INTERLEUKIN-1 GENE CLUSTER POLYMORPHISMS AND ALCOHOLISM IN SPANISH MEN

Abstract

Aims: In an attempt to explain differences in susceptibility to alcoholism and alcohol liver disease (ALD), different genes have been analysed, among them those encoding inflammatory cytokines. Thus, it has been reported recently that both the interleukin 1 receptor antagonist (IL1RN) and the IL1β (IL1B) genes may influence the risk of ALD in Japanese alcoholics. We analysed the distribution of single nucleotide polymorphisms (SNPs) located in the IL1A, IL1B, IL1R1 and IL1RN genes in alcoholic and non-alcoholic Spanish subjects. Methods: DNA samples were obtained from 139 male alcoholics, 78 of whom were diagnosed as alcohol dependent (32 patients with liver cirrhosis and 46 without ALD) and 61 as alcohol abusers (25 with liver cirrhosis and 36 without ALD). As a control, we studied 81 age- and sex-matched healthy volunteers. Results: Alleles −511 IL1B*1 and IL1RN*1 were represented more in alcoholic patients than in the control group. We did not find any association of alcoholism or ALD with polymorphisms in the IL1A and IL1R1 genes. Conclusions: We conclude that the proteins encoded by the IL1RN and IL1B genes may be involved in susceptibility to alcoholism in Spanish men, probably through a different pathway from that involved in the regulation of the inflammatory response.

( Received 25 November 2004; first review notified 14 December 2004; in revised form 23 February 2005; accepted 8 March 2005; Advance Access publication 29 March 2005 )

INTRODUCTION

It is well established that only a minority of alcoholic patients have alcohol liver cirrhosis (ALC); furthermore, alcoholic patients may develop either alcohol dependence or alcohol abuse. In an attempt to explain these findings, genetic studies have been conducted. Previous reports have shown that polymorphic alleles of alcohol dehydrogenase ( Couzigou et al ., 1990 ; Sherman et al ., 1993 ; Borras et al ., 2000 ; Eriksson et al ., 2001 ), aldehyde dehydrogenase ( Poupon et al ., 1992 ; Iwahashi et al ., 1995 ) and cytochrome CYP2E1 ( Tsutsumi et al ., 1994 ; Maezawa et al ., 1995 ; Savolanien et al ., 1997; Grove et al ., 1998 ; Parsian et al ., 1998 ) are associated with alcoholism. Pro-inflammatory cytokines, such as interleukin 1 (IL-1) and tumour necrosis factor-α (TNF-α) emerged as proteins involved in the pathogenesis of alcohol liver disease (ALD) ( McClain et al ., 1993 ; Neuman et al ., 2001 ; Laso et al ., 2002 ), and polymorphisms in genes encoding these cytokines could change the susceptibility to develop ALD. Moreover, in Japanese alcoholics Takamatsu et al . (1998) have reported that a polymorphism in the IL1 receptor antagonist gene (IL1RN) may influence the risk of hepatic fibrosis. Additionally, a polymorphism in the promoter region and in exon 5 of the IL1B gene has been associated with the development of ALD ( Takamatsu et al ., 2000 ).

Recent work conducted by our group has shown that the presence of the IL1RN*1 allele is not associated with ALD but with greater risk of alcoholism in Spanish men ( Pastor et al ., 2000 ). Thus, we seek to investigate whether the association of IL1RN gene polymorphism and alcoholism in Spanish men can be extended to other members of the IL1 gene family located on chromosome 2q ( Steinkasserer et al ., 1992 ; Nicklin et al ., 2002 ), such as IL1A, which encodes IL-1α, ILB, which encodes IL-1β, and ILRI, which encodes the IL1 type I receptor.

SUBJECTS AND METHODS

Subjects

We studied 139 male alcoholic patients referred to the Alcoholism Unit of the University Hospital of Salamanca, Spain, aged 24 to 77 years (50 ± 13 years, mean ± SE) who had consumed >120 g of ethanol daily for at least 10 years. Seventy-eight patients had alcohol dependence (32 with ALC and 46 without ALD) and 61 patients had alcohol abuse (25 with ALC and 36 without ALD) according to the DSM-IV criteria for alcohol dependence or alcohol abuse ( American Psychiatric Association, 1994 ).

