goes hand in hand with lymphocyte infiltration, thereby suggesting a

possible pathogenic role of HLA-mediated endothelial vasculopathy in the development of type 1

diabetes. Greening and colleagues (6) demonstrated that both pancreatic and aorta endothelial

cells expressing HLA class II molecules have the ability to process and present the islet

autoantigen GAD65. It is therefore conceivable to assume that endothelial cells of other

peripheral arterial beds migh also owe HLA-mediated antigen presenting capacity in type 1

diabetes (6). Particular HLA-DQ phenotypes, including DQ2/8, appear to facilitate the presence

of potentially pathogenic T cells in the peripheral circulation via thymic selection (10).

Lymphocyte accumulation within the arterial wall is an important mechanistic

component of atherosclerosis development (8), and contributes to endothelial injury (18).

Endothelial injury promotes in turn additional immune events, including release of chemokines

and cytokines, with further endothelial transmigration of immune cells and CRP synthesis via

Page 10 of 21

Copyright Information

11

liver activation by interleukin-6 (15). The increased inflammatory activity leads to alteration of

lipoprotein metabolism characterized in part by increase in LDL and decrease in HDL

cholesterol. Dyslipidemia prevails in type 1 diabetes (12), and has an important role in

endothelial injury and plaque development (24, 25). CRP is slightly, yet significantly elevated in

diabetic children (21). Nevertheless, a slight rise in CRP, ie, over 1 mg/l, appears to be predictive

of the relative risk of future cardiovascular events in apparently healthy adults (20). Järvisalo and

colleagues found that even less plasma concentrations of CRP (ie, 0.7 mg/l) were associated with

decreased endothelial vasodilatory function of the brachial artery (9). How exactly the putative

association of CRP with the endothelial and lipidemic disturbances is amplified in HLA DQ2/8

diabetic patients is difficult to speculate, being to our opinion a tantalizing task for future studies

addressing this topic.

In diabetic children and adolescents, measures of both functional and structural

changes precursive to atherosclerosis seem to be influenced by age and diabetes duration. It is

therefore possible that the observed associations with diabetes-risk HLA might be more

pronounced in older patients or in those with longer disease duration. Also, further studies are

needed to investigate whether a better diabetes control would translate into a lesser impact of the

genotype on atherogenic endothelial and lipid phenotypes. In our study, these associations

remained significant after adjustment for age, diabetes duration, and HbA1c, but a possible

additive effect of these factors on the reported links cannot be ruled out. Whether the genotyping

might be used to identify individuals for whom early lipid management may be warranted is

perhaps an additional important task for future trials given the impact of diabetes dyslipidemia on

cardiovascular risk.

Several limitations need to be considered. The size of the study group is rather

small although invitation to participate was submitted to all diabetic patients registered at our

hospital. Our intention is to recruit additional patients aged 6 to 18 y across the entire part of

Southernmost Sweden (Skåne county). Other HLA genotypes known to predispose to diabetes

Page 11 of 21

Copyright Information

12

will be studied as well. We also intend to include nondiabetic HLA-matched siblings in order to

assess whether the HLA predisposition to diabetes could influence per se the vascular status and

lipid profile in otherwise healthy children. Another limitation resides in the cross-sectional

nature of this study. Follow-up studies over a 2-year period are under way. Finally, even though

flow-mediated dilation of the brachial artery is generally accepted as a surrogate of early

atherosclerosis, it remains uncertain whether it could in time become a clinically useful measure

of cardiovascular risk. Therefore, the precise clinical significance of the observed endothelial

disturbances in relation to HLA-DQ2/8 genotype remains uncertain.

In conclusion, the present study suggests independent associations of HLA-DQ 2/8

genotype, which is known to confer the highest risk for type 1 diabetes, with atherogenic lipid

and vascular endothelial phenotypes in children and adolescents with type 1 diabetes. The

findings warrant large-scale, prospective studies to verify these findings and to assess whether

this and other diabetes-susceptible HLA genotypes could accelerate atherosclerosis already

during the preclinical phase of type 1 diabetes.

Acknowledgements

The study was supported by FAMRI, USA. Additional grants were received from Lund

University, the Swedish Research Council, and the National Institute of Health. We thank Annica

Maxedius, registered nurse, for the excellent technical assistance in ultrasound scanning and

blood sampling throughout the study.

REFERENCES

1. Barker JM. Clinical review: Type 1 diabetes-associated autoimmunity: natural history,

genetic associations, and screening. J Clin Endocrinol Metab 91: 1210-1217, 2006.

