Main

HIV+ individuals have an increased risk of serious bacterial infections, and IPD accounts for a significant proportion of these cases(1, 2). Cumulative indices of IPD in HIV+ children far exceed those of the general population(3, 4), even those for children with sickle cell disease(5). Antibody-dependent opsonophagocytosis is crucial for effective host defense against Streptococcus pneumoniae(6, 7), and defects in antibody-mediated immunity have been cited to explain the increased susceptibility of HIV+ patients to pneumococcal infection(1, 2). However, the mechanism for this phenomenon, including the possible contribution of effector cell function, remains to be determined.

Since its introduction in the 1980s, passive immunotherapy with i.v. immune globulin has been successfully used in immunodeficient patients as replacement therapy to prevent bacterial infections(8). A landmark study documented that monthly i.v. immune globulin therapy in HIV+ children with CD4+ counts >200/mm3 markedly decreased their incidence of invasive S. pneumoniae infections(9). This supports the concept that antibody therapy for pneumococcal disease can be effective in a high risk population, but the nature of the protective antibodies in polyclonal preparations is unknown. Given the importance of effector cell phagocytosis for host defense against S. pneumoniae, we evaluated FcγRIIa polymorphisms in HIV+ patients who developed pneumococcal disease.

Fc receptors are transmembrane glycoproteins that bind different Ig isotypes and subclasses and mediate antibody-dependent phagocytosis. Of the three human FcγR, FcγRII (CD32) is constitutively expressed in most immune cells, and it is the only FcγR to bind IgG2 (human IgG2)(10). FcγRIIa expression is not age-dependent(11) and does not appear to be affected by HIV infection(12). A polymorphism of the FcγRII isoform “a” (FcγRIIa) has been associated with increased susceptibility of some clinical groups to infections with capsulated pathogens(11, 13). It has been hypothesized that the FcγRIIa R/R-131 genotype, which binds human IgG2 with lower affinity than the other genotypes H/H and H/R, is a risk factor for capsular infections, because IgG2 is the predominant subclass of adult antibodies against capsular organisms, including S. pneumoniae(6, 13).

The frequencies of FcγRIIa genotypes have been reported in Japanese, Chinese, Asian Indian, Caucasian(14), and African American(15) ethnic groups, but not in HIV+ cohorts or Hispanics. The goals of this work were: 1) to define the distribution of the three FcγRIIa genotypes in HIV+ children including those with IPD, and 2) to determine whether this genetic factor contributes to the susceptibility of HIV+ children to pneumococcal infections.

METHODS

FcγRIIa genotypes were determined according to published techniques by PCR amplification of the FcγRIIa locus from genomic DNA samples and hybridization of the PCR products with ASO(14).

Study population. We studied 48 HIV+ children who attend the Bronx-Lebanon Hospital Center Pediatric Infectious Disease Clