Association of Fcγ receptor IIB polymorphism with cryptococcal meningitis in HIV-uninfected Chinese patients.

Background As important regulators of the immune system, the human Fcγ receptors (FcγRs) have been demonstrated to play important roles in the pathogenesis of various infectious diseases. The aim of the present study was to identify the association between FCGR polymorphisms and cryptococcal meningitis. Methodology/Principal Findings In this case control genetic association study, we genotyped four functional polymorphisms in low-affinity FcγRs, including FCGR2A 131H/R, FCGR3A 158F/V, FCGR3B NA1/NA2, and FCGR2B 232I/T, in 117 patients with cryptococcal meningitis and 190 healthy controls by multiplex SNaPshot technology. Among the 117 patients with cryptococcal meningitis, 59 had predisposing factors. In patients with cryptococcal meningitis, the FCGR2B 232I/I genotype was over-presented (OR = 1.652, 95% CI [1.02–2.67]; P = 0.039) and the FCGR2B 232I/T genotype was under-presented (OR = 0.542, 95% CI [0.33–0.90]; P = 0.016) in comparison with control group. In cryptococcal meningitis patients without predisposing factors, FCGR2B 232I/I genotype was also more frequently detected (OR = 1.958, 95% CI [1.05–3.66]; P = 0.033), and the FCGR2B 232I/T genotype was also less frequently detected (OR = 0.467, 95% CI [0.24–0.91]; P = 0.023) than in controls. No significant difference was found among FCGR2A 131H/R, FCGR3A 158F/V, and FCGR3B NA1/NA2 genotype frequencies between patients and controls. Conclusion/Significance We found for the first time associations between cryptococcal meningitis and FCGR2B 232I/T genotypes, which suggested that FcγRIIB might play an important role in the central nervous system infection by Cryptococcus in HIV-uninfected individuals.


Introduction
Cryptococcal meningitis is the most common opportunistic fungal infection of the central nervous system in AIDS patients. Among HIV-uninfected patients, several predisposing factors for cryptococcal meningitis such as corticosteroid medication, solid organ transplantation and malignancy, etc, have been indentified. Yet cryptococcal infections in apparently healthy individuals are also increasingly being reported, especially from Asian data [1][2][3]. Our previous study has demonstrated an association between mannose-binding lectin (MBL) genetic deficiency and cryptococcal meningitis in HIV-uninfected patients [4]. However, MBL deficiency was present in only 21% of the cases, and for the remaining 79% of patients the underlying mechanism for susceptibility remained unclear.
Fc gamma receptors (FccRs) mediate a variety of immune responses after binding to IgG-opsonized pathogens or immune complexes, and therefore act as immune regulators in both autoimmune and infectious diseases [5][6][7][8][9]. According to their affinity to IgG, FccRs are categorized into high-affinity and lowaffinity receptors. FccRI is the only known high-affinity receptor.
FCGR polymorphisms had been associated with the susceptibility and severity of various infections. FCGR2A 131R/R had been reported to attribute to the susceptibility of meningococcal infection, community-acquired pneumonia (CAP) caused by Haemophilus influenza, and the development of severe malaria [10][11][12]. FCGR2A 131H/H was reported to contribute to higher risk of bacteremia in pneumococcal CAP patients [13]. Another study showed that HIV-infected patients with FCGR2A 131R/R genotype progressed to a low CD4 + cell count at a faster rate, but conversely in individuals carried FCGR2A 131H/H there was an increased risk of Pneumocystis jiroveci pneumonia [14]. FCGR3A 158F/V gene polymorphism was not associated with progression of HIV infection, but predicted the risk of Kaposi's sarcoma [14]. A study on infections during induction chemotherapy found that FCGR2A 131H/H was associated with a decreased risk of pneumonia, FCGR3B NA1/NA1 associated with infections, and FCGR3A polymorphisms not associated with infections [15]. Sadki et al. investigated the influence of FCGR3A 158V/F and FCGR2A 131H/R polymorphisms on susceptibility to pulmonary tuberculosis in the Moroccan population but no association was found [16]. A study in East Africa found that the FCGR2B 232T/T genotype provided protectiveness for children against severe malaria [17].
A previous study by Meletiadis et al. investigated FCGR polymorphisms in patients with cryptococcosis, and found that FCGR2A 131R/R and FCGR3A 158V/V were over-presented, and FCGR3B NA2/NA2 was under-presented in patients with cryptococcosis [18]. The purpose of this study was to investigate FCGR polymorphisms in our series of patients to further verify the association between FCGR and cryptococcal meningitis.

