Abstract The efficacy of IgG-induced Fc gamma receptor(FcR) function displays interindividual heterogeneity dueto genetic polymorphisms of three FcR subclasses:FcRIIa, FcRIIIa and FcRIIIb. FcR polymorphisms maycontribute to disease susceptibility or may alter diseasecourse. The aim of this study is to examine FcR gene poly-morphisms in Turkish children with recurrent respiratorytract infections and without well known humoral immun-odeficiencies. For the patients in the study group (n=52),recurrent infection was defined as the presence of at leastsix infection episodes a year. Seventy-one healthy childrenwith a maximum of two infections in a year were enrolledas the control group. Subjects in both groups had no abnor-malities in serum immunoglobulins, IgG subsets and spe-cific antibody levels. For FcRIIa: H131H, H131R, R131Rgenotypes and 131R, 131H alleles; for FcRIIIa: F158F,F158V, V158V genotypes and 158F, 158V alleles; and forFcRIIIb: NA1/NA1, NA1/NA2, NA2/NA2 genotypesand NA1, NA2 alleles were determined by using amplifi-cation refractory mutation system polymerase chain reac-tion (ARMS-PCR). Compared with the control group, theFcRIIa-R131R genotype and 131R allele were found to be

significantly elevated in the study group, and FcRIIa-H131H genotype and 131H allele in the study group weresignificantly lower than in the control group. Genotypesand alleles related with FcRIIIa and FcRIIIb gene poly-morphisms did not show any significant difference betweenthe study and control groups. FcRIIa gene polymorphism(R131R) may increase the risk and susceptibility for recur-rent infectious diseases in children.

Key words FcR gene polymorphisms Recurrent infec-tions Congenital immune deficiencies

Introduction

Fc receptor (FcR)-induced leukocyte functions includingantibody-dependent cellular cytotoxicity, phagocytosis,superoxide generation, degranulation, cytokine productionand regulation of antibody production are all essential forhost defence. The efficacy of IgG-induced FcR functiondisplays interindividual heterogeneity due to genetic poly-morphisms of three FcR subclasses: FcRIIa, FcRIIIaand FcRIIIb [1].

Three FcR subclasses display functionally relevantgenetically determined polymorphisms [1]. FcRIIa dis-plays a G to A point mutation in the region specifying itsligand-binding domain, causing an arginine (R) to histidine(H) amino acid substitution at position 131. FcRIIa-H131allotype displays higher binding efficiency for human IgG2and IgG3, compared to FcRIIa-R131 [25]. Importantly,FcRIIa-H131 represents the sole receptor capable of effi-cient interaction with IgG2 [2]. The phagocytic capacity ofleukocytes with R131R for IgG2- and IgG3-opsonised bac-teria is lower than the ones with H131H [2].

The FcRIIIa gene displays a T to G substitution atnucleotide 559, resulting in a valine (V) to phenylalanine(F) substitution at amino acid position 158. FcRIIIa-

Clin Exp Med (2006) 6:2732DOI 10.1007/s10238-006-0090-y

C. Ozturk G. Aksu A. Berdeli N. Kutukculer

Fc gamma RIIa, IIIa and IIIb polymorphisms in Turkish children susceptible to recurrent infectious diseases

Received: 22 June 2005 / Accepted: 4 February 2006

O R I G I N A L

C. Ozturk G. Aksu A. Berdeli N. Kutukculer ()Department of Pediatrics,Faculty of Medicine,Ege University,35100 Bornova-zmir, Turkeye-mail: necil.kutukculer@ege.edu.trFax: +90-232-388-9900

V158V genotype displays higher affinity for IgG1, IgG3and IgG4 [1]. FcRIIIb bears the neutrophil antigen (NA)polymorphism in its membrane-distal Ig-like domain.FcRIIIb-NA1 induces phagocytosis of IgG1- and IgG3-opsonised particles and binds immune complexed IgG3more efficiently than the FcRIIIb-NA2 allotype [6].

