www.elsevier.com/locate/diabres

Diabetes Research and Clinical Practice 71 (2006) 299–305

Interleukin-6–174 G > C polymorphism affects the

association between IL-6 plasma levels and insulin

resistance in type 2 diabetic patients

R. Testa a,*, F. Olivieri b, A.R. Bonfigli a, C. Sirolla a, M. Boemi a,F. Marchegiani b, M. Marra a, S. Cenerelli a, R. Antonicelli c,

A. Dolci d, G. Paolisso e, C. Franceschi b,f,g

aDiabetology Unit, INRCA, Via Della Montagnola, 81, I-60131, Ancona, ItalybCenter of Molecular Biology and Genetics, INRCA, Ancona, Italy

cCardiology Unit, INRCA, Ancona, ItalydMultimedica Laboratories, Milan, Italy

eDepartment of Geriatrics and Metabolic Diseases, Second University of Naples, ItalyfDepartment of Experimental Pathology, Bologna University, Bologna, Italy

g Interdepartmental Center ‘‘Galvani’’ (CIG), Bologna University, Bologna, Italy

Received 10 March 2005; received in revised form 15 July 2005; accepted 18 July 2005

Available online 2 September 2005

Abstract

Interleukin-6 (IL-6), a powerful inflammatory mediator, plays a pivotal role in the pathogenesis of insulin resistance and

type 2 diabetes. Recently, the IL-6 promoter polymorphism, at position �174 (G > C), has been associated to insulin sensitivity

althoughcontrastingdatahavebeenreported.Theaimofthisstudywastoevaluate theeffectof theIL-6–174G > Cpolymorphismon

insulinresistance. In238type2diabeticpatientswithoutdiabeticcomplicationsandin255controlsubjects,ageandgender-matched,

we evaluated the IL-6 –174 G > C genotype, the IL-6 plasma levels and the insulin resistance by the homeostasis model

assessment (HOMA). The levels of IL-6 and HOMA were not genotype-dependent and were higher in diabetic patients

( p < 0.01).Controlsubjects,bothC+(CG + CCgenotypes)andC� (GGgenotype)carriers, showedIL-6plasmalevelssignificantly

related to BMI, fasting insulin and HOMA. The same relationships were found in C+ diabetic carriers. Differently, diabetic

C� carriers did not show any relationship between IL-6 levels and all the evaluated variables. Interestingly, all the correlations

were dependent on BMI. These findings highlight that IL-6–174G > C polymorphism affects insulin resistance in type 2 diabetes,

where C+ carriers have an insulin resistance ‘‘IL-6-sensitive’’, while C� carriers do not. The identification of two categories of

diabetic patients may, therefore, lead to different therapeutic strategies in the management of insulin resistance.# 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Interleukin-6; –174 G > C interleukin-6 polymorphism; Insulin resistance; HOMA; Type 2 diabetes

* Corresponding author. Tel.: +39 071 8003854; fax: +39 071 8003556.

E-mail address: r.testa@inrca.it (R. Testa).

0168-8227/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.diabres.2005.07.007

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305300

1. Introduction

Chronic sub clinical inflammation is closely

involved in the pathogenesis of type 2 diabetes [1–

3]. Interleukins are regulatory proteins able to

accelerate or inhibit these inflammatory processes.

Circulating levels of interleukin-6 (IL-6) are increased

in insulin-resistant states, such as obesity, impaired

glucose tolerance and type 2 diabetes [4]. IL-6 gene

harbours a G to C polymorphism at its 50-upstream.

This sequence variation affects the rate of IL-6 gene

transcription [5] and IL-6 production [6,7]. Recent

data have focused the interest of this polymorphism on

the pathogenesis of the insulin-resistant state, demon-

strating that it is involved in the risk of developing

insulin resistance and type 2 diabetes mellitus [8,9].

Contrasting data were reported regarding the relation-

ship between this polymorphism and insulin sensi-

tivity. Fernandez-Real et al. demonstrated that

subjects homozygous for the C allele showed

significantly lower post-load glucose levels, lower

glycosylated haemoglobin, lower fasting insulin levels

and an increased insulin sensitivity index than carriers

of the G allele [10]. On the contrary, Kubaszek et al.

found lower energy expenditure and lower rates of

whole-body glucose uptake in subjects homozygous

for the C allele than subjects with GC or GG genotypes

[11]. Starting from these findings, the aim of this study

was to evaluate the effect of the IL-6�174 G > C

polymorphism on insulin resistance in healthy and

type 2 diabetic subjects.

Table 1

General characteristics of the studied subjects

Control subjects (n = 255)

Age (years) 60.2 � 9.5

BMI (kg/m2) 27.2 � 4.4

Sex (male %) 36.9

Hypertension (%) 19.6

Fasting glucose (mmol/l) 5.22 � 0.59

HbA1c (%) 5.40 � 0.67

Fasting insulin (mU/l) 4.98�/�1.92

HOMA 1.14�/�1.97

Total cholesterol (mmol/l) 5.72 � 0.97

HDL (mmol/l) 1.51 � 0.79

Triglycerides (mmol/l) 1.01�/�0.02

Fibrinogen (mg/dl) 277.3 � 66.7

CRP (mg/l) 0.95�/�3.46

IL-6 (pg/ml) 0.57�/�2.74

2. Materials and methods

2.1. Subjects

Two hundred and thirty-eight type 2 diabetic

patients (98 males and 140 females, mean

age � S.D., 61.2 � 9.7 years) and 255 healthy

controls (94 males and 161 females, mean

age � S.D., 60.2 � 9.5 years) matched for age and

sex were studied after informed consent was

obtained from each subject and the protocol of the

study was approved by the Ethical Committee of

INRCA Hospital. Clinical characteristics of the

participants are shown in Table 1. The diabetic

patients, diagnosed according to the American

Diabetes Association Criteria [12], presented

absence of micro and macrovascular complications,

assessed as follows: retinopathy by fundoscopy

through dilated pupils, nephropathy by urinary

albumin excretion rate (<30 mg/24 h) and by serum

creatinine (<1.3 mg/dl), neuropathy by somatosen-

sory threshold test (Semmes–Weinstein monofila-

ments 10 g) and by vibratory threshold test, ischemic

heart disease by clinical history and by resting

electrocardiogram, peripheral vascular disease by

clinical history and, for lower limbs, by ankle-

brachial index. The choice of a diabetic group

without complications was performed in order to

avoid that the onset of overt diabetic complications

may affect the studied associations. One hundred and

forty-five (60.9%) diabetic patients were treated by

Diabetic subjects (n = 238) p

61.2 � 9.7 0.31

28.4 � 4.7 0.01

41.2 0.32

51.0 <0.01

8.68 � 2.43 <0.01

7.48 � 1.50 <0.01

6.32�/�1.87 <0.01

2.36�/�2.02 <0.01

5.66 � 0.96 0.51

1.20 � 0.31 <0.01

1.29�/�0.02 <0.01

289.0 � 80.4 0.13

2.01�/�3.20 <0.01

0.82�/�3.54 <0.01

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305 301

Fig. 1. Plasma concentration of IL-6 in C+ (CG + CC genotypes,

closed triangles) and C� carriers (GG genotype, open circles) of the

IL-6�174 G > C polymorphism in healthy and diabetic subjects.

Each symbol represents one individual, the bar indicates the geo-

metric mean per group; *p < 0.05.

diet alone and 93 (39.1%) by metformin. People

were defined as hypertensive when they were under

active treatment or when their systolic blood

pressure was >140 mmHg and/or their diastolic

blood pressure was >90 mmHg, on at least three

different occasions. The selected individuals did not

show any other illness and did not take any other

drug. Smokers and individuals with alcohol abuse

history were excluded. All the selected patients

consumed a Mediterranean diet.

2.2. Laboratory assays

Overnight fasting venous blood samples were

collected from 8:30 to 9:00 a.m. in plain, EDTA and

citrate added tubes (Venoject, Terumo, Europe, NV).

The samples were either analyzed immediately or

stored at �808 C for no more than 10 days. Glucose,

total and HDL cholesterol, triglycerides, fibrinogen,

insulin, c-reactive protein (CRP) and HbA1c were

analyzed by routine laboratory methods.

2.3. Genotyping

DNA was extracted from peripheral blood mono-

nuclear cells (PBMC) using phenol/chloroform,

according to standard procedures. Amplification of

the �174 locus was performed as described in a

previous work [13]. PCR amplification was followed

by an overnight restriction digest of 15 ml PCR

product with Nla III enzyme, according to conditions

recommended by the supplier. The presence of a

cytosine (C allele) at nucleotide-174 was revealed by

the presence of the Nla lII cutting site. The two alleles

were revealed by electrophoretic analysis in 2%

agarose gels.

2.4. IL-6 plasma levels

Blood was collected for the IL-6 measurement in

pyrogen-free glass tubes with EDTA and centrifuged

at 600 � g for 10 min. The plasma was removed and

stored at �80 8C in pyrogen-free plastic tubes until

analysis. Plasma IL-6 levels were measured using

a commercially available dual antibody sandwich

enzyme-linked immunoassay (BioSource Cytoscreen

Human IL-6 UltraSensitive kit). This assay detects as

little as 0.1 pg/ml with no cross-detection of other

cytokines. The inter- and intra-assay percentage

coefficients of variation of the method were 3.6 and

5.3%, respectively (1.1 pg/ml).

2.5. Calculation and statistical analysis

Insulin resistance was calculated by the home-

ostasis model assessment (HOMA) model, which

assumes that normal weight and healthy subjects

aged < 35 years have a 100% b-cell function and an

insulin resistance of unity. HOMA was expressed as a

product of insulin and glucose levels divided by 22.5

[14]. Data were analyzed with SPSS/Win program

(Version 10.0; Spss Inc., Chicago, IL). The skewed

distributions were log-transformed before statistical

analyses to achieve a normal distribution. Their results

are shown as geometric means �/� antilog S.D.

Differences between control and diabetic patients

were compared in univariate analyses using Student’s

t-test for continuous variables and x2 test for

categorical variables. The distributions of IL-6�174

G > C genotypes, allele frequencies and the Hardy–

Weinberg equilibrium were analyzed by x2 test. The

two-way analysis of variance with contrast analysis

was performed to investigate the interaction effect

between IL-6 genotype and diabetes. Pearson correla-

tion coefficients were calculated to analyze the

relationship between IL-6 and the other variables

stratified by groups and IL-6 genotypes. Partial

correlations were performed to adjust the simple

correlations for the effect of BMI. Probability value

less than 0.05 was considered statistically significant.

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305302

3. Results

Clinical characteristics of the two study groups are

reported in Table 1. As expected, diabetic patients

showed a higher percentage of hypertension, higher

BMI and higher levels of fasting glucose, HbA1c,

fasting insulin, HOMA, triglycerides, HDL, CRP and

IL-6 plasma levels. No differences in total cholesterol

and fibrinogen levels were detected between healthy

and diabetic subjects. The frequencies of the different

genotypes at C-174 G were as follows: 8% CC, 44%

GC and 48% GG genotypes in the control group and

10% CC, 40% GC and 50% GG genotypes in the

diabetic group. The genotype distributions were not

different between the control and the diabetic groups

(x2 = 1.21, d.f. = 2, p = 0.55). The genotype frequen-

cies were in Hardy–Weinberg equilibrium both in the

control and in the diabetic group. For the successive

Fig. 2. BMI, fasting glucose, HbA1c and HDL-C in healthy and diabetic su

carriers (GG genotype, open circles) of the IL-6�174 G > C polymorphism

statistical analyses, genotypes at �174 G > C locus

were grouped as ‘‘C+’’ (CC and CG) and ‘‘C�’’ (GG)

carriers. Fig. 1 shows the IL-6 plasma levels in

controls and diabetics, C+ and C� carriers. No IL-6

genotype–diabetes interaction effect was detected for

IL-6 levels (F = 0.14, d.f. = 1, p = 0.71), confirming

that IL-6 plasma levels increase only due to the effect

of diabetes (F = 10.22, d.f. = 1, p < 0.01). Figs. 2–4

show the means and the variability of the other

variables reported in Table 1, stratified by group and

genotype. For all the variables, no IL-6 genotype–

diabetes interaction effect was detected. Table 2 shows

Pearson’s and partial correlations between IL-6

plasma levels and the evaluated parameters in both

the control and diabetic groups stratified by C+ and

C� carriers. Significant correlations were found in

the control group between IL-6 levels and BMI,

fasting insulin, HOMA and CRP both in C+ and in

bjects divided in C+ (CG + CC genotypes, closed triangles) and C�. Columns are arithmetic means, the error bars are S.D.; *p < 0.05.

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305 303

Fig. 3. Fasting insulin and HOMA in healthy and diabetic subjects

divided in C+ (CG + CC genotypes, closed triangles) and C�carriers (GG genotype, open circles) of the IL-6�174 G > C poly-

morphism. Each symbol represents one individual, the bar indicates

the geometric mean per group; *p < 0.05.

Fig. 4. Triglycerides and CRP in healthy and diabetic subjects

divided in C+ (CG + CC genotypes, closed triangles) and C�carriers (GG genotype, open circles) of the IL-6�174 G > C poly-

morphism. Each symbol represents one individual, the bar indicates

the geometric mean per group; *p < 0.05.

Table 2

Pearson’s and partial correlations adjusted for BMI between IL-6 plasma levels and the evaluated variables in healthy and diabetic C+ (CG + CC

genotypes) and C-carriers (GG genotype)

Control subjects Diabetic subjects

C+ (n = 133) C� (n = 122) C+ (n = 120) C� (n = 118)

r Partial r r Partial r r Partial r r Partial r

Age �0.16 0.04 0.18 0.07

BMI 0.24* 0.30* 0.28** 0.10

Fasting glucose �0.06 0.17 0.02 0.06

HbA1c �0.16 �0.16 0.07 0.05

Fasting insulin 0.30** 0.22 0.37** 0.19 0.28* 0.16 0.18

HOMA 0.26* 0.17 0.36** 0.22 0.27* 0.14 0.18

Total cholesterol �0.09 �0.05 0.04 0.03

HDL 0.12 �0.19 �0.11 �0.16

Triglycerides 0.08 0.11 0.21 0.17

Fibrinogen 0.22 0.12 0.14 0.15

CRP 0.24* 0.18 0.25* 0.02 0.17 0.19

* p < 0.05.** p < 0.01.

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305304

C� carriers. In the diabetic group, only C+ subjects

showed a significant relationship between IL-6 plasma

levels and BMI, fasting insulin and HOMA, whereas

no significant correlation were found in the C�diabetic subjects. To evaluate the effect of BMI on the

above mentioned significant relationships, the partial

correlations were performed controlling for BMI. As

shown in Table 2, all the correlations disappeared.

4. Discussion

The main findings of our study are: (i) the effect of

the IL-6�174 G > C polymorphism on insulin

resistance appears when IL-6 levels are related to

insulin resistance in diabetic subjects; (ii) the C+

diabetic carriers show a relation between IL-6 levels

and insulin resistance, whereas the C� diabetic

carriers do not; (iii) BMI seems to sustain all the

relationships between IL-6 and insulin resistance. The

association between IL-6 genotype and insulin

resistance is highlighted by the data reported in

Table 2. This table clearly evidences that the

significant relationships between fasting insulin,

HOMA and IL-6 in the control subjects are not

influenced by IL-6 genotype. On the contrary, this

genotype becomes a determinant factor in the diabetic

group, where the same relationships are present only

in the C+ carriers. The IL-6�174 G > C polymorph-

ism could be, therefore, considered a genetic marker

able to discriminate two diabetic populations with

different sensitivity toward IL-6. In fact, the presence

of C allele defines a group of diabetic subjects in

which insulin action is ‘‘IL-6-sensitive’’, as in the

control group, whereas the C� diabetic carriers appear

to be ‘‘IL-6-insensitive’’. This ‘‘insensitivity’’ to the

effect of IL-6 levels in C� diabetic carriers suggests

that in these subjects insulin resistance depends on

other factors, probably not linked to inflammation.

One possible explanation may be found in the fact that

type 2 diabetes is a condition of metabolic impairment

with altered insulin sensitivity and inflammation. In

this stressed condition, the C/G polymorphism could

come into play, causing the different behaviour

between C+ and C� diabetic carriers. The underlying

mechanisms are not known. However, it was

previously reported that C+ carriers have a tight IL-

6 gene expression regulation as the presence of

cytosine in the position-174 creates a possible

inhibitor-transcription-factor-binding-site [5]. This

report leads us to hypothesize that this genetic

predisposition allows the C+ carriers to maintain

the relationship between IL-6 and insulin resistance

also in presence of diabetes. The disappearance of any

significant relationships between insulin resistance

and IL-6 plasma levels, found in control and C+

diabetic subjects after controlling for BMI, suggests

that adipose tissue plays an important role in the

balancing of inflammatory process. Our results are in

accordance with data recently reported in which it has

been demonstrated that fat mass and not insulin

responsiveness is the main factor in determining IL-6

levels [15]. On the whole, our study suggests that IL-6-

174 C+/C� genetic variant could constitute a feasible

genetic marker to identify two categories of diabetic

patients in which this genetic information could be

useful in the development of therapeutic strategies to

manage insulin resistance through pharmacogenomi-

cal and dietetic approaches.

References

[1] J.M. Fernandez-Real, W. Ricart, Insulin resistance and chronic

cardiovascular inflammatory syndrome, Endocr. Rev. 24

(2003) 278–301.

[2] A Festa, R. D’Agostino Jr., L. Howard, L. Mykkanen, R.P.

Tracy, S.M. Haffner, Chronic subclinical inflammation

as part of the insulin resistance syndrome: the insulin resis-

tance atherosclerosis study (IRAS), Circulation 102 (2000)

42–47.

[3] J.C. Pickup, M.B. Mattock, G.D. Chusney, D. Burt, NIDDM as

a disease of the innate immune system: association of acute-

phase reactants and interleukin-6 with metabolic syndrome X,

Diabetologia 40 (1997) 1286–1292.

[4] J.W. Stephens, S.J. Hurel, J.A. Cooper, J. Acharya, G.J. Miller,

S.E. Humphries, A common functional variant in the inter-

leukin-6 gene is associated with increased body mass index in

subjects with type 2 diabetes mellitus, Mol. Genet. Metab. 82

(2) (2004) 180–186.

[5] D. Fishman, G. Faulds, R. Jeffery, V. Mohamed-Ali, J.S.

Yudkin, S. Humphries, et al. The effect of novel polymorph-

isms in the interleukin-6 (IL-6) gene on IL-6 transcription and

plasma IL-6 levels, and an association with systemic-onset

juvenile chronic arthritis, J. Clin. Invest. 102 (1998) 1369–

1376.

[6] M. Bonafe, F. Olivieri, L. Cavallone, S. Giovagnetti, F.

Marchegiani, M. Cardelli, et al. A gender-dependent genetic

predisposition to produce high levels of IL-6 is detrimental for

longevity, Eur. J. Immunol. 31 (2001) 2357–2361.

R. Testa et al. / Diabetes Research and Clinical Practice 71 (2006) 299–305 305

[7] F. Olivieri, M. Bonafe, S. Giovagnetti, R. Stecconi, M. Car-

delli, L. Cavallone, et al. In vitro IL-6 production by EBV-

immortalized B lymphocytes from young and elderly people

genotyped for �174 C/G polymorphism in IL-6 gene: a model

to study the genetic basis of inflamm-aging, Mech. Ageing

Dev. 124 (2003) 549–553.

[8] B. Vozarova, J.M. Fernandez-Real, W.C. Knowler, L. Gallart,

R.L. Hanson, J.D. Gruber, et al. The interleukin-6 (�174) G/C

promoter polymorphism is associated with type-2 diabetes

mellitus in Native Americans and Caucasians, Hum. Genet.

112 (2003) 409–413.

[9] A. Kubaszek, J. Pihlajamaki, V. Komarovski, V. Lindi, J.

Lindstrom, J. Eriksson, et al. Finnish Diabetes Prevention

Study. Promoter polymorphisms of the TNF-alpha (G-308A)

and IL-6 (C-174G) genes predict the conversion from impaired

glucose tolerance to type 2 diabetes: the Finnish Diabetes

Prevention Study, Diabetes 52 (2003) 1872–1876.

[10] J.M. Fernandez-Real, M. Broch, J. Vendrell, C. Gutierrez, R.

Casamitjana, M. Pugeat, et al. Interleukin-6 gene polymorph-

ism and insulin sensitivity, Diabetes 49 (2000) 517–520.

[11] A. Kubaszek, J. Pihlajamaki, K. Punnonen, P. Karhapaa, I.

Vauhkonen, M. Laakso, The C-174G promoter polymorphism

of the IL-6 gene affects energy expenditure and insulin sensi-

tivity, Diabetes 52 (2003) 558–561.

[12] Report of the expert committee on the diagnosis and classi-

fication of diabetes mellitus, Diab. Care 20 (1997) 1183–

1197.

[13] O. Olomolaiye, N.A. Wood, J.L. Bidwell, A novel Nla III

polymorphism in the human IL-6 promoter, Eur. J. Immuno-

genet. 25 (1998) 267.

[14] K. Kain, A.J. Catto, P.J. Grant, Clustering of thrombotic

factors with insulin resistance in south Asian patients

with ischaemic stroke, Thromb. Haemost. 88 (2002) 950–

953.

[15] A.L. Carey, C.R. Bruce, M. Sacchetti, M.J. Anderson, D.B.

Olsen, B. Saltin, et al. Interleukin-6 and tumor necrosis

factor-a are not increased in patients with type 2 diabetes:

evidence that plasma interleukin-6 is related to fat mass and

not insulin responsiveness, Diabetologia 47 (2004) 1029–

1037.