Histopathological examination of the livers of 47 cirrhotic patients in which biopsy was performed revealed micronodular cirrhosis. Liver biopsy was not performed in the remaining 10 patients owing to blood coagulation abnormalities. In these cases, the diagnosis of cirrhosis was established on the basis of the presence of the physical stigmata of chronic liver disease, a history of ascitis, variceal bleeding or hepatic encephalopathy, as well as on gastroscopic and/or ultrasonographic findings ( Erlinger and Benhamou, 1999 ). None of the alcoholic patients without ALD had the physical stigmata of chronic liver disease (i.e. cutaneous signs, hepatosplenomegaly, gynaecomastia, testicular atrophy). Haemoglobin concentration, prothrombin time and serum albumin level were normal in all subjects; ALT and AST serum levels were required to be less than twice the upper normal limits (40 IU/l). Liver biopsy was not performed for ethical reasons, liver disease being excluded on the basis of ultrasonographic studies. Hepatitis B surface antigen (HbsAg) and antibodies to hepatitis C virus were negative in all alcoholic patients.

Eighty-one age- and sex-matched healthy volunteers were included in the study. All of them consumed <10 g of ethanol/day and none had a history of alcohol abuse or alcohol dependence and neither did their first-degree relatives.

Patients and control subjects gave informed consent. The study was approved by the Ethical Committee of the University Hospital of Salamanca.

Genetic analysis

Genomic DNA was extracted from nucleated peripheral blood cells using standard proteinase K digestion, phenol– chloroform extraction and ethanol precipitation, and was stored at −20°C.

The C/T transition at position −889 in the promoter region of the IL1A gene was analysed after PCR amplification using the primers 5′-AAGCTTGTTCTACCACCTGAACT AGGC-3′ and 5′-TTACATATGAGCCTTCCAATG-3′. PCR conditions consisted of 30 cycles of denaturation at 95°C for 1 min, annealing at 55°C for 1 min, and extension at 70°C for 1 min followed by a final extension step at 70°C for 5 min. The PCR products were digested overnight at 37°C with the NcoI restriction enzyme (Fermentas UAB, Vilnius, Lithuania) and the fragments were separated on 4% agarose gels ( McDowell et al ., 1995 ).

The region containing the C/T transition at position −511 in the IL1B gene promoter was amplified by PCR using the primers 5′-TGGCATTGATCTGGTTCATC-3′ and 5′-GTTTA GGAATCTTCCCATTT-3′. PCR conditions were similar to those described for the amplification of the VNTR of the ILIRN gene ( Pastor et al ., 2000 ). The PCR products were digested overnight at 37°C with the AvaI restriction enzyme (Fermentas UAB, Vilnius, Lithuania) and the fragments were visualized on 2.5% agarose gels ( Di Giovine et al ., 1992 ).

The region containing the silent T/C transition at position +3953 within exon 5 of the IL1B gene was amplified using the primers 5′-GTTGTCATCAGACTTTGACC-3′ and 5′-TTCAGTTCATATGGACCAGA-3′. DNA samples were amplified under the same conditions as above. The PCR products were digested overnight at 65°C with the TaqI restriction enzyme (Fermentas UAB, Vilnius, Lithuania) and the fragments were visualized on 3% agarose gels ( Pociot et al ., 1992 ).

The region containing the G/A transition in the 5′UTR of the IL1R1 gene was amplified using the primers 5′-CAGGGATGACAGTCTCCACCTT-3′ and 5′-GTGTCCAAATGGCAGTTCTGAA-3′. Denaturing was performed at 95° for 30 s, annealing was performed at 60° for 45 s and extension at 72° for 90 s. The PCR products were digested overnight at 37°C with the HinfI restriction enzyme (Fermentas UAB, Vilnius, Lithuania) and the fragments were visualized on 2.5% agarose gels ( Bergholdt et al ., 2000 ).

Molecular analysis of the intron 2 VNTR of the IL1RN gene was performed as reported previously ( Pastor et al ., 2000 ).

Statistical analysis

The results are presented as genotypic, allelic and haplotypic frequencies. The different groups were compared using the chi-square and Fisher's exact probability tests, with Yates correction for small numbers. Continuous variables were compared using the non-parametric Mann–Whitney U - and Kruskal–Wallis tests, thereby avoiding any distributional assumptions. P < 0.05 was regarded as significant. The odds ratio and 95% confidence intervals (OR 95% CI) were also calculated. All analyses were performed using SPSS v.11.0.

RESULTS

IL1A NcoI polymorphism (position −889 in the promoter)

The distribution of −889 IL1A alleles and genotypes in the control subjects were similar to that reported previously in Caucasians ( Mansfield et al ., 1994 ; McDowell et al ., 1995 ). No significant differences were observed in genotype or allelic distribution between the patients and the healthy subjects or between alcoholics with ALC and those without ALD. Moreover, the genotype and allele frequencies of the −889 IL1A polymorphism were similar among the alcohol-dependent subjects and alcohol abusers ( Table 1 ).

IL1B AvaI polymorphism (position −511 in the promoter)

The frequency of −511 IL1B genotypes and alleles in the healthy subjects included in our study was similar to that reported previously in Caucasians ( Mansfield et al ., 1994 ; Santtila et al ., 1998 ). Both the frequency of genotype −511 IL1B*1/1 and the frequency of allele −511 IL1B*1 were significantly higher in the alcoholics than in the control group [ P = 0.017 (OR = 2.08) and P = 0.002 (OR = 1.95), respectively], but no significant differences were detected between the alcoholics with ALC and those without ALD. Moreover, both the genotype and allele frequencies of the −511 IL1B polymorphism were similar among the alcohol-dependent subjects and alcohol abusers ( Table 2 ).

IL1B Taq I polymorphism (position +3953 in exon 5)

The distribution of +3953 IL1B genotypes and alleles in the healthy subjects included in our study was similar to that reported previously in Caucasians ( Pociot et al ., 1992 ). As shown in Table 3 , the distribution of alleles and genotypes was similar in the alcoholics and the healthy controls. Nevertheless, allele +3953 IL1B*1 was significantly better represented in the alcohol-dependent subjects than in the alcohol abusers ( P = 0.006; OR = 1.83), and allele +3953 IL1B*2 was more prevalent in patients with alcohol abuse than in those with alcohol dependence ( P = 0.011; OR = 0.45). No significant differences were observed between alcoholics with ALC and without ALD.

IL1R1 HinfI polymorphism (in the 5′ UTR)

Analysis of the distribution of alleles and genotypes of a polymorphism located in the 5′ UTR of the IL1R1 gene revealed no differences either between the alcoholic patients and the healthy subjects or between the alcoholics with ALC and those without ALD. Moreover, the genotype and allele frequencies were similar among the alcohol-dependent subjects and alcohol abusers ( Table 4 ).

IL1RN-VNTR polymorphism

The genotypic and allelic distribution of the VNTR of the IL1RN gene in the control group was similar to that reported previously in Caucasians ( Mansfield et al ., 1994 ). As shown in Table 5 , the IL1RN*1 allele was over-represented in Spanish alcoholics compared with the healthy subjects ( P = 0.000; OR = 3.69). No significant differences in allelic frequencies were observed either between the alcoholics with liver cirrhosis and those without ALD, or between the alcohol-dependent subjects and the alcohol abusers.

IL 1 gene cluster haplotypes

Haplotypes IL1RN*1/−511 IL1B*1 and IL1RN*1/−511 IL1B*1/+3953 IL1B*1 were more prevalent in the alcoholics than in the control subjects [ P = 0.003 (OR = 3.08) and P = 0.042 (OR = 2.01), respectively]. Moreover, the IL1RN*1/+3953 IL1B*2 and ILRN*1/−511 IL1B*1/+3953 IL1B*2 haplotypes were more prevalent in patients with alcohol abuse than in those with alcohol dependence [ P = 0.002 (OR = 0.311) and P = 0.002 (OR = 0.312), respectively] ( Table 6 ). We failed to find any difference in the distribution of haplotypes that include the other genotypes.

DISCUSSION

In a previous work, we have shown that the IL1RN*1 allele is associated with alcoholism in Spanish subjects ( Pastor et al ., 2000 ). However, we were not able to rule out if this association was the consequence of a linkage disequilibrium with other gene or genes located in the vicinity of the IL1RN locus. Thus, in the present work, we have increased the number of subjects included in our series and analysed other genes that are clustered with IL1RN in chromosome 2q ( Steinkasserer et al ., 1992 ; Nicklin et al ., 2002 ).