Certain studies proved that FcR polymorphism leads torecurrent infectious diseases as a result of relatively inef-fective bacterial phagocytosis [79]. These polymorphismsmay contribute to the disease by either increasing vulnera-bility to the disease or changing the disease course [1].

The aim of this study is to examine FcR gene poly-morphisms in Turkish children with recurrent respiratorytract infections and without well known humoral immun-odeficiencies.

Patients and methods

The study consisted of 52 Turkish children with recurrent febrileinfections (20 male and 32 female; ages 5.202.90 years) in thestudy group (SG) and 71 healthy children with a maximum oftwo infections in a year (32 male and 39 female; ages 6.072.46years) in the control group (CG). In this study, other than healthychildren in the control group, children with recurrent respiratorytract infections and without well known humoral and cellularcongenital immune deficiencies were examined for FcR poly-morphisms. Recurrent infection was defined as the presence of atleast six febrile infection episodes in a year. Patients who havesome abnormalities in serum immunoglobulins, IgG subsets andspecific antibody levels (IgG antibodies against polyribosyl-ribitol-phosphate of Haemophilus influenzae type B and anti-tetanus toxoid antibodies) and who have clinical symptoms andlaboratory findings of a suspected cellular immune deficiency orchronic diseases or malignancies were excluded. Parents provid-ed written informed consent prior to the study.

Methods

The quantifications of serum IgA, IgG, IgM and IgG1, IgG2, IgG3were performed by the nephelometric method on the Dade BehringBN2 Nephelometer Analyzer and commercially available kits byDade Behring (Germany). The enzyme immunoassay techniquewas used for the determination of IgG antibodies against polyribo-sylribitol-phosphate of H. influenzae type B and anti-tetanus tox-oid antibodies (Immunozym HiB IgG and Immunozyme Tetanus,Progen Biotechnik, Heidelberg, Germany).

DNA purification

Genomic DNA from patients and healthy controls was extractedfrom peripheral blood leukocytes with QIAamp DNA BloodMini Kits (Qiagen GmbH, Hilden, Germany) according to themanufacturers instructions.

28

FcRIIa genotyping

FcRIIa genotyping was performed by using the method definedby Flesch et al. [10]. We used a 25-l PCR mixture containing 2.5I of genomic DNA, 2.5 I of 10x PCR buffer (AppliedBiosystems, Foster City, CA, USA), 2 mM MgCl2, 200 mol/I ofeach dNTP (Promega, Madison, USA), and 0.5 U AmpliTag DNAPolymerase (Applied Biosystems). In addition, 0.5 mol/l ofH131-specific sense primer (5-ATCC-CAGAAATTCTCCCA-3)from the second extracellular domain or 0.5 mol/l R131-specificsense primer (5-ATCCCAGAAATTCTCCCG-3) was used. Allprimer oligonucleotides were synthesised by Qiagen Operon Co.and 0.5 mol/I common anti-sense primer from an area down-stream of the intron (5-CAATTTTGCTGCTATGGGC-3). Theresulting fragment was 253 bp in length. As internal PCR control,we used 0.125 mol/I human growth hormone (HGH)-1 forwardprimer (5-CAGTGCCTTCCCAACCATTCCCTTA-3) and 0.125mol/I HGH-II reverse primer (5-ATCCACTCACGGATTTCT-GTTGTGTTTC-3), which resulted in a 439-bp fragment.

A thermal cycler (GeneAmp 9700, Applied Biosystems) wasused to perform a hot-start PCR as follows: 5 min at 95C, 10cycles of 1 min at 95C, 2 min at 57C, and 1 min at 72C; there-after, to enhance the sensitivity, we used 22 cycles of 1 min at95C, 2 min at 54C, 1 min at 72C and a final extension step for5 min at 72C. The PCR amplification products were separatedon a 2% agarose gel and visualised by ethidium bromide staining.