Page 12 of 21

Copyright Information

13

2. Creager MA, Lüscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease:

Pathophysiology, clinical consequences, and medical therapy. Circulation 108: 1527-32,

2003.

3. Daneman D. Type 1 diabetes. Lancet 367: 847-858, 2006.

4. Davignon J, Ganz P. Role of endothelial dysfunction in atherosclerosis. Circulation 109:

27-32, 2004.

5. Eckel RH, Wassef M, Chait A, Sobel B, Barrett E, King G, Lopes-Virella M, Reusch J,

Ruderman N, Steiner G, Vlassara H. Prevention Conference VI: Diabetes and

Cardiovascular Disease: Writing Group II: pathogenesis of atherosclerosis in diabetes.

Circulation 105: 138-143, 2002.

6. Greening JE, Tree TI, Kotowicz KT, van Halteren AG, Roep BO, Klein NJ, Peakman M.

Processing and presentation of the islet autoantigen GAD by vascular endothelial cells

promotes transmigration of autoreactive T-cells. Diabetes 52: 717-725, 2003.

7. Goldberg IJ, Dansky HM. Diabetic vascular disease: an experimental objective.

Arterioscler Thromb Vasc Biol 26: 1693-1701, 2006.

8. Hansson GK, Libby P, Schonbeck U, Yan ZQ. Innate and adaptive immunity in the

pathogenesis of atherosclerosis. Circ Res 91: 281-291, 2002.

9. Jarvisalo MJ, Harmoinen A, Hakanen M, Paakkunainen U, Viikari J, Hartiala J,

Lehtimäki T, Simell O, Raitakari OT. Elevated serum C-reactive protein levels and early

arterial changes in healthy children. Arterioscler Thromb Vasc Biol 22: 1323-1328, 2002

10. Jordan MS, Riley MP, von Boehmer H, Caton AJ. Anergy and suppression regulate

CD4(+) T cell responses to a self peptide. Eur J Immunol 30: 136-144, 2000.

11. Joshua IG, Zhang Q, Falcone JC, Bratcher AP, Rodriguez WE, Tyagi SC. Mechanisms of

endothelial dysfunction with development of type 1 diabetes mellitus: role of insulin and

C-peptide. J Cell Biochem 96: 1149-1156, 2005.

Page 13 of 21

Copyright Information

14

12. Kershnar AK, Daniels SR, Imperatore G, Palla SL, Petitti DB, Pettitt DJ, Marcovina S,

Dolan LM, Hamman RF, Liese AD, Pihoker C, Rodriguez BL. Lipid abnormalities are

prevalent in youth with type 1 and type 2 diabetes: the SEARCH for Diabetes in Youth

Study. J Pediatr 149: 314-319, 2006.

13. Lernmark A. Type 1 diabetes as a model for prediction and diagnosis. Autoimmunity 37:

341-345, 2004.

14. Lernmark B, Lynch K, Lernmark A. Cord blood islet autoantibodies are related to stress

in the mother during pregnancy. Ann N Y Acad Sci 1079: 345-349, 2006.

15. Libby P. Inflammation in atherosclerosis. Nature 420: 868-74, 2002.

16. Liuba P, Persson J, Luoma J, Ylä-Herttuala S, Pesonen E. Acute infections in children are

accompanied by oxidative modification of LDL and decrease of HDL cholesterol, and are

followed by thickening of carotid intima-media. Eur Heart J 24: 515-521, 2003.

17. Marelli-Berg FM, Frasca L, Weng L, Lombardi G, Lechler RI. Antigen recognition

influences transendothelial migration of CD4+ T-cells. J Immunol 162: 696-703, 1999.

18. Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated

signalling in immune, inflammatory and angiogenic processes: the cascade of events

leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 3: 35-42, 2004.

19. Nordlie MA, Wold LE, Kloner RA. Genetic contributors toward increased risk for

ischemic heart disease. J Mol Cell Cardiol 39: 667-679, 2005.

20. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO 3rd, Criqui M, Fadl

YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC Jr, Taubert K, Tracy RP,

Vinicor F; Centers for Disease Control and Prevention; American Heart Association.

Markers of inflammation and cardiovascular disease: application to clinical and public

health practice: A statement for healthcare professionals from the Centers for Disease

Control and Prevention and the American Heart Association. Circulation 107: 499-511,

2003.

Page 14 of 21

Copyright Information

15

21. Picardi A, Valorani MG, Vespasiani Gentilucci U, Manfrini S, Ciofini O, Cappa M,

Guglielmi C, Pozzilli P; IMDIAB Group. Raised C-reactive protein levels in patients with

recent onset type 1 diabetes. Diabetes Metab Res Rev 23: 211-214, 2007.

