Fcγ receptor polymorphisms in populations in Ethiopia and Norway

Abstract

Seventy-seven healthy Ethiopians were genotyped for polymorphisms in the immunoglobulin G Fc receptors (FcγR) FcγRIIa, FcγRIIIa and FcγRIIIb, including the SH allele. The genotype and allele frequencies were compared with those of 96 healthy Norwegians. Ethiopians had higher frequencies of the SH-FcγRIIIb (P = 0·001), FcγRIIIa-158 V (P = 0·026) and FcγRIIIb-Na2 (P = 0·046) alleles. The genotype distributions of FcγRIIa, FcγRIIIa and FcγRIIIb, however, did not differ significantly from those of the Norwegians. The data were also compared with those reported from studies on other ethnic groups. The variation of different polymorphisms both within and between ethnic groups may influence differences in the incidence rates of infectious and autoimmune diseases.

Introduction

The three subgroups of human immunoglobulin G (IgG) Fc receptors (FcγRI, FcγRII and FcγRIII) each form a heterogeneous family of molecules, encoded by distinct genes with alternative splicing variants.^1,2 The receptors differ in their distribution on different cell types, strength and capacity of binding different IgG subclasses, and type of intracellular signals. FcγRI (CD64), which is primarily expressed on monocytes and macrophages, binds monomeric IgG with high affinity. The others are all low-affinity receptors. Members of the FcγRII (CD32) group are found on most leucocytes and platelets. FcγRIIIa (CD16) is mainly expressed on natural killer cells and macrophages and FcγRIIIb on neutrophils. With the help of these receptors, the IgG molecule with its constant region can start a broad range of immune responses, such as antibody-dependent cellular cytotoxicity, endocytosis, phagocytosis, release of inflammatory mediators and augmentation of antigen presentation, depending on the identity of the FcγR-expressing cell. These immune responses can, however, be inhibited by cross-binding of FcγRIIb, primarily found on B lymphocytes, macrophages, neutrophils and mast cells. Co-ligation of both activation and inhibitory FcγR will therefore determine the magnitude of the effector cell responses.^1,2

Polymorphisms in the human FcγR genes IIa, IIIa and IIIb further enhance the heterogeneity of this molecule class. FcγRIIa has two co-dominantly expressed alleles, H131 and R131, based on their interaction with murine IgG1. Their single amino acid difference turns out to be essential for IgG binding; H131 has a substantially higher affinity for IgG2.³ Moreover, phagocytes of the H/H genotype proved to have a higher phagocytosis capability than H/R or R/R cells.⁴ FcγRIIIa is also biallelic; the 158F allele binds IgG1, IgG3 and IgG4 less avidly than 158V.⁵ Linked to this polymorphism is a triallelic polymorphism at amino acid position 48.⁶ The two allotypes of FcγRIIIb (Na1 and Na2) differ in at least five nucleotides, resulting in four different amino acids. Polymorphonuclear leucocytes of the Na1/Na1 genotype have been reported to show higher phagocytosis than those of the other genotypes.⁴ In addition, another variant of this gene (FcγRIIIb-SH) also exists.⁷ Only one nucleotide differs between this SH gene and the Na2 allele. Gene deletions or duplications of FcγRIIIb have frequently been observed, the latter especially in SH⁺ individuals.⁸

FcγR polymorphisms may influence the vigour of the inflammatory response and may contribute to differences in susceptibility to infectious and autoimmune diseases.⁹ There is ethnic variation in the frequency distribution of the different allotypes.¹⁰ Since the prevalence of autoimmune and infectious diseases also varies between different ethnic groups, this might be of clinical importance. So far there has been no study of FcγR polymorphisms in negroid people from the African continent. In this project, we measured the frequencies of the three polymorphisms in a population of 77 healthy blood donors in Ethiopia and compared them with the frequencies from a group of blood donors in Norway.

Materials and methods

Subjects

The African blood samples were obtained from 77 healthy Ethiopian blood donors (17% female) with a mean age of 29·8 years (SD = 10·0). The donors were all from Addis Ababa but were not related. Data on caucasoid people have partly been published before and were obtained from 96 healthy Norwegian subjects.¹¹

DNA purification

Genomic DNA was purified from peripheral blood leucocytes using QIAamp DNA Blood Mini Kits 50 (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions.

Polymerase chain reaction (PCR) for FcγRIIa

Briefly, genotyping was performed using amplification refractory mutation system–PCR modified from Botto et al.¹² Five oligonucleotide primers were used. Two primers from the T-cell receptor Vα22 gene (Ctrl-1: 5′-GAT TCA GTG ACC CAG ATG GAA GGG-3′ and Ctrl-2: 5′-AGC ACA GAA GTA CAC CGC TGA GTC-3′) amplified a fragment of 270 bp and were used as an internal positive control. The other primers bound to sequences in the FcγRIIa gene exon. EC2-131R (5′-CCA GAA TGG AAA ATC CCA GAA ATT CTC TCG-3′) and EC2-131H (5′-CCA GAA TGG AAA ATC CCA GAA ATT CTC TCA-3′) were allele-specific primers. In co-operation with the antisense downstream primer TM1 (5′-CCA TTG GTG AAG AGC TGC CCA TGC TGG GCA-3′), these primers amplified a 980-bp fragment. Two separate PCR reactions were performed for genotyping of each individual.

The 25 µl PCR reaction mixture contained 100 ng genomic DNA, 2·5 µl 10× PCR buffer (Applied Biosystems, Stockholm, Sweden), 57·5 nmol MgCl₂, 1 nmol each of the four dNTPs, 1·1 pmol each of the two control primers, 0·011 nmol of the other three primers and 1·75 U of Taq DNA polymerase (Perkin Elmer).

PCR conditions were denaturation for 5 min at 94°, followed by 45 cycles of 94° for 45 seconds, 63° for 30 seconds and 72° for 1·5 min, and a final extension step at 72° for 10 min.

PCR for FcγRIIIa

In this amplification refractory mutation system–PCR, the allele-specific primers (KIM-G (V): 5′-TCT CTG AAG ACA CAT TTC TAC TCC CTA C-3′ and KIM-1 (F): 5′-TCT CTG AAG ACA CAT TTC TAC TCC CTA A-3′) amplified a 160-bp fragment with their antisense downstream primer A013 (5′-ATA TTT ACA GAA TGG CAC AGG-3′). The control primers were Ctrl-1 and Ctrl-2 (see FcγRIIa), amplifying a 270-bp piece of DNA. Again, two PCR procedures were required for genotyping of a subject.

Reactions were performed with 175 ng of genomic DNA in a 25-µl reaction volume, containing 2·5 µl PCR Gold Buffer, 62·5 nmol MgCl₂, 1·25 nmol each of the four dNTPs, 11·9 pmol KIM-G (V), 66·4 pmol KIM-G (F), 15·4 pmol A013, 1·1 pmol each of the two control primers and 0·5 U Ampli Taq Gold (Applied Biosystems).

PCR conditions were denaturation for 10 min at 94°; 32 cycles of 95° for 30 seconds, 57° for 20 seconds and 72° for 25 seconds; and a final extension step at 72° for 7 min.

PCR for FcγRIIIb

Two human growth hormone (hGH) primers (hGH-I: 5′-CAG TGC CTT CCC AAC CAT TCC CTT A-3′ and hGH-II: 5′-ATC CAC TCA CGG ATT TCT GTT GTG TTT C-3′) were used as internal positive controls, amplifying a 439-bp fragment. The Na1-specific primer (5′-CAG TGG TTT CAC AAT GTG AA-3′) gave a 141-bp fragment, and the Na2-specific primer (5′-CAA TGG TAC AGC GTG CTT-3′) amplified a 219-bp fragment. The reverse primer (5′-ATG GAC TTC TAG CTG CAC-3′) did not discriminate between the two allotypes. Since there was a substantial difference in length between the Na1-specific and Na2-specific reaction products, both alleles could be detected in the same reaction.

The reaction mixture (volume 25 µl) contained 100 ng genomic DNA, 3·7 µl 10× PCR buffer, 25·0 nmol MgCl₂, 1 nmol each of the four dNTPs, 4·0 pmol each of the two control primers, 0·013 nmol of the Na1 and Na2 primers, 0·025 nmol of the reverse primer and 0·5 U Taq DNA polymerase.

PCR conditions were denaturation for 3 min at 94° followed by 30 cycles of 94° for 1 min, 57° for 2 min and 72° for 1 min, and a final extension step at 72° for 10 min.

PCR for FcγRIIIb-SH

The SH⁺ (5′-ACT GTC GTT GAC TGT GTC AT-3′) and SH⁻ (5′-ACT GTC GTT GAC TGT GTC AG-3′) primers both amplified 191-bp products. Their reverse primer was Anti-SH (5′-AAG ATC TCC CAA AGG CTG TG-3′). Only subjects that turned out to be SH-positive were tested with the SH⁻ primer. The hGH primers were used as an internal positive control.

The 23 µl reaction mixture contained 2 µl genomic DNA, 2·25 µl 10× PCR buffer, 56·3 nmol MgCl₂, 1·8 nmol each of the four dNTPs, 180 nmol of the SH⁺ or SH⁻ primer and of the anti-SH primer, 2·9 pmol each of the two control primers and 0·8 U Taq DNA polymerase.

PCR conditions were denaturation for 3 min at 94°; 30 cycles of 95° for 30 seconds, 60° for 1 min and 71° for 30 seconds; and a final extension step at 72° for 7 min.

Analysis of the products

After electrophoresis in 1·5% agarose gel and staining with ethidium bromide, the amplification products were visualized with ultraviolet light.

Statistical analysis

Pearson's χ² test was used for analysing the data. Statistical significance was defined as P = 0·05. Allele frequencies were compared using 2 × 2 tables containing the number of alleles in the study groups. Tests were performed on 3 × 3 tables for checking whether the genotypes on two loci were independent.

Results

No significant differences were found between the Ethiopian and Norwegian genotype and allele frequencies of FcγRIIa (Table 1). FcγRIIIa and FcγRIIIb genotype frequencies did not differ either, but significant differences were found when allele frequencies were compared (Tables 2, 3). Ethiopian subjects had a higher frequency of FcγRIIIa-158V (P = 0·026) and FcγRIIIb-Na2 (P = 0·046). All subjects had at least one copy of the FcγRIIIb gene. The Ethiopian group contained a higher percentage of SH-positive individuals than the Norwegian group (P = 0·001). SH-FcγRIIIb was only found in Na2-positive people. The SH⁻ primer amplified a 191-bp band in all 13 SH-positive individuals. The five of these 13 that had been genotyped as Na1/Na2 heterozygotes are thus expected to possess at least three genes coding for FcγRIIIb, assuming that the SH⁻ primer was not able to attach to the Na1 sequence.

Table 1.

Distribution of FcγRIIa genotypes and allele frequencies in Ethiopians and Norwegians

	Genotype, n (%)			Allele freq.
	H/H	H/R	R/R	H	R
Ethiopians	19	34	24	0·47	0·53
(n = 77)	(24·7%)	(44·1%)	(31·2%)
Norwegians	18	45	33	0·42	0·58
(n = 96)	(18·7%)	(46·9%)	(34·4%)

Table 2.

Distribution of FcγRIIIa genotypes and allele frequencies in Ethiopians and Norwegians

	Genotype, n (%)			Allele freq.*
	V/V	V/F	F/F	V	F
Ethiopians	18	32	24	0·46	0·54
(n = 74)	(24·3%)	(43·3%)	(32·4%)
Norwegians	13	32	41	0·34	0·66
(n = 86)	(15·1%)	(37·2%)	(47·7%)

^*Ethiopians had a higher frequency of the V allele and lower frequency of the F allele as compared with Norwegians, P = 0·026.

Table 3.

Distribution of FcγRIIIb genotypes and allele frequencies in Ethiopians and Norwegians

	Genotype, n (%)*					Allele frequency†
	Na1/Na1	Na1/Na2	Na2/Na2	SH+	SH–	Na1	Na2
Ethiopians	5	30	42	12	65	0·26	0·74
(n = 77)	(6·5%)	(39·0%)	(54·5%)	(15·6%)	(84·4%)
Norwegians	11	41	35	1	86	0·36	0·64
(n = 87)	(12·7%)	(47·1%)	(40·2%)	(1·1%)	(98·9%)

^*There was a higher frequency of SH-positive and a lower frequency of SH-negative individuals among Ethiopians than Norwegians, P = 0·001.

^†Ethiopians had a higher frequency of the Na2 allele and lower frequency of the Na1 allele as compared with Norwegians, P = 0·046.

The genotypes for FcγRIIa, FcγRIIIa and FcγRIIIb were independent. The two populations were also compared for combinations of receptor genotypes. No significant differences were found for the combinations FcγRIIa–FcγRIIIb and FcγRIIa–FcγRIIIa. However, the genotypes for the FcγRIIIa–FcγRIIIb combination differed between Ethiopians and Norwegians (P = 0·030). The combinations V/V–Na1/Na1, V/V–Na1/Na2 and F/V–Na2/Na2 occurred more often in Ethiopian subjects. However, the number of subjects in each category of this analysis was small, so the data only indicate possible differences.