FcRIIIa genotyping

FcRIIIa genotyping was performed by using the method of vanden Berg et al. [11]. In this amplification refractory mutation sys-tem-PCR, the allele-specific primers (KIM-G (V): 5TCTCTGAA-GACACATTTCTACTCCCTAC-3 and KIM-1 (F): 5-TCTCT-GAAGACACATTTCTACTCCCTAA-3) amplified a 160-bp frag-ment with their antisense downstream primer A013 (5-ATATT-TACAGAATGGCACAGG-3). The internal control primers wereCtrl-1 and Ctrl-2, amplifying a 270-bp piece of DNA. Again, twoPCR procedures were required for genotyping of a subject.

Reactions were performed with 175 ng of genomic DNA in a 25-I reaction volume, containing 2.5 I PCR Gold buffer, 62.5 nmolMgCl2, 1.25 nmol each of the four dTNPs, 11.9 pmol KIM-G (V),66.4 pmol KIM-G (F), 15.4 pmol A013, 1.1 pmol each of the twocontrol primers and 0.5 U Ampli Tag Gold (Applied Biosystems).

PCR conditions were denaturation for 10 min at 95C; 32cycles of 95C for 30 s, 57C for 20 s, 72C for 25 s, and a finalextension step at 72C for 7 min.

FcRIIIb genotyping

FcRIIIb gene genotyping was performed by using the PCRmethod of Bux et al. [12]. The NA1-specific primer (5-CAGTG-GTTTCACAATGTGAA-3) gave a 141-bp fragment, and theNA2-specific primer (5-CAATGGTACAGCGTGCTT-3) ampli-fied a 219-bp fragment. The reverse primer (5-ATGGACTTC-TACCTGCAC-3) did not discriminate between the two allo-types. Because there was a substantial difference in lengthbetween the NA1-specific and NA2-specific reaction products,

C. Ozturk et al.: Fc gamma receptor polymorphisms in Turkish children

both alleles could be detected in the same reaction. The 25-lreaction mixture used contained 100 ng of genomic DNA, 3.7 l10x PCR buffer (Applied Biosystems), 25 nM MgCl2, 1 nmol/l ofeach dNTP (Promega), 4 pmol of each of the control primers,0.012 nmol of the NA1 and NA2 primers, 0.025 nmol of thereverse primer and 0.5 U AmpliTag DNA polymerase (AppliedBiosystems). After denaturation for 3 min at 94C, 30 PCR cyclesconsisting of 94C for 1 min, 57C for 2 min and 72C for 1 minwere run. After a final extension of 1 min at 72C, samples wereresolved in 2% agarose gels stained with ethidium bromide.

To validate accuracy and reproducibility of the resultsobtained using the above described techniques, we randomly ran20% of samples from both groups (study and control groups) foreach FcRIIa, FcRIIIa and FcRIIIb genotype using a directDNA sequencing technique with ABI PRISM 310 GeneticAnalyzer System (Applied Biosystems).

Statistical analysis

Statistical analysis of the data was performed with SPSS package v.11.0 program. Differences in demographic factors between the SGand CG were tested with the chi-square test. Genotypic distributionand allelic frequencies were statistically defined by the chi-squaretest. Odds ratios (OR) with 95% confidence intervals (95% CI) werecalculated using the same software. Deviation from the Hardy-Weinberg expectancy was examined with the chi-square test.

Results

There was no significant difference between the SG andCG in ages, immunoglobulin levels, immunoglobulin G

subsets and specific antibody responses (meansstandarddeviations are listed in Table 1). Serum immunoglobu-lins, IgG subsets of each subject in both groups were alsocompared to age-related normal levels [13] one by oneand all of them were within normal ranges. Specific anti-body responses against tetanus and H. influenzae in sub-jects who were vaccinated before were all above protec-tive levels (protective antibody levels against tetanus andH. influenzae are 0.1 IU/ml and 1 g/ml, respectively[14, 15]. In other words, none of the subjects in the SGor CG had signs of a well known congenital humoralimmunodeficiency.