As shown in Tables 1 and 4, our results indicate that alcoholism is not associated with changes in the allelic frequency of polymorphisms located in the promoter region of the IL1A gene (position −889) or in the promoter region of the IL1R1 gene. However, as shown in Table 2 , allele −511 IL1B*1 was more frequent in Spanish alcoholics than in healthy subjects, whereas alleles of the +3953 IL1B polymorphism had a similar distribution in both groups ( Table 3 ). Moreover, the IL1RN*1 allele was over-represented in the alcoholics, in keeping with previous data from our group ( Pastor et al ., 2000 ) ( Table 5 ). In addition, our results show that both the IL1RN*1/−511 ILB*1 and the IL1RN*1/511 IL1B*1/+3953 IL1B*1 haplotypes were more frequent in the alcoholics than in the control subjects ( Table 6 ).

The lack of association of alcoholism with polymorphisms in both the IL1A and IL1R1 genes indicates that the relationship between IL1RN and IL1B gene variants and alcoholic disease cannot be attributed to linkage disequilibrium with other gene or genes located on chromosome 2 q. The functional significance of allele 1 of IL1RN is unknown. Allele IL1RN*2 has been linked to an increased synthesis of IL1Ra and a decreased production of IL1α in stimulated monocytes ( Danis et al ., 1995 ). Moreover, healthy carriers of the IL1RN*2 allele have significantly higher plasma levels of IL1-Ra than non-carriers ( Hurme and Santtila, 1998 ). The fact that we found a decrease in the frequency of allele 2 in Spanish alcoholics with and without liver disease suggests that the IL1Ra protein would not be involved in the development of ALD in Spanish alcoholics.

Interestingly, when we analysed the groups of alcoholics separately, we found that allele +3953 IL1B2 ( Table 3 ) and both haplotypes IL1RN*1/+3953 IL1B*2 and ILRN*1/−511 IL1B*1/+3953 IL1B*2 ( Table 6 ) were more frequent in the alcohol abusers than in the alcohol-dependent subjects, suggesting that the carriers of allele +3953 IL1B*2 may be less susceptible to developing alcohol dependence. Furthermore, the presence of allele +3953 IL1B*2 has been associated with an increase in IL1β secretion in vitro after lipopolysaccharide-induced activation of peripheral blood cells obtained from healthy subjects ( Pociot et al ., 1992 ), although in another report the authors failed to find such a relationship ( Stokkers et al ., 1998 ). No differences were detected between the groups included in our study with respect to the frequency of haplotype ILRN*2/+3953 IL1B*1, which is associated with a higher production of IL1β in immune cells ( Hurme and Santtila, 1998 ; Santtila et al ., 1998 ). In contrast to the findings reported for Japanese alcoholic patients ( Takamatsu et al ., 2000 ), we did not find any association between haplotype −511IL1B*2/+3953 IL1B*1 and ALC.

Our results show a genetic association between alcoholism and genes that encode proteins involved in the immune response. The relationship between cytokines and the central nervous system (CNS) is well known ( Reichlin, 1993 ; Hopkins and Rothwell, 1995 ; Rothwell and Hopkins, 1995 ; Allan and Rotwell, 2001 ), and there is also evidence that several members of the cytokine family are present in the brain. Thus, IL-1 activity in the brain has been related to the differentiation of dopaminergic neurons ( Ling et al ., 1998 ) and the modulation of central monoaminergic activity ( MohanKumar et al ., 1998 ). Similarly, IL-1, IL-2 and IL-6 can modify the release of neurotransmitters from the nucleus accumbens and other brain regions of the CNS that play a central role in alcohol abuse and alcohol dependence ( Koob et al ., 1994 ; Song et al ., 1999 ). It has also been reported that allele −511 IL1B*2 may be involved in the development of the structural changes seen in the CNS of subjects with schizophrenia ( Katila et al ., 1999 ), and that some alleles of the IL1RN gene have been associated with attention deficit hyperactivity disorder ( Segman et al ., 2002 ). Our results confirm that the IL1RN gene may be involved in susceptibility to alcoholism in Spanish subjects, and show for the first time that IL1B gene may also play a role in determining individual susceptibility to alcoholism. The mechanisms that define this relationship are yet to be determined.

The authors are grateful to Mrs Nieves Mateos, Mercedes García Mingo and Pilar Rodríguez for their technical help. This work was supported by Grants from the Junta de Castilla y León SA016/99, SA090/01, SA015/03 and MCYT PM00-70.

REFERENCES

Author notes

1Unidad de Medicina Molecular, Departamento de Medicina

2Unidad de Alcoholismo. Servicio de Medicina Interna, Hospital Universitario de Salamanca

3Instituto de Neurociencias de Castilla y León (INCYL), Universidad de Salamanca, Salamanca, Spain