Background
Plasmodium falciparum malaria is a major cause of worldwide mortality and morbidity. Host genetic factors have been shown to influence malaria infection intensity and clinical malaria. Several candidate genes have been associated with resistance against severe malaria, whereas linkage or association analyses mapped several loci controlling mild malaria and/or parasitemia [
1]. Noticeably, chromosomes 5q31-q33 and 6p21-p23 have been linked to parasitaemia or mild malaria [
2,
3], whereas genes located within those chromosomal regions have been associated with parasitaemia, mild or severe malaria [
4‐
10]
. The
HBB locus has been shown to be a major locus based on a genome association study for severe malaria, whereas haemoglobin S (HbS) and haemoglobin C (HbC) have been associated with protection against mild and severe malaria in a large number of studies. It should be stressed that a limited number of genes have been associated with mild or severe malaria in several independent studies; these include
FCGR2A that encodes the human IgG receptor FcγRIIa.
Anti-
P. falciparum IgG antibodies are thought to play a critical role in immune protection against asexual blood stages of the parasite. Passive transfer of IgG has provided protection against the
P. falciparum blood stages in humans.
In vitro, human IgGs that recognize either infected erythrocytes or merozoites act in cooperation with monocytes to eliminate the parasite [
11]. Cytophilic IgGs that activate effector cells are, therefore, considered protective, while non-cytophilic IgGs against the same epitope may block the protective effect of the cytophilic ones. This hypothesis has been supported by several immune-epidemiological studies. High levels of the cytophilic IgG3 subclass have been associated with reduced parasitaemia, and protection against mild and severe malaria [
12,
13]. Interestingly, high levels of IgG2 could be correlated with protection in individuals carrying the H131 variant of monocytes FcγRIIa receptor, which efficiently binds to IgG2. In contrast, high levels of non-cytophilic IgG4 antibodies have been associated with susceptibility to malaria [
14]. In this context, several investigators have provided evidence of the genetic control of IgG levels. Twin studies have shown a better concordance in monozygotic twins than in dizygotic twins for IgG levels [
15]. In addition, high sib-sib correlations for IgG subclass levels have been detected [
16‐
18]. Further evidence of a genetic component has been provided by a survey conducted in several sympatric ethnic groups having different genetic backgrounds [
19]. Some candidate genes have been associated with IgG or IgG subclass levels. These include
HBB, IL4, TNF, and
FCGR2A, which have also been associated with both malaria resistance and IgG levels [
1,
20‐
25]. This suggests that those genes control the production of cytophilic IgG subclasses. More generally, genes that have been associated with malaria resistance may be associated with the level of protective IgG subclasses. The linkage or association of
HBB, IL4, IL12,
TNF, LTA, and
NCR3 with mild malaria or parasitaemia has been previously reported in a population living in Burkina Faso [
2,
3,
6,
9,
10,
26]. The objective of this study was to determine the influence of
HBB, IL4, IL12,
TNF, LTA, NCR3 and
FCGR2A polymorphisms on the IgG subclass patterns of antibodies against
P. falciparum antigens in the same population by using a family-based approach.
Results
Table
1 presents the summary of the correlation between IgG sub-class levels. There were significant positive correlations between all measured IgG subclasses (
P < 0.0001). In particular, IgG1 and IgG3 levels were highly correlated (r = 0.631). The best correlations were between IgG levels and IgG1 levels on the one hand and IgG levels and IgG3 levels on the other hand (r >0.7).
Table 1
Correlation between anti-malarial IgG subclass levels in individuals living in an endemic area in Burkina Faso
IgG1 | 1 | | | |
IgG2 | 0.373 a | 1 | | |
IgG3 | 0.631 a | 0.558 a | 1 | |
IgG4 | 0.467 a | 0.326 a | 0.46 a | 1 |
IgG | 0.71 a | 0.537 a | 0.787 a | 0.507 a |
Allele frequencies for all single-nucleotide polymorphisms (SNPs) and previously reported family-based associations with parasitaemia and mild malaria are shown in Table
2 and
Additional file 1, respectively.
Additional file 1 also shows the family-based associations of SNPs with parasitaemia and mild malaria in individuals (n = 193), for which the level of IgG subclasses has been measured. HbC,
LTA + 80,
TNF- 1031,
TNF-238,
TNF 1304, and
NCR3-412 were associated with parasitaemia or mild malaria in this sub-population (
Additional file 1). Moreover, FcγRIIa H/R131 was associated with mild malaria in individuals, for which the level of IgG2 was higher than the median (
Additional file 1).
Table 2
Family-based association analyses for anti-
Plasmodium falciparum
IgG levels
IL12Bpro (promoter) | rs17860508 | TTAGAG/GC (0.3068) | P > 0.05 |
IL12B (3'UTR) | rs3212227 | A/C (0.3275) | P > 0.05 |
IL4-590 (promoter) | rs2243250 | C/T (0.8017) | P > 0.05 |
FCGR2A H/R131 (exon 4) | rs1801274 | C/T (0.5371) | Td (0.0166f) |
HbC (exon1) | rs33930165 | A/C (0.1682) |
C
d
(0.0091
e,g
)
|
HbS (exon1) | rs334 | A/T (0.0193) | P > 0.05 |
LTA + 80 (intron 1) | rs2239704 | C/A (0.3542) | P > 0.05 |
TNF- 1031(promoter) | rs1799964 | T/C (0.1106) | P > 0.05 |
TNF-863 (promoter) | rs1800630 | C/A (0.0858) | Ad (0.0440e) |
TNF-857 (promoter) | rs1799724 | C/T (0.0228) |
T
d
(0.0064
e,g
)
|
TNF-308 (promoter) | rs1800629 | G/A (0.1063) | P > 0.05 |
TNF-238 (promoter) | rs361525 | G/A (0.0264) | P > 0.05 |
TNF 1304 (intron 3) | rs3093664 | A/G (0.075) |
A
d
(0.0083
e,g
)
|
NCR3-412 (promoter) | rs27362191 | G/C (0.2267) | P > 0.05 |
NCR3*3790 (3'UTR) | rs986475 | T/C (0.0272) | Cd (0.0307e) |
The family-based association of SNPs with the level of IgG subclasses (IgG1, IgG2, IgG3, and IgG4) and IgG against
P. falciparum crude extract was further assessed. The additive, the dominant, and the recessive models were used for each SNP. The FBAT analysis gave a number of significant results at the nominal level of 0.05, although there was no association of HbS,
IL12B pro, and
IL12B 3’ UTR,
IL4-590,
NCR3-412,
NCR3-3790,
TNF-238,
TNF-308, and
TNF- 1031 polymorphisms with the IgG responses (Tables
2 and
3). After correcting for multiple tests, there were only some SNPs that were significantly associated with anti-malarial IgG subclass levels, as shown in Tables
2 and
3. The most significant finding was a positive association between IgG2 levels and FcγRIIa H131 based on a recessive model. Conversely, IgG3 levels were negatively associated with FcγRIIa H131 based on a dominant model. HbC was positively associated with IgG levels based on an additive model. Three
TNF polymorphisms were associated with IgG responses based on an additive model:
TNF-863 and
TNF-857 were associated with IgG3 levels and IgG levels, respectively, whereas
TNF 1304 was associated with IgG3, IgG4, and IgG levels.
Table 3
Family-based association analyses for anti-
Plasmodium falciparum
IgG subclass levels
Locus | Pa | Pa | Pa | Pa |
IL12Bpro
| P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
IL12B3’UTR
| P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
IL4-590 | P > 0.05 | P > 0.05 | Cb (0.0472d) | P > 0.05 |
FCGR2A H/R131 | P > 0.05 |
T
b
(0.0003
e,f
)
|
C
b
(0.0010
d,f
)
| P > 0.05 |
HbC
| P > 0.05 | Cb (0.0139d) | P > 0.05 | P > 0.05 |
HbS
| P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
LTA + 80
| Ab (0.0305c) | Ab (0.0505c) | Ab (0.0214c) | Ab (0.0139c) |
TNF- 1031 | P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
TNF-863 | Ab (0.0379c) | P > 0.05 |
A
b
(0.0082
c,f
)
| P > 0.05 |
TNF-857 | P > 0.05 | P > 0.05 | Tb (0.0413c) | Tb (0.0249c) |
TNF-308 | P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
TNF-238 | P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
TNF 1304 | Ab (0.0495c) | P > 0.05 |
A
b
(0.0099
c,f
)
|
A
b
(0.0108
c,f
)
|
NCR3-412 | P > 0.05 | P > 0.05 | P > 0.05 | P > 0.05 |
NCR3*3790 | P > 0.05 | P > 0.05 | Cb (0.0276c) | Cb (0.0196c) |
Discussion
The aim of this study was to determine the effect of several candidate gene polymorphisms on the IgG subclass responses against
P. falciparum extract in a population living in an endemic area in Burkina Faso. Most of the candidates have been associated with parasitaemia, mild or severe malaria [
1].