Mean frequency of annual infections was higher in theSG than CG (10.513.24 and 0.550.63 respectively,P

When we examine the allele frequencies, the odds ratiosthat are calculated for R131-SG/CG in relation to H131-SG/CG result in a single odds ratio of 2.43 (P=0.001)(Table 3). The others (F158-V158 and NA1-NA2) did notshow any significant difference (P>0.05) (Table 3).

FcRIIa-R131R, FcRIIIa-F158F and FcRIIIb-NA2/NA2 genotype combinations, which represent theleast efficient FcR combinations, were determined in theSG (only two patients, 3.8%) and in the CG (one child,1.4%). FcRIIa-R131R and FcRIIIa-F158F genotypecombinations were also found in small numbers in the SGand CG (n=5, 9.6% and n=2, 2.8%, respectively). In theSG, FcRIIa-R131R and FcRIIIb-NA2/NA2 genotypecombination and FcRIIIa-F158F and FcRIIIb-NA2/NA2genotype combination were found to be 7.7% and 11.5%,respectively, while these are less than 4% in the CG. Thenumber of cases having these least efficient combinationswas not enough, so that statistical analysis was not per-formed in order not to have false positive results.

Discussion

FcR polymorphisms are now considered to be heritablerisk factors for autoimmune and infectious diseases and

30

support for a relevant role of these polymorphisms hasbeen obtained in previous studies [16]. Results from asso-ciation studies are sometimes contradictory, possibly dueto differences in inclusion criteria, sample size, ethnicityand definitions of disease severity [1]. However, there isconvincing evidence that FcRIII and FcRII alleles arerisk factors for systemic lupus erythematosus [17, 18].

The FcRIIa-R131R genotype was found to be enrichedamong American bacteraemic pneumonia patients, suggest-ing that inefficient FcRIIa-IgG2 interactions increase hostsusceptibility to S. pneumonia [5]. On the other hand,decreased frequencies of FcRIIa-H131H genotypesamong Dutch children with upper airway infections by S.pneumonia have been reported [4]. FcRIIa-R131R andFcRIIIb-NA2/NA2 genotype combination, which repre-sents the least efficient FcR combination on neutrophils,was increased among complement-deficient patients withhigh frequencies of meningococcal disease [19]. Tezcan etal. [20] have reported that the FcRIIa polymorphismmight be a contributing factor to increased susceptibility tomeningitis with encapsulated bacteria in children belowtwo years of age. In some other reports, it has been shownthat patients carrying the FcRIIa-R131R polymorphismsuffer from different severe infections and infection com-plications [21, 22]. In our previous study, a significantincrease was found in the FcRIIa-R131R genotype in

C. Ozturk et al.: Fc gamma receptor polymorphisms in Turkish children

Table 2 Genotype frequency of FcRIIa, FcRIIIa and FcRIIIb in the study and control groups

Gene Fc Study group Study group Control group Control group P OR OR (95% CI)observed expected, na observed expected, na observed observed

n % n %

IIa-HH 17 32.7 15.02 37 52.1 38.76 1.000 1.00 0.35

patients with rheumatic fever caused by group A strepto-cocci leading to high risk, but the FcRIIa-H131R geno-type was shown to cause an intermediate risk [23].

In this study, it was attempted to determine whetherFcRIIa, FcRIIIa and FcRIIIb polymorphisms lead tosusceptibility to febrile infections in Turkish children.With this aim, children with normal serum immunoglobu-lins, IgG subgroups and specific antibody responses werestudied. In other words, control and study groups had nowell known congenital immunodeficiency predisposing toinfections. The only difference between the two groupswas annual infection rate and ages at the beginning ofinfections.