Demographic Characteristics
A total of 117 HIV-uninfected patients with cryptococcal meningitis were included. Subjects from both the patient and control groups were of Chinese Han ethnicity. Clinical information and predisposing factors of the patients are summarized in Table 1. Of the 190 healthy control subjects, 111 were male (58.4%). The median age of the control subjects was 44 years (range, 12-79 years).

Genotype Distribution
Two samples failed genotyping of FCGR3A and 2 samples failed in genotyping of FCGR2B. Allele distributions of the tested FCGR genes in the control group were in Hardy-Weinberg equilibrium. The frequencies of FCGR2A, FCGR3A, FCGR3B and FCGR2B genotypes were shown in Table 2. An association was found between FCGR2B 232I/T genotypes and cryptococcal meningitis based on dominant and over-dominant model. The FCGR2B 232I/I genotype was over-presented (OR = 1.652, 95% CI [1.02-2.67]; P = 0.039) and the FCGR2B 232I/T genotype was underpresented (OR = 0.542, 95% CI [0.33-0.90]; P = 0.016) in patients with cryptococcal meningitis in comparison with controls. No significant difference was found in the distribution of FCGR2A 131H/R, FCGR3A 158 F/V and FCGR3B NA1/NA2 genotypes.
We further compared the genotype distribution of FCGR2A, FCGR3A, FCGR3B and FCGR2B between the 58 patients without predisposing condition and controls. Similar to results from the overall patient group, associations were also found between FCGR2B 232I/T genotypes and cryptococcal meningitis based on dominant and over-dominant model. Specifically, FCGR2B 232I/I genotype was also more frequently detected (OR = 1.958, 95% CI [1.05-3.66]; P = 0.033), and FCGR2B 232I/T genotype was also less frequently detected (OR = 0.467, 95% CI [0.24-0.91]; P = 0.023) in patients without predisposing factor than in controls. For the genotype distribution of other polymorphisms (FCGR2A 131H/R, FCGR3A 158 F/V and FCGR3B NA1/NA2), there was also no significant difference between patients and controls.
The four polymorphisms of low-affinity receptors genotyped in our study have each been demonstrated to affect functions of their encoded receptors. In FCGR2A, the G.A SNP at amino acid position 131 results in a histidine (H) to arginine (R) change (FCGR2A 131H/R), resulting in reduced affinity of the correspondent receptor to IgG2 [33,34]. The T.G SNP at position 158 of FCGR3A causes a phenylalanine (F) to valine (V) substitution (FCGR23A 131F/V) and FCGR3A 158V/V encoded receptors show higher affinity to IgG1 and IgG3 [35,36]. In the FCGR3B gene, five nucleotides (141,147,227,277 and 349) are combined to form two main haplotypes termed FCGR3B NA1 and FCGR3B NA2, and receptor encoded by FCG3B NA1 haplotype binds to IgG1 and IgG3 more easily [37]. Finally, FCGR2B 232I/T causes an isoleucine (I) to threonine (T) substitution in the transmembrane domain [22,38] and receptors encoded by FCGR2B 232T are unable to interact with activating receptors [39].
Although FCGR polymorphisms have been demonstrated to be associated with susceptibility and severity of numerous infections, there has only been one previous genetic association study on the relationship of FCGR genotypes and cryptococcosis [18]. Meletiadis and colleagues genotyped FCGR2A 131H/R, FCGR3A 158F/V and FCGR3B NA1/NA2 in 103 cryptococcosis patients and 395 healthy controls. They found that in patients with cryptococcosis FCGR2A 131R/R and FCGR3A 158V/V were over-presented (Pvalues were 0.04 and 0.04), while FCGR3B NA2/NA2 was underpresented (P-value was 0.04).
In our study, we found for the first time that cryptococcal meningitis was associated with the FCGR2B 232I/T genotypes, which was not reported in Metediatis' study. As the only known inhibitory FccR, FccRIIB has an immunoreceptor tyrosine-based inhibitory motif (ITIM) in its cytoplasmic domain, and thus it plays an important role in regulating the immune system [40]. FCGR2B 232I/T is located in the transmembrane domain, and the FccRIIB receptors encoded by FCGR2B 232T are unable to interact with activating receptors and exert inhibitory activity [38]. Published data have suggested the mutation genotype FCGR2B 232T/T to be a susceptible genotype for systemic lupus erythematosus [17,22,32], and this genotype also provided protective effect for severe malaria in East African children [17]. The role of FccRIIB in cryptococcal infection is still not very clear. Like the activatory FccRs, FccRIIB can also recognize the major component of the capsule of C. neoformans, glucuronoxylomannan (GXM). In a previous study by Monari et al., the immunosuppressive effect of GXM was demonstrated to be dependent on FcRcIIB, based on the evidences that FccRIIB blockade inhibits GXM-induced IL-10 production and induces TNF-a secretion, and that the addition of monoclonal antibody to GXM reverses GXM-induced immunosuppression by shifting recognition from FccRIIB to FccRIIA [41]. Another study subsequently demonstrated that GXM triggered NO-induced macrophage apoptosis, which was dependent on FccRII [42]. These data support that FccRIIB plays a critical role in the pathogenesis of cryptococcal infection. In our study, it is the FCGR2B 232I/T heterozygote instead of the minor homozygote 232T/T that is under-presented in patient group. One study on children with idiopathic thrombocytopenia (ITP) also showed a similar pattern, that the FCGR2B 232I/T was less frequently detected in acute ITP in comparison with chronic ITP [27]. The reason for the heterozygotes 232I/T rather than 232T/T under-presenting in our patients and those acute ITP children has not been clarified.
Unlike results from Meletiadis' study, no association among FCGR2A 131H/R, FCGR3A 158F/V, FCGR3B NA2/NA2 and cryptococcal meningitis was found in our study. The cause for discrepant results may be multifactorial. One was the ethnic differences between the two studies. Subjects in our study were of Chinese Han ethnicity, while the majority of subjects in Meletiadis' study were Caucasians (60%). As a matter of fact, the FCGR3A 158V allele was significantly increased only in patients who were Caucasian in Meletiadis' study. Secondly, all the cases in our study were diagnosed with cryptococcal meningitis, while some patients from Meletiadis' study were cryptococcosis without central nervous system involvement. Furthermore, both studies had relatively small sample sizes, which could be underpowered to generate positive results.
In conclusion, our study suggested that FccRIIB genetic polymorphism may contribute to the susceptibility of cryptococcal meningitis. The overall association is relatively weak, which warrants validation in larger population.

Ethics Statement
This study was reviewed and approved by the Ethic Committee/Institutional Review Board (HIRB) of Huashan Hospital, Fudan University, and informed written consent was obtained from each participant.

Subjects
A total of 200 volunteers and 117 unrelated patients with proven or probably diagnosed cryptococcal meningitis who were referred to Huashan Hospital, Fudan University, China, from 2001 through 2011 were recruited for the present study. Patients who met at least one of the following criteria were considered as proven cryptococcal meningitis: (1) Isolation of C. neoformans from cerebrospinal fluid (CSF) by culture or positive India ink smear, and (2) compatible histopathological findings, which are 5-10 mm encapsulated yeasts observed in brain tissue. Patients who had no microbiological or pathological documentation but present with positive cryptococcal antigen titer ($1:10) in CSF and met at least one of the following criteria were regarded as probable cryptococcal meningitis: (1) abnormal laboratory tests or an increased open pressure ($200 mmH 2 O) of CSF, (2) abnormalities of cranial imaging (Computerized Tomography or Magnetic Resonance Imaging) which could not be explained by other factors, and (3) comorbidities that compromise the host immune system. Cryptococcal antigen was determined using diluted CSF with the Latex-Cryptococcus antigen detection system (Immuno-Mycologics). Patients and volunteers were assessed for predisposing factors as follow, immunocompromising diseases (liver cirrhosis, chronic kidney diseases, autoimmune diseases, malignancies, solid organ transplantation) [2,3,43], and corticosteroid (at prednisone equivalent dose of .0.3 mg/kg/day of for .3 weeks) or immunosuppressive medications (within 90 days before onset of cryptococcal meningitis) [44], and idiopathic CD4 + T lymphocytopenia (unexplained CD4 + T lymphocytopenia with CD4 + T lymphocyte count ,300 cells/mm 3 ) [45]. Diabetes mellitus was also included, although this common condition is a controversial predisposing factor [3,46]. Patients without any of the above mentioned predisposing factors were considered as apparently healthy hosts. Ten volunteers were excluded because of disclosed predisposing conditions, and the remaining 190 healthy volunteers were included in the control group.
Venous blood was obtained by venepuncture from each subject. Genomic DNA was extracted using the QIAamp DNA kit (Qiagen, Hilden, Germany) according to manufacturer's instructions. Genotyping of 8 SNPs in FCGRs (Table 3) was performed by multiplex SNaPshot technology using an ABI fluorescence-based assay discrimination method (Applied Biosystems, Foster city, CA, USA), which has been described in detail in previous studies [47,48]. The multiplex SNaPshot detection of single-base extended probe primers was based on fluorescence and extended length detected by capillary electrophoresis on ABI3130XL Sequencer (Applied Biosystems, Foster City, CA, USA).
The PCR reactions were performed with 1 mL of DNA and 1 mL multiple PCR primers (the concentration was 1 mM) in a total volume of 20 mL containing 16 HotStarTaq buffer, 2.0 mM Mg 2+ , 0.3 mM dNTP, and 1 U HotStarTaq polymerase (Qiagen, Hilden, Germany). The cycling conditions for FCGR2A and FCGR3A were 95uC for 2 min, 35 cycles using 96uC for 20 s, 62uC for 2 min, and 72uC for 3 min, then 72uC for 10 min, and finally kept at 4uC. The cycling conditions for FCGR2B and FCGR3B were 95uC for 2 min, 7 cycles using 96uC for 20 s, 55uC for 2 min, and 72uC for 3 min, then 72uC for 10 min, and finally kept at 4uC. PCR products were then purified (add 1U SAP enzyme to 10 mL PCR products, incubate at 37uC for 1 hour, then, inactivate at 75uC for 15 min).
The extension reaction to identify single nucleotide polymorphisms in the PCR products was performed in a total volume of 10 mL containing 2 mL purified PCR product, 1 mL primer (the concentration was 0.8 mM), 5 mL SNaPshot Multiplex Kit (Applied Biosystems, Foster City, CA, USA), and 2 mL ultrapure water. The cycling conditions for extension were 96uC for 1 min, 28 cycles of 96uC for 10 s, 52uC for 5 s, and 60uC for 30 s, and kept at 4uC. Then each extended product was added to 1 U shrimp alkaline phosphatase, incubated at 37uC for 1 hour, and the enzyme inactivated at 75uC for 15 min. Then, 0.5 mL was added to 0.5 mL Liz120 SIZE STANDARD (Applied Biosystems, Foster City, CA, USA), 9 mL Hi-Di (Applied Biosystems, Foster City, CA, USA), and sequenced by ABI3130XL Sequencer (Applied Biosystems, Foster City, CA, USA). Finally, the primary data was analyzed by GeneMapper 4.0 (Applied Biosystems, Foster City, CA, USA). Genotypes were determined by the type of nucleotide presented at SNP site, which was visualized by one or two different color peaks on the figures.
For quality control, a random sample of 5% of the cases and controls was genotyped twice by different researchers, with a reproducibility of 100%. The minor allele counts were compared with database (http://www.ncbi.nlm.nih.gov/projects/SNP), and the data were matched well. Genotyping was performed blind to group status.

Statistical Analysis
Dominant, over-dominant, recessive and allelic models were applied for the analysis of genotype distribution. Hardy-Weinberg equilibrium, differences in gene polymorphism distributions between patients and controls were analyzed with 262 x 2 tests or Fisher's exact test where appropriate. P-values, odds ratios (ORs) and 95% confidence intervals (CIs) were obtained by SPSS 16.0 for Windows (SPSS, Inc, Chicago, IL). P-values ,0.05 were considered statistically significant.