There was positive correlation between all measured antibody levels. Anti-malarial IgG levels were correlated with all anti-malarial IgG subclasses. The best correlations were the correlations of anti-malarial IgG levels with anti-malarial IgG1 and IgG3 levels. A correlation between all anti-
P. falciparum IgG subclass levels was observed, the best correlation being between IgG1 and IgG3 levels. The results are similar to a study in Thailand, where anti-malarial IgG3 levels were correlated to IgG1 and IgG2 levels [
34]. The results are also coherent with a recent study conducted in Sudan, where anti-malarial IgG levels were correlated with each anti-malarial IgG subclass against malarial antigen, and where there was an age-dependent association of protective IgG2 and IgG3 subclasses [
12,
35]. Cytokines that are effective in most of IgG subclass production could partly explain the correlations between the IgG subclass levels. IL4 induces switching to ϵ, γ1, γ3, γ4 and also enhances the expression of ϵ, γ1, γ3 and γ4 germline transcript, and the secretion of corresponding proteins [
36].
IL4-590 polymorphism does not explain, nevertheless, variation in the level of IgG subclasses, suggesting that other
IL4 polymorphisms should be investigated. In addition, other cytokines, such as IL2, IL6, or IL10 are also known to enhance the IgG production
in vitro by B cells activated by polyclonal activators or plasmodial antigens [
37,
38]. Besides, IL10, which causes the switch towards IgG1 and IgG3, likely explains the high correlation between IgG1 and IgG3 levels [
39]. Interestingly, IgG2 and IgG3 levels were also strongly correlated, whereas the correlation between IgG1 and IgG2 levels was low. IFN-gamma seems to inhibit IgG1 production
in vitro, and to enhance IgG2 production [
40]. Although IFN-gamma is unlikely an IgG2 switch factor, T cells producing IFN-gamma, namely the Th1 (T helper 1) lymphocytes, have been shown to produce an IgG2 switch factor acting on CD40-activated naïve B cells [
41]. These observations suggest that cytokines produced by Th1 lymphocytes may collectively increase the production of IgG2. In all, the correlations between IgG subclass levels encourage to further assess the production of cytokines in individual living in endemic areas.
The family-based associations of candidate polymorphisms on the IgG subclass levels were further investigated. There were some significant results after applying a multiple test correction. There was no association of IgG subclass levels with
IL4 and
IL12B polymorphisms, while others reported an association of anti-malarial IgG levels with
IL4-590 [
20]. There was no association between anti-malarial IgG subclass levels and
NCR3 polymorphisms, although those may alter the production of IFN-gamma by NK cells. The association of
LTA + 80 with all anti-malarial IgG subclass levels was significant at the nominal level of 5%, but was no more significant after applying a multiple test correction; this polymorphism may, nevertheless, require further investigation to confirm or invalidate the trend.
HbS and HbC have been associated with protection against mild and severe malaria, and it has been proposed that these polymorphisms could increase the elimination of the parasite through acquired immune mechanisms [
42]. There was no association between HbS and anti-malarial IgG responses, although other studies provided evidence for a positive association [
21,
22]. These conflicting results may be explained by the low frequency of HbS in the study population. In contrast, HbC that was negatively associated with maximum parasitemia and mild malaria was positively associated with anti-malarial IgG levels in the same population, further suggesting that
HBB polymorphisms alter the anti-malarial IgG production. This result is consistent with previous reports also providing evidence for an involvement of HbC in the immune response against infected erythrocytes through enhanced IgG production [
43,
44]. This further supports the hypothesis that the protection against malaria of HbC may be partly mediated by acquired immunity against malaria, more particularly by IgG-mediated effector mechanisms. Interestingly, there was also a trend for the production of anti-malarial IgG2 levels; although the association was not significant after applying a multiple test correction, this trend may require further attention in other populations.