In our study, the FcRIIa-R131R genotype in the SG(25.0%) was significantly higher than that in the CG(4.2%) (P=0.001). This finding led us to suggest that theR131R genotype makes Turkish children vulnerable toinfectious diseases. Similarly, the R131 allele was higherin the SG, whereas H131 was higher in the CG, which sug-gests that existence of the R131 allele alone increases therisk of infectious diseases. However, it has to be noted thatin genetic association studies with such small groups ofpatients and healthy controls, interpretation of the datasometimes yields false positive results.

Although in the literature there are studies suggestingthe existence of a relationship between infectious diseaseattacks and FcRIIIa and RIIIb polymorphisms [2426],we could not confirm this relationship in this study. If thisstudy is considered to be a preliminary study due to thelow number of subjects, in a more comprehensive study,which could be carried out in Turkish population, the rela-tionship between genetic polymorphism of the two differ-ent forms of the FcRIII gene and infections may be estab-lished. In addition, the least effective FcR gene polymor-phism combinations were not found to be high enough inboth groups to compare statistically.

In conclusion, it has been found that the FcRIIa genepolymorphism (R131R) is a risk factor for infectious dis-ease and this polymorphism may contribute to susceptibil-ity to recurrent infections in children.

References

1. van Sorge NM, van der Pol WL, van de Winkel JG (2003)FcgammaR polymorphisms: implications for function, dis-ease susceptibility and immunotherapy. Tissue Antigens61:189202

2. Sanders LA, Feldman RG, Voorhorst-Ogink MM, de Haas M,Rijkers GT, Capel PJ, Zegers BJ, van de Winkel JG (1995)Human immunoglobulin G (IgG) Fc receptor IIa (CD32)polymorphism and IgG2-mediated bacterial phagocytosis byneutrophils. Infect Immun 63:7381

3. Parren PW, Warmerdam PA, Boeije LC, Arts J, WesterdaalNA, Vlug A, Capel PJ, Aarden LA, van de Winkel JG (1992)

On the interaction of IgG subclasses with the low affinity Fcgamma RIIa (CD32) on human monocytes, neutrophils, andplatelets. Analysis of a functional polymorphism to humanIgG2. J Clin Invest 90:15371546

4. Lehrnbecher T, Foster CB, Zhu S, Leitman SF, Goldin LR,Huppi K, Chanock SJ (1999) Variant genotypes of the low-affinity Fcgamma receptors in two control populations and areview of low-affinity Fcgamma receptor polymorphisms incontrol and disease populations. Blood 94:42204232

5. Salmon JE, Edberg JC, Brogle NL, Kimberly RP (1992)Allelic polymorphisms of human Fc gamma receptor IIA andFc gamma receptor IIIB. Independent mechanisms for differ-ences in human phagocyte function. J Clin Invest89:12741281

6. Salmon JE, Millard SS, Brogle NI, Kimberly RP (1995) Fc receptor IIIb enhances Fc receptor IIa function in an oxidantdependent and allele-sensitive manner. J Clin Invest95:28772885

7. Rodriguez ME, van der Pol WL, Sanders LA, van de WinkelJG (1999) Crucial role of FcgammaRIIa (CD32) in assess-ment of functional anti-Streptococcus pneumoniae antibodyactivity in human sera. J Infect Dis 179:423433

8. Sanders LA, van de Winkel JG, Rijkers GT, Voorhorst-OginkMM, de Haas M, Capel PJ, Zegers BJ (1994) Fc gammareceptor IIa (CD32) heterogeneity in patients with recurrentbacterial respiratory tract infections. J Infect Dis 170:854861

9. Yee AM, Phan HM, Zuniga R, Salmon JE, Musher DM (2000)Association between FcgammaRIIa-R131 allotype and bac-teremic pneumococcal pneumonia. Clin Infect Dis 30:2528

10. Flesch BK, Bauer F, Neppert J (1998) Rapid typing of humanFc RIIa receptor polymorphism by polymerase chain reac-tion amplification with allele specific primers. Transfusion38:174176

11. Van den Berg L, Mhyr KM, Kluge B, Vedeler CA (2001) Fc receptor polymorphisms in populations in Ethiopia andNorway. Immunology 104:8791

