The Association of Interleukin-4Haplotypes With Chronic Periodontitisin a Czech PopulationLydie Izakovicova Holla,*† Antonin Fassmann,† Peter Augustin,† Tomas Halabala,† Vladimir Znojil,*and Jiri Vanek†

Background: Cytokine gene polymorphisms are known toinfluence the susceptibility and disease course of manychronic disorders. Recently, interleukin (IL)-4 gene polymor-phisms were associated with aggressive periodontitis. Theaim of this study was to test differences in the distribution ofthe IL-4 alleles, genotypes, and haplotypes between patientswith chronic periodontitis (CP) and healthy controls in a Czechpopulation.

Methods: The association study was conducted using anage- and smoking status–matched case-control design in pa-tients with CP(n = 194) and healthy controls (n = 158) using thepolymerase chain reaction–restriction fragment length poly-morphism methods for the -590C/T, -33C/T, and intron 3 var-iable number tandem repeat (VNTR) variants of the IL-4 gene.

Results: No significant differences between patients andcontrols were found in allele and genotype frequencies ofall three polymorphisms. Nevertheless, complex analysisrevealed significant differences in haplotype frequenciesbetween the groups (P = 0.005). The haplotype T(-590)/T(-33)/allele 2 VNTR (70 base pairs)2 of the IL-4 gene was sig-nificantly more frequent in patients with CP than in controls(17.0% versus 11.0%; odds ratio = 1.85; 95% confidence inter-val: 1.19 to 2.87).

Conclusion: The three polymorphisms in the IL-4 gene actin a cooperative fashion and suggest that the high-productionIL-4 haplotype was associated with an increased risk for CP inthe Czech population. J Periodontol 2008;79:1927-1933.

KEY WORDS

Gene; haplotype; interleukin; interleukin-4; periodontitis;polymorphism.

Periodontal diseases are probablyone of the most common chronicdiseases and the major cause of

tooth loss in adults. Although periodontalbacteria are the causative agents in peri-odontitis, subsequent progression anddisease severity are determined by thehost immune response.1 The local re-sponse to periodontopathogens and theirproducts includes the recruitment of poly-morphonuclear neutrophils and macro-phages and the subsequent release ofinflammatory mediators, matrix metallo-proteinases, and cytokines. There is acomplex network of pro- and anti-inflam-matory cytokines acting in the inflamedperiodontal tissues.2 Dichotomy in thepattern of cytokine production and,hence, the balance between T helper(Th) 1 and Th2 cytokines led to formula-tion of several hypotheses about theirrole in periodontitis.3-5 Seymour et al.4

suggested that non-susceptibility to peri-odontal breakdown may involve a pre-dominantly Th1-like response, resultingin T-cell activation, cell-mediated im-munity, and interferon-gamma (IFN-g)enhancement of innate immunity. Insusceptible individuals, disease progres-sion involves a predominantly Th2-likeresponse in which T cells produce the cy-tokines required for B-cell proliferation,leading to polyclonal B-cell activation.4

However, it isunknownhowthestableTh1cell–dominated6 condition of marginal gin-givitis is transformed into the progressive

* Department of Pathophysiology, Faculty of Medicine, Masaryk University, Brno, CzechRepublic.

† Clinic of Stomatology, Faculty of Medicine, Masaryk University.

doi: 10.1902/jop.2008.080035

J Periodontol • October 2008

1927

periodontal lesion in which Th2 activity, with a crucialrole for interleukin (IL)-4, is prominent.7,8

IL-4 is a pleiotropic cytokine that inhibits Th1 cellswhile stimulating a Th2-type of immune response. It isalso a potent downregulator of macrophage function3

and induces apoptosis in monocytes.9 IL-4 inhibits thepersistence of macrophages in periodontitis lesionsand downregulates CD14 receptor, one of the key re-ceptors for lipopolysaccharides.Thus, itwasproposedthat a lack of downregulation of monocytes by IL-4leads to tissue destruction in periodontitis.3

The human gene for IL-4 maps to the long arm (q31to 33) of chromosome 5, and it is located within a clus-ter of other Th2 cytokine genes.10 Several polymor-phisms in the IL-4 gene have been described; someof them are implicated in the regulation of IL-4 produc-tion. Data from in vitro and in vivo studies11,12 sug-gested that the -590 C/T*T allele, which is in tightlinkage disequilibrium with the -33 C/T*T allele, is as-sociated with increased IL-4 expression. Another im-portant variant is a variable number tandem repeat(VNTR) polymorphism located in the third intron.13

It was proposed that an increased responsiveness ofthe (70 base pair [bp])2 repeat allele to transcriptionalactivation might lead to overexpression of IL-4.14,15 Inaddition, combinations of the three variants men-tioned above form two major ‘‘high’’ and ‘‘low’’ IL-4producer haplotypes.14,16

With respect to the key role of interleukins in the in-flammatory response in periodontitis, the IL-4 gene isan obvious functional candidate for this disease. Thereare several studies17-23 of polymorphisms of the IL-4gene in chronic periodontitis (CP) and/or aggressiveperiodontitis with controversial findings. However, toour knowledge, no study of the potential associationbetween IL-4 haplotypes formed by the three variantsused by us (i.e., a single nucleotide polymorphism[SNP] in the IL-4 promoter region [C/T] at position-590 [-524 relative to the transcription start site],one SNP at -33 C/T [+ 33 relative to the transcriptionstart site], and VNTR in the third intron [70-bp motif intwo or three copies]) and the risk for periodontitis hasbeen published. This study was undertaken to com-pare the distributions of IL-4 alleles, genotypes, andhaplotypes in patients with CP and controls in a rela-tively homogenous middle European population.

MATERIALS AND METHODS

Study PopulationAll subjects were white and of Czech nationality. Phe-notype status was assigned without knowledge ofgenotypes by two independent investigators. Onehundred ninety-four Czech patients with CP (98men and 96 women; age range: 35 to 50 years; meanage: 44.2 – 7.2 years), referred to the PeriodontalClinic of Masaryk University and St. Anne Hospital,

Brno, Czech Republic, in 2000 to 2006, were in-cluded in the study. All subjects had ‡18 teeth andwere in good general health. Exclusion criteria werehistory of cardiovascular disorders (such as coro-nary artery diseases or hypertension), diabetes mel-litus, malignant diseases, immunodeficiency, andcurrent pregnancy or lactation. The control groupconsisted of 158 unrelated white subjects (75 menand 83 women; mean age 44.5 – 7.9 years), residingin the same geographic area as the included subjects,who did not have a clinical history of periodontal dis-ease and presented to the clinic for non-periodontaltreatment. Controls were selected randomly duringthe same period as patients and were matched forage, gender, and smoking status. All control subjectshad ‡20 teeth and were in good general health. Exclu-sion criteria were identical to the patients with CP.These subjects were screened using a World HealthOrganization probe, and the Community PeriodontalIndex of Treatment Needs (CPITN)24 was assessed;CPITN index was <3 in controls.

The diagnosis of CP was based on physical exam-ination, medical and dental history, probing depth,assessment of attachment loss, tooth mobility, andradiographs as described previously.25 All patientsfulfilled the diagnostic criteria defined by the Interna-tional Workshop for a Classification of PeriodontalDiseases and Conditions for Chronic Periodontitis.26

Patients were classified according to the severity oftheir periodontal disease (based on the amount ofclinical attachment loss [AL]) into one of two diseasecategories. The moderate (n = 35) classificationrequired clinical AL of 3 to 5 mm, and the severe(n = 139) classification required clinical AL ‡6 mm.The loss of alveolar bone was determined radio-graphically. We used the index of Muhlemann andMazor27 to evaluate decreases in alveolar bone level.To adjust for the effect of smoking history on peri-odontal disease, the subjects (patients and controls)were classified into the following groups: subjectswho never smoked (referred to as non-smokers) andsubjects who were former smokers for ‡5 pack-yearsor current smokers (referred to as smokers). Thepack-years were calculated by multiplying the num-ber of years of smoking by the average number ofcigarette packs smoked per day.28 The study was per-formed with the approval of the Committee for Ethicsof the Medical Faculty, Masaryk University, and writ-ten informed consent was obtained from all partici-pants, in line with the Helsinki Declaration, beforeinclusion in the study.

DNA ExtractionThe standard protein K digestion and phenol/chloro-form method was used to extract genomic DNA fromEDTA-preserved blood samples.29

IL-4 Haplotypes in Periodontitis Volume 79 • Number 10

1928

GenotypingThenucleotidepositions in this studyaregiven relativeto the translation start site. The GenBank accessionnumber was AF395008 for the reference genomic se-quence used for IL-4 alignments.

The IL-4 polymorphisms, -590 C/T and -33C/T,were genotyped by polymerase chain reaction (PCR)–restriction fragment length polymorphism analysisaccording to previously published methods.30 VNTR(IL-4 70-bp repeat) polymorphism in intron 3 of theIL-4 gene was detected by a modification of the PCRdescribed by Mout et al.13 The primers, PCR condi-tions, and restriction enzymes used are listed in Table1. All fragments were resolved on agarose gels. Con-trols and patients were not genotyped in separatedbatches, and the analysis was performed masked withrespect to case-control status.

Statistical AnalysisA computer package‡ was used to estimate the statis-tical power to detect significant differences betweenthe study groups.

Differences in allele frequencies were tested bythe Fisher exact test, and the x2 test was used for gen-otypefrequencies.P<0.05wasconsideredstatisticallysignificant. Pairwise linkage disequilibrium amongthree common polymorphisms was measured by theLewontin standardized disequilibrium coefficient|D9|31 in each group separately.

Haplotype analysis was performed using a pro-gram§ developedbyStephensetal.32,33 Theapproachunderlying the program is a Bayesian haplotype re-construction method using coalescent-based modelsto improve theaccuracyofhaplotypes forunrelated in-dividuals, although extensions to related individualsare possible. The algorithm uses a flexible model forthe decay of linkage disequilibrium with distanceand explicitly incorporates an assumption about re-

combination rate variation. The program uses Gibbssampling, a Markov-Chain Monte Carlo algorithm forthe estimation of the posterior distribution. Hence,the individual haplotypes can be estimated from thisposterior distribution by choosing the most likely hap-lotype reconstruction for each individual. Practically,the program was run with a single file including geno-types of all subjects; for c-option (two groups), 10,000permutations were performed. Global differences inhaplotype frequency profiles between the groups weredetermined by the permutation test, and P <0.05 wasconsidered statistically significant.

Contingency table analysis, odds ratios (ORs), 95%confidence intervals, and significance values were esti-mated with the use of a statistical program package.i

RESULTS

The power calculation performed for this study dem-onstrated that the sample size required to detect a sig-nificant association of CP with genetic polymorphismsselected with an a-value of 0.05, OR = 1.75, and 80%power was 148 for controls and 177 for cases. The es-timate of the percentage of controls with the risk factorwas 17%. This shows that our sample size was largeenough to detect the association with an acceptablelevel of significance. However, considering the fact thatthe number of subjects included in the study in bothgroups was close to the calculated values, non-signifi-cant differences between both groups for single nucleo-tide polymorphism (in a single locus analysis) could bedue to the lack of statistical power to detect minor differ-ences and, therefore, should be interpreted with caution.

Onehundredninety-fourpatientswithCPand158con-trol subjectswereenrolled in thestudy.Thedemographic

Table 1.

PCR–Restriction Fragment Length Polymorphism Conditions for Analysis of the IL-4Gene Polymorphisms

Polymorphism Primers (59-39)

PCR Product

(bp/enzyme)

Digested

Fragments/Alleles

-590C/T BsmFI C: 601 + 45 bpT: 646 bp

59-ACTAGGCCTCACCTGATACG-39 (sense)59-AGGTGTCGATTTGCAGTGAC-39 (antisense) 646

-33C/T BsmAI C: 646 bpT: 622 + 24 bp

VNTR, intron 3 59-TAGGCTGAAAGGGGGAAAGC-39 (sense)59-CTGTTCACCTCAACTGCTCC-39 (antisense)

– Allele 1: 254 bp (70 bp)3Allele 2: 184 bp (70 bp)2

– = no restriction enzyme was needed for a reaction. For this method, we only used PCR (without restriction fragment length polymorphism); we detected avariable number of tandem repeats.

‡ EpiInfo 6, Centers for Disease Control and Prevention, Atlanta, GA.§ PHASE v. 2.1, University of Washington, Seattle, WA.i Statistica v. 6.0, Statsoft, Tulsa, OK.

J Periodontol • October 2008 Holla, Fassmann, Augustin, Halabala, Znojil, Vanek

1929

data of these subjects are shown in Table 2. We ana-lyzed three common polymorphisms in the IL-4 gene:one in the 59 flanking region, one in exon 1, and oneVNTR in intron 3. All variants were in tight linkage dis-equilibrium in both groups (Table 3). The genotype fre-quencies of all studied variants are presented in Table4. All three polymorphisms examined in this studyfollowed Hardy-Weinberg equilibrium in the controlgroup. None of the polymorphisms was found to besignificantly associated with CP. In addition, whenthe patient population was divided into two groups ac-cording to the severity of CP, there were no differencesbetween the groups with regard to the frequencies ofalleles or genotypes of all three variants (data not pre-sented). The same analysis was done for smoking andnon-smoking subjects, which showed similar frequen-cies of IL-4 variants in all groups (data not presented).

Whenwe combinedmultiple SNPsites andanalyzedhaplotypes, we observed only two haplotypes withfrequency >5% because all three SNPs in the IL-4 geneexhibited variability in the degree of linkage disequilib-rium. The frequencies of the IL-4 haplotypes are sum-marized in Table 5. Although a comparison of thegenotype or allele distributions in healthy subjects topatients provided essentially negative results, the dis-

tribution of IL-4 haplotype frequencies between thegroupsshowedsignificantdifferences(P=0.005).Spe-cifically, the -590T/-33T/VNTR (70 bp)2 haplotypewas more common in CP than in the control group(17% versus 11%; OR = 1.85; 95% CI: 1.19 to 2.87),suggesting that this haplotype is significantly associ-ated with periodontitis.

DISCUSSION

Periodontal diseases are complex, common diseasescaused by combinations of environmental factorsand genes or, more precisely, the genetic variabilityin these loci. Our study investigated the role of theIL-4 gene in CP in the Czech population. Despite bio-logic plausibility, only a few studies have examinedthe influence of selected variants in the IL-4 gene inaggressive and/or CP. Michel et al.17 studied two(-590C/T and 70-bp repeat) IL-4 polymorphisms inGerman patients with aggressive periodontitis(AgP). The investigators found a significant associa-tion between these variants and disease susceptibility.Scarel-Caminaga et al.18 evaluated the same pro-moter SNP in Brazilian patients with CP, but they didnot determine any significant differences in the alleleor genotype frequencies in this group. Similarly, Kangetal.,19 Pontesetal.,21andKaraetal.23didnotfindanysignificant correlation between the risk for periodonti-tis and the -590C/T and 70-bp repeat variants inKorean,Brazilian,orTurkishpopulations,respectively.Recently, Gonzales et al.22 investigated the relation-ship between -590C/T and -33C/T gene polymor-phisms and the risk for AgP. They observed that-590T/T and -33T/T genotypes were significantlyassociated with AgP. However, the findings in thesestudies are contradictory, possibly reflecting a smallsample size, the study of different populations, andtesting polymorphisms in isolation, which can be mis-leading. Selection of phenotype is also important be-cause periodontitis is a heterogeneous entity.

Our study evaluated and compared the distributionof IL-4 haplotypes, formed by the three selected vari-ants, between patients with CP and healthy subjects.Testing of haplotypes overcomes some of the prob-lems encountered with using single polymorphismsin genetic association studies, particularly when inter-actionsbetweenpolymorphismsonthesamechromo-some are more important in determining the diseaserisk.34 In accordance with this hypothesis, we founddifferences in the frequency of IL-4 haplotypes be-tween the groups of subjects included in the presentstudy but no significant associations between suscep-tibility toCPand individual IL-4polymorphisms.There-fore, our finding supports the idea that the interactionof multiple genetic markers within a haplotype, ratherthan individual markers, is a major determinant of dis-ease susceptibility.35

Table 2.

Demographic Data for Patients With CPand Controls

Controls Patients With CP

Subjects (n) 158 194

Age (years; mean – SD) 44.5 – 7.9 44.2 – 7.2

Age range (years) 33 to 55 35 to 50

Gender (n; male/female) 75/83 98/96

Smokers (%) 31.3 31.3

Table 3.

Pairwise Linkage Disequilibrium for AllPossible Two-Way Comparisons AmongThree Polymorphisms in the IL-4 Genein Control and CP Groups

SNP1 -590 SNP2 -33 VNTR

SNP1 controlsCP

1.000.98

0.760.95

SNP2 controlsCP

0.760.95

Linkage disequilibrium was measured by Lewontin standardized disequi-librium coefficient jD9j.

IL-4 Haplotypes in Periodontitis Volume 79 • Number 10

1930

Analysis of cytokine polymorphisms is crucial forour understanding of the pathogenesis and clinical ex-pression of several multifactorial diseases. Gene poly-morphisms may cause significant changes in functionbyaltering the levelsoractivityof specificproteins.Re-cently, the levels of blood serum IL-4 in patients withCPwere found to besignificantlyhigher than inhealthycontrols.36 Unfortunately, we did not analyze plas-matic IL-4 levels in patients and controls and theirassociations with gene polymorphisms and/or peri-odontal status in our study. However, all three typedmarkers in thisgenecombine to formtwomajor ‘‘high’’[haplotype I - T/T/(70 bp motif)2] and ‘‘low’’ [haplo-type II - C/C/(70 bp motif)3] IL-4 producer haplotypes.Associations were found between haplotype I and pul-monary dysfunction37 or fatal or near-fatal asthma;38

this suggests that IL-4 might be a disease enhancer

in this clinically heterogeneouscondition. In contrast, in rheuma-toidarthritis, theTh1-typedisease,the presence of ‘‘high-producer’’alleles of haplotype I, was associ-ated with a less severe or a non-destructive clinical course.39,40

Periodontal lesions contain Tcells and macrophages, which canproduce many cytokines, someof them (e.g., IL-4) were shownto play a role in Th2-like responsewith B-cell activation, prolifera-tion, and differentiation.7 In addi-tion, after in vitro stimulation withPorphyromonas gingivalis andFusobacterium nucleatum, mostlymphocytesobtained frombiop-sies of periodontitis-affected tis-sues were of the Th2 phenotype,

i.e., they formed large amounts of IL-4and only small amounts of IL-2 andIFN-g.4 After in vitro stimulation of pe-ripheral blood mononuclear cells, a sig-nificantly higher production of IL-4 andlower production of IFN-g were found inpatients with AgP compared to theirhealthy siblings.41 In addition, inactiva-tion of IFN-g and IL-12 by the gingipainsfrom P. gingivalis at inflammatory sitescould also downregulate Th1 responses(associated with non-aggressive peri-odontal lesions) andpromote Th2 path-ways and polyclonal B-cell activation inadvanced periodontitis.42 Therefore,although IL-4 genetically determineddifferences in cytokine expressionlevels are likely to play a role in suscep-tibility to periodontitis, they can also in-

teract with several other factors (e.g., periodontalpathogens and other genes) that increase or decreasethe chance of disease onset; this can explain contra-dictory findings from different studies.

CONCLUSIONS

The association between IL-4 haplotypes and CPfound by us supports some of the previous findingsabout the role of IL-4 gene variants in periodontal dis-eases.17,22 CP is a common, multifactorial diseasewhose manifestation is influenced by the interactiveeffects of multiple gene variants and environmentalfactors. However, this factor probably is influentialonly in cooperation with others (considering the lowfrequency of the associated haplotype). Moreover,our results need to be verified in different populations

Table 4.

Genotype Frequencies of IL-4 Polymorphisms inBoth Groups

Locus Allele Genotype

Controls

(n [%])

Periodontitis

(n [%])

P

Value OR (95% CI)

SNP1 -590 C/T CC 118 (74.7) 132 (68.1) 0.085 1.0CT 36 (22.8) 54 (27.8) 1.34 (0.82 to 2.19)TT 4 (2.5) 8 (4.1) 1.79 (0.52 to 6.09)

SNP2 -33 C/T CC 118 (74.7) 133 (68.6) 0.099 1.0CT 36 (22.8) 53 (27.3) 1.31 (0.80 to 2.13)TT 4 (2.5) 8 (4.1) 1.77 (0.52 to 6.04)

VNTR intron 3 1/2 11 118 (74.7) 132 (68.1) 0.072 1.012 36 (22.8) 53 (27.3) 1.32 (0.81 to 2.15)22 4 (2.5) 9 (4.6) 2.01 (0.60 to 6.70)

Comparisons performed with the Fisher exact test for allele frequencies.Reference group (healthy controls) designated as OR = 1.0.

Table 5.

Estimated Frequencies and Standard Errors of theReconstructed IL-4 Haplotypes Characterized by theThree Polymorphisms Analyzed in Each Group

Haplotypes Controls Patients With CP

-590C/T -33C/T VNTR E (frequency) SE E (frequency) SE

C C 1 (70 bp)3 0.831919 0.001111 0.805308 0.001589

T T 2 (70 bp)2 0.110400 0.001111 0.169768 0.000868

Five others* 0.057682 0.002222 0.024923 0.004993

†P = 0.005

OR = 1.85; 95% CI = 1.19 to 2.87

Bold denotes significant findings.* Five pooled remaining haplotypes with very low frequencies of occurrence.† Omnibus P value assessed by permutation testing.

J Periodontol • October 2008 Holla, Fassmann, Augustin, Halabala, Znojil, Vanek

1931

with a larger sample size to confirm the usefulness ofIL-4 haplotypes as genetic markers in CP.

ACKNOWLEDGMENTS

Theauthors thankProf.AnnaVasku,chair,Departmentof Pathophysiology, Faculty of Medicine, MasarykUniversity, for her support of this research andA. Stejskalova, Department of Pathophysiology, Fac-ulty of Medicine, Masaryk University, for excellentlaboratory assistance. The study was supported by agrant fromtheMinistryofHealthof theCzechRepublic,Prague, Czech Republic (Internal Grant Agency:NR9129-3/06). The authors report no conflicts of in-terest related to this study.

REFERENCES1. Kornman KS, Knobelman C, Wang HY. Is periodontitis

genetic? The answer may be Yes! J Mass Dent Soc2000;49:26-30.

2. Okada H, Murakami S. Cytokine expression in peri-odontal health and disease. Crit Rev Oral Biol Med1998;9:248-266.

3. Shapira L, Van Dyke TE, Hart TC. A localized absenceof interleukin-4 triggers periodontal disease activity: Anovel hypothesis. Med Hypotheses 1992;39:319-322.

4. Seymour GJ, Gemmell E, Reinhardt RA, Eastcott J,Taubman MA. Immunopathogenesis of chronic inflam-matory periodontal disease: Cellular and molecularmechanisms. J Periodontal Res 1993;28:478-486.

5. Dennison DK, Van Dyke TE. The acute inflammatoryresponse and the role of phagocytic cells in periodontalhealth and disease. Periodontol 2000 1997;14:54-78.

6. Seymour GJ, Powell RN, Cole KL, et al. Experimentalgingivitis in humans: A histochemical and immuno-logical characterization of the lymphoid cell subpop-ulations. J Periodontal Res 1983;18:375-385.