22. Raza J, Movahed A. Current concepts of cardiovascular disease in diabetes mellitus. Int J

Cardiol 89: 123-134, 2003.

23. Redondo MJ, Fain PR, Eisenbarth GS. Genetics of type 1 diabetes. Recent Prog Horm

Res 56: 69-89, 2001.

24. Renard CB, Kramer F, Johansson F, Lamharzi N, Tannock LR, von Herrath MG, Chait A,

Bornfeldt KE. Diabetes and diabetes-associated lipid abnormalities have distinct effects

on initiation and progression of atherosclerotic lesions. J Clin Invest 114: 659-668, 2004.

25. Ross R, Aqius L. The process of atherogenesis--cellular and molecular interaction: from

experimental animal models to humans. Diabetologia 35: 34-40, 1992.

26. Savage CO, Brooks CJ, Harcourt GC, Picard JK, King W, Sansom DM, Willcox N.

Human vascular endothelial cells process and present autoantigen to human T-cell lines.

Int Immunol 7: 471-479, 1995.

27. Todd JA. Genetic analysis of type 1 diabetes using whole genome approaches. Proc Natl

Acad Sci U S A 92: 8560-8565, 1995.

28. Tree TI, Duinkerken G, Willemen S, de Vries RR, Roep BO. HLA-DQ-regulated T-cell

responses to islet cell autoantigens insulin and GAD65. Diabetes 53: 1692-1699,

2004.Genetic analysis of type 1 diabetes using whole genome approaches. Proc Natl Acad

Sci U S A 92: 8560-8565, 1995.

Page 15 of 21

Copyright Information

16

TABLE

Baseline diameter, GTN-induced dilatation and reactive hyperemia of the brachial artery (BA),

intima-media thickness (IMT) of the carotid artery (CA), and diastolic (DBP) and systolic (SBP)

blood pressure values in patients with (“DQ2/8”) and without (“non-DQ2/8”) HLA-DQ2/8

genotype. All data are expressed as mean±SE.

non-DQ2/8 DQ2/8 p-value

DBP (mmHg) 71±2 68±1 0.2

SBP (mmHg) 126±2 123±3 0.4

Baseline diameter of BA (mm) 2.99±0.07 2.29±0.09 0.9

GTN-induced vasodilatation of BA (%) 11.4±0.6 12.2±0.8 0.4

Reactive hyperemia (%) 283±15 292±20 0.7

IMT (mm) of CA 0.40±0.01 0.39±0.01 0.8

Page 16 of 21

Copyright Information

17

FIGURE LEGEND

Figure 1: Box plot distribution of age (years), diabetes duration (years), body mass index (BMI,

kg/m2), and glycosylated hemoglobin levels (%) of diabetic patients with (“DQ2/8“n=34) and

without (DQ2/8, n=52) HLA-DQ2/8 genotype. The box plot displays the 25th percentile, median,

and 75th percentile, as well as the 10th and 90th percentiles as horizontal lines outside the box.

* denotes p value.

Figure 2: Panel A-Box plot illustrating the differences in LDL-to-HDL ratio between the DQ2/8

and non-DQ2/8 groups. The box plot displays the 25th percentile, median, and 75th percentile, as

well as the 10th and 90th percentiles as horizontal lines outside the box. Panel B- association of

LDL-to-HDL ratio with CRP in DQ2/8 and non-DQ2/8 patients. * denotes p value, r denotes

correlation coefficient, and n denotes number of patients.

Figure 3: Panel A- Box plot illustrating the differences in brachial artery`s flow mediated

dilation between the DQ2/8 and non-DQ2/8 groups. The box plot displays the 25th percentile,

median, and 75th percentile, as well as the 10th and 90th percentiles as horizontal lines outside the

box. Panel B- Association of brachial artery’s flow-mediated dilation with CRP in in DQ2/8 and

non-DQ2/8 patients. * denotes p value, r denotes correlation coefficient, and n denotes number of

patients.

Figure 4: Box plot distribution of LDL-to-HDL ratio (Panel A) and brachial artery`s flow

mediated dilation (Panel B) in relation to the HLA-DQ2/8 genotype and CRP (cut-point 1 mg/l).

The box plot displays the 25th percentile, median, and 75th percentile, as well as the 10th and 90th

percentiles as horizontal lines outside the box. * denotes p value, and n denotes number of

patients.

Page 17 of 21

Copyright Information

18

Page 18 of 21

Copyright Information

19

Page 19 of 21

Copyright Information

20

Page 20 of 21

Copyright Information

21

Page 21 of 21

Copyright Information