12. Bux J, Stein EL, Santoso S, Mueller-Eckhardt C (1995) NAgene frequencies in the German population, determined bypolymerase chain reaction with sequence-specific primers.Transfusion 35:5457

13. Kutukculer N, Genel F, Koturoglu G, Aksu G, Kurugol Z,Caglayan Z (2006) Serum immunoglobulins (IgG, IgM, IgA)and IgG subclass concentrations in healthy children: a studyusing nephelometric technique. Turk J Pediatr 48:1924

14. Schroder JP, Kuhlmann WD (1992) Preventive tetanus immu-nization and avoidance of side effects of booster immuniza-tion (in German). Dtsch Med Wochenschr 117:19031906

15. Breukels MA, Spanjaard L, Sanders LA, Rijkers GT (2001)Immunological characterization of conjugated Haemophilusinfluenza type b vaccine failure in infants. Clin Infect Dis32:17001705

16. Rascu A, Repp R, Westerdaal NA, Kalden JR, van de WinkelJG (1997) Clinical relevance of Fc gamma receptor polymor-phisms. Ann N Y Acad Sci 815:282295

17. Koene HR, Kleijer M, Swaak AJ et al (1998) TheFcgammaRIIIA:158F allele is a risk factor for systemic lupuserythematosus. Arthritis Rheum 41:18131818

18. Karassa FB, Trikalinos TA, Ioannidis JP (2002) Role of theFcgamma receptor IIa polymorphism in susceptibility to sys-temic lupus erythematosus and lupus nephritis: a meta-analy-sis. Arthritis Rheum 46:15631572

C. Ozturk et al.: Fc gamma receptor polymorphisms in Turkish children 31

19. Fijen CA, Bredius RG, Kuijper EJ (1993) Polymorphism ofIgG Fc receptors in meningococcal disease. Ann Intern Med119:636638

20. Tezcan I, Berkel AI, Ersoy F, Sanal O, Kanra G (1998) Fcgamma receptor allotypes in children with bacterial meningi-tis. Turk J Pediatr 40:533538

21. Bredius RG, Derkx BH, Fijen CA, de Wit TP, de Haas M,Weening RS, van de Winkel JG, Out TA (1994) Fc gammareceptor IIa (CD32) polymorphism in fulminant meningococ-cal septic shock in children. J Infect Dis 170:848853

22. Platonov AE, Shipulin GA, Vershinina IV, Dankert J, van deWinkel JG, Kuijper EJ (1998) Association of human Fcgamma RIIa (CD32) polymorphism with susceptibility to andseverity of meningococcal disease. Clin Infect Dis 27:746750

23. Berdeli A, Celik HA, Ozyurek R, Aydin HH (2004)Involvement of immunoglobulin FcgammaRIIa and

32

FcgammaRIIIb gene polymorphisms in susceptibility torheumatic fever. Clin Biochem 37:925929

24. Kobayashi T, Westerdaal NA, Miyazaki A, van der Pol WL,Suzuki T, Yoshie H, van de Winkel JG, Hara K (1997)Relevance of immunoglobulin G Fc receptor polymorphismto recurrence of adult periodontitis in Japanese patients. InfectImmun 65:35563560

25. Sugita N, Yamamoto K, Kobayashi T, Van Der Pol W,Horigome T, Yoshie H, Van De Winkel JG, Hara K (1999)Relevance of Fc gamma RIIIa-158V-F polymorphism torecurrence of adult periodontitis in Japanese patients. ClinExp Immunol 117:350354

26. Meisel P, Carlsson LE, Sawaf H, Fanghaenel J, Greinacher A,Kocher T (2001) Polymorphisms of Fc gamma-receptorsRIIa, RIIIa, and RIIIb in patients with adult periodontal dis-eases. Genes Immun 2:258262

C. Ozturk et al.: Fc gamma receptor polymorphisms in Turkish children