7. Yamazaki K, Nakajima T, Hara K. Immunohistochem-ical analysis of T cell functional subsets in chronicinflammatory periodontal disease. Clin Exp Immunol1995;99:384-391.

8. Gemmell E, Marshall RI, Seymour GJ. Cytokines andprostaglandins in immune homeostasis and tissuedestruction in periodontal disease. Periodontol 20001997;14:112-143.

9. Mangan DF, Robertson B, Wahl SM. IL-4 enhancesprogrammed cell death (apoptosis) in stimulated hu-man monocytes. J Immunol 1992;148:1812-1816.

10. Marsh DG, Neely JD, Breazeale DR, et al. Linkage ana-lysis of IL4 and other chromosome 5q31.1 markersand total serum immunoglobulin E concentrations.Science 1994;264:1152-1156.

11. Rosenwasser LJ, Klemm DJ, Dresback JK, et al.Promoter polymorphisms in the chromosome 5 genecluster in asthma and atopy. Clin Exp Allergy 1995;25:74-78.

12. Rosenwasser LJ, Borish L. Genetics of atopy andasthma: The rationale behind promoter-based candi-date gene studies (IL-4 and IL-10). Am J Respir CritCare Med 1997;156:S152-S155.

13. Mout R, Willemze R, Landegent JE. Repeat polymor-phism in the interleukin-4 gene (IL4). Nucleic AcidsRes 1991;19:3763.

14. Vandenbroeck K, Goris A. Cytokine gene polymor-phisms in multifactorial diseases: Gateways to novel

targets for immunotherapy? Trends Pharmacol Sci2003;24:284-289.

15. Vandenbroeck K, Martino G, Marrosu MG, et al.Occurrence and clinical relevance of an interleukin-4gene polymorphism in patients with multiple sclerosis.J Neuroimmunol 1997;76:189-192.

16. Nakashima H, Miyake K, Inoue Y, et al. Associationbetween IL-4 genotype and IL-4 production in theJapanese population. Genes Immun 2002;3:107-109.

17. MichelJ,GonzalesJR,WunderlichD,DieteA,HerrmannJM, Meyle J. Interleukin-4 polymorphisms in early onsetperiodontitis. J Clin Periodontol 2001;28:483-488.

18. Scarel-Caminaga RM, Trevilatto PC, Souza AP, BritoRB Jr., Line SRP. Investigation of IL4 gene polymor-phism in individuals with different levels of chronicperiodontitis in a Brazilian population. J Clin Peri-odontol 2003;30:341-345.

19. Kang BY, Choi YK, Choi WH, et al. Two polymor-phisms of interleukin-4 gene in Korean adult peri-odontitis. Arch Pharm Res 2003;26:482-486.

20. Gonzales JR, Kobayashi T, Michel J, Mann M, YoshieH, Meyle J. Interleukin-4 gene polymorphisms inJapanese and Caucasian patients with aggressiveperiodontitis. J Clin Periodontol 2004;31:384-389.

21. Pontes CC, Gonzales JR, Novaes AB Jr., et al.Interleukin-4 gene polymorphism and its relation toperiodontal disease in a Brazilian population of Africanheritage. J Dent 2004;32:241-246.

22. Gonzales JR, Mann M, Stelzig J, Bodeker RH, Meyle J.Single-nucleotide polymorphisms in the IL-4 and IL-13promoter region in aggressive periodontitis. J ClinPeriodontol 2007;34:473-479.

23. Kara N, Keles GC, Sumer P, et al. Association of thepolymorphisms in promoter and intron regions of theinterleukin-4 gene with chronic periodontitis in aTurkish population. Acta Odontol Scand 2007;65:292-297.

24. Ainamo J, Barmes D, Beagrie G, Cutress T, Martin J,Sardo-Infirri J. Development of the World HealthOrganization (WHO) community periodontal index oftreatment needs (CPITN). Int Dent J 1982;32:281-291.

25. Holla LI, Buckova D, Fassmann A, Halabala T, VaskuA, Vacha J. Promoter polymorphisms in the CD14receptor gene and their potential association with theseverity of chronic periodontitis. J Med Genet 2002;39:844-848.

26. Armitage GC. Development of a classification systemfor periodontal diseases and conditions. Ann Peri-odontol 1999;4:1-6.

27. Muhlemann HR, Mazor ZS. X-ray diagnosis (in Ger-man). Schweiz Mschr Zahnheilk 1955;65:1005-1013.

28. McDevitt MJ, Wang HY, Knobelman C, et al. Interleu-kin-1 genetic association with periodontitis in clinicalpractice. J Periodontol 2000;71:156-163.

29. Sambrook J, Fritsch EF, Maniatis T. Molecular Clon-ing: A Laboratory Manual, 2nd ed. New York: ColdSpring Laboratory Press; 1989:9-17.

30. Hegab AE, Sakamoto T, Saitoh W, et al. Polymor-phisms of IL4, IL13, and ADRB2 genes in COPD. Chest2004;126:1832-1839.

31. Lewontin RC. On measures of gametic disequilibrium.Genetics 1988;120:849-852.

32. Stephens M, Smith NJ, Donnelly P. A new statisticalmethod for haplotype reconstruction from populationdata. Am J Hum Genet 2001;68:978-989.

IL-4 Haplotypes in Periodontitis Volume 79 • Number 10

1932

33. Stephens M, Donnelly P. A comparison of Bayesianmethods for haplotype reconstruction. Am J HumGenet 2003;73:1162-1169.

34. Hoehe MR. Haplotypes and the systematic analysis ofgenetic variation in genes and genomes. Pharmacoge-nomics 2003;4:547-570.

35. Crawford DC, Nickerson DA. Definition and clinicalimportance of haplotypes. Annu Rev Med 2005;56:303-320.

36. Zong M, Yang PS, Qi XM, Yi XH. Changes of circulat-ing IFN-gamma, IL-4 in patients with chronic peri-odontitis before and after periodontal initial therapy(in Chinese). Shanghai Kou Qiang Yi Xue 2005;14:131-133.

37. Burchard EG, Silverman EK, Rosenwasser LJ, et al.Association between a sequence variant in the IL-4gene promoter and FEV1 in asthma. Am J Respir CritCare Med 1999;160:919-922.

38. Sandford AJ, Chagani T, Zhu S, et al. Polymorphismsin the IL4, IL4RA, and FCERIB genes and asthmaseverity. J Allergy Clin Immunol 2000;106:135-140.

39. Buchs N, Silvestri T, di Giovine FS, et al. IL-4 VNTRpolymorphism in chronic polyarthritis. The rare allele

is associated with protection against destruction.Rheumatology (Oxford) 2000;39:1126-1131.

40. Genevay S, Di Giovine FS, Perneger TV, et al. Asso-ciation of interleukin-4 and interleukin-1b gene vari-ants with Larsen score progression in rheumatoidarthritis. Arthritis Rheum 2002;47:303-309.

41. Bartova J, Kratka-Opatrna Z, Prochazkova J, et al.Th1 and Th2 cytokine profile in patients with earlyonset periodontitis and their healthy siblings. Media-tors Inflamm 2000;9:115-120.

42. Yun LW, Decarlo AA, Collyer C, Hunter N. Enhance-ment of Th2 pathways and direct activation of B cellsby the gingipains of Porphyromonas gingivalis. ClinExp Immunol 2003;134:295-302.

Correspondence: Dr. Lydie Izakovicova Holla, Departmentof Pathophysiology, Faculty of Medicine, Masaryk Uni-versity, Kamenice, pav. A18, 625 00 Brno, CzechRepublic. Fax: 420-549494340-2330; e-mail: holla@med.muni.cz.

Submitted January 16, 2008; accepted for publicationMarch 31, 2008.

J Periodontol • October 2008 Holla, Fassmann, Augustin, Halabala, Znojil, Vanek

1933