Discussion

The distributions of the FcγRIIa, FcγRIIIa, FcγRIIIb and SH-FcγRIIIb genotypes were approximately similar in the Norwegians compared with other caucasoid groups.^{5,7,8,10,13,14,15,16} However, the Norwegians had a higher FcγRIIa-131R frequency than did some other caucasoid populations, such as the Dutch and Russians.^17,18 The FcγRIIa, FcγRIIIa and FcγRIIIb genotype distributions were similar in the Ethiopian and Norwegian populations except for the FcγRIIIb-SH genotype, which was significantly increased in the Ethiopians compared with the Norwegians. The Ethiopian SH⁺ frequency was also elevated compared with four other studies of caucasoids.^7,8,10,15 Although the genotypes were similar, we found that the FcγRIIIa-158V and FcγRIIIb-Na2 allele frequencies were significantly higher among the Ethiopians compared with the Norwegians. Compared with other caucasoid populations, there seems to be a trend towards higher V and Na2 allele frequencies among Ethiopians.^{5,10,13,16,19}

We know of no other publication on FcγR polymorphisms in African populations. The distribution of the FcγRIIa genotypes was, however, similar between Ethiopians and African-Americans^10,17,20 just like that of SH-FcγRIIIb.^10,15 The frequency of FcγRIIIa-158V and FcγRIIIb-Na2 was higher in the Ethiopians than in the African-Americans.^10,15

The Ethiopians had a low FcγRIIa-131H frequency compared with Japanese, Chinese and Korean subjects but not with Asian Indians.¹², 21–23 This difference is mainly caused by the over-representation of the H/H genotype in the mongoloid populations. To our knowledge, no information is available on the FcγRIIIa frequencies in mongoloid populations at present. For FcγRIIIb, Asian Indians and Hispanics did not differ significantly from the Ethiopians, whereas Chinese, Native Americans and Malaysians had an elevated Na1 frequency.^24–26 The SH-FcγRIIIb frequency in Ethiopians was higher than in Native Americans, Koreans and a mixed group of Malays/Chinese but not in Asian Indians or Hispanics.^15,24

Our results are in agreement with the extensive review presented by Lehrnbecher et al.¹⁰ in which there are actual differences in ethnic populations that become apparent when large enough populations are compared. These observations underscore the subtle differences between populations of the same ethnic background yet different geographical location.

FcγR polymorphisms may influence the vigour of the inflammatory response and may contribute to interindividual differences in susceptibility to infectious and autoimmune diseases.⁹ Studies have shown that patients homozygous for FcγRIIIb-Na1 (and, to a lesser extent, FcγRIIa-131H) have less recurrence of adult periodontitis.²² Patients homozygous for FcγRIIa-131H have a better clinical course of meningococcal disease^18,27 and less frequent recurrence of bacterial respiratory tract infections¹⁷ than do patients heterozygous or homozygous for FcγRIIa-131R.

Of great interest is that FcγRI and FcγRII also bind C-reactive protein. It was recently found that FcγRIIa-131R binds C-reactive protein with the highest affinity.²⁸ This interaction leads to a pro-inflammatory stimulus similar to that seen with IgG complexes.²⁹ The reciprocal relationship between IgG and C-reactive protein binding avidity to FcγRIIa may affect the contribution of FcγRIIa alleles to host defence and autoimmunity.

Several lines of evidence suggest that immune complexes may influence the course of autoimmune diseases. We have found that patients with multiple sclerosis¹¹ and Guillain–Barré syndrome³⁰ homozygous for the FcγRIIIb-Na1 allele have a significantly more benign course than patients heterozygous or homozygous for the Na2 allele. A more effective processing of circulating immune complexes may be one mechanism for better clinical outcome. The FcγRIIa-131R and FcγRIIIa-158F alleles have been associated with systemic lupus erythematosus or with lupus nephritis,^5,31,32 and higher susceptibility for rheumatoid arthritis has been found with the FcγRIIIa-158F allele.³³ This may also be caused by reduced clearance of immune complexes. In addition, these low-affinity FcγR alleles increase the risk of chronic infections.⁹ Persistent triggering with bacterial antigens might therefore also influence the inflammatory activity in autoimmune diseases.

The association of variant alleles and disease susceptibility may differ between ethnic groups.¹⁰ The importance of looking at different populations with sufficient numbers is therefore critical for determining the validity of a proposed association. However, increased death rates among infants in Ethiopia compared with Norway may influence case selection. Furthermore, if the mean age of the population studied is low, as in our study of Ethiopians, one cannot be sure whether they will develop certain diseases later in life. This would also be influenced by the reduced life expectancy in the Ethiopian population as compared with the Norwegians. In addition, differences between the health-care systems in different countries may influence the diagnostic process and thereby the prevalence rates of different diseases. This might prove important when the data are linked to prevalence rates of these diseases. The uneven sex distribution in our Ethiopian population is not expected to influence the results, however, since, like other authors,^1,16 we did not find any difference in frequencies between Norwegian men and women.

In conclusion, differences in FcγR frequencies both within and between ethnic groups may indicate variation in the incidence of several infectious and autoimmune diseases. Large genetic and epidemiological studies are needed to further elucidate this important issue.