The FcγRIIa H131 allele has been associated with protection from severe malaria in several populations [
1]. Furthermore, it has been demonstrated to be the only receptor for IgG2, the high level of which have been associated with protection against
P. falciparum malaria in a population living in Burkina Faso [
14], and in other populations [
45,
46]. The FcγRIIa H131 allele has been found to be particularly prevalent in the Fulani, who are less affected by clinical malaria than individuals from other ethnic groups; interestingly, the Fulani also showed high levels of anti-malarial IgG2 [
23,
47]. On this basis, an effect of the FcγRIIa H131 allele on anti-malarial IgG2 levels was anticipated in this study. The data provided evidence of a significant association of FcγRIIa H131 allele with IgG2 levels based on a recessive model. This family-based result is consistent with population-based results previously published [
22,
23]. Together, these observations indicate that the FcγRIIa H131 allele at the homozygous state increases the production of anti-malarial IgG2. In the same way, there is a trend for anti-malarial IgG levels based on a recessive model, suggesting that the FcγRIIa H131 allele at the homozygous state increases the production of anti-malarial IgG, although the association was not significant after correcting for multiple tests. Thus, the FcγRIIa H131 allele may increase the phagocytosis by monocytes or macrophages, and may favour the antigen presentation by these cells to the T helper lymphocytes; this hypothesis remains to be assessed. Conversely, there was a negative association of the FcγRIIa H131 allele with anti-malarial IgG3 levels based on recessive model, suggesting that the FcγRIIa H131 allele at the homozygous state diminishes the level of anti-malarial IgG3. In the same way, Israelsson
et al. found that the FcγRIIa R131 allele carriers in the Fulani had higher anti-malarial IgG3 levels than the individuals homozygous for the FcγRIIa H131 allele. One might speculate that more IgG3 will be present in the circulation of individuals with the FcγRIIa R131 allele, because the FcγRIIa H131 allele has a higher binding than the FcγRIIa R131 allele [
48]. Nevertheless, the FcγRIIa H131 allele was associated with higher levels of anti-malarial IgG3 in another study in the Fulani [
49]. It is likely that other factors affecting the production of IgG subclass antibodies, such as the cytokines or the antigens used, could partly explain the conflicting results [
38].
In previous studies,
TNF polymorphisms have been related to severe malaria. SNPs at position −1031,-857,-308,-238 and −863 in the promoter region of
TNF gene exhibit differential associations to malaria and TNF production in different populations suggesting that individual TNF responses may be genetically determined [
7,
8,
50,
51].
TNF 1304 within intron 3 was associated with variation in mild malaria and parasitaemia [
10], and
TNF-308A and
TNF-238A allele have been associated with high anti-
P. falciparum antibodies [
24,
25]. In this study, there was no association of IgG subclass levels with
TNF-308 and
TNF-238 polymorphisms.
TNF-863A and
TNF-857 T were, however, associated with high anti-malarial IgG3 and IgG levels, respectively. Moreover, there was an association of low IgG3, IgG4, and IgG levels with
TNF 1304, which has been associated with mild malaria and parasitaemia in the same population [
10]; this further supports the role of
TNF 1304 on malaria phenotypes, although the effect of the polymorphism on the molecular function or the gene expression has not been reported. In contrast, the molecular effect of polymorphisms within the
TNF promoter has been studied. Other authors reported that the transcription factor OCT-1 binds specifically to the
TNF-863A and
TNF-857 T alleles, that the transcriptional promoter activity of a haplotype harbouring
TNF-863A and
TNF-857 T was higher than the one of other haplotypes in Japan [
50,
51]. An explanation for association of
TNF polymorphism with IgG response could be that
TNF gene is important in the development of humoral response as an autocrine B-cell growth factor [
52].
Competing interests
The authors declare that they have no conflicting interests.
Authors’ contributions
SA calculated the IgG subclass phenotypes, evaluated the correlation between IgG subclass phenotypes, and carried out most of the FBAT analyses. AA and SG participated in the FBAT analyses, and the multiple test correction. FF participated in the design of the study, and revised the results and the manuscript. PR performed the design of the study, supervised the IgG phenotype determination and the statistical analyses, and wrote the manuscript. All authors read and approved the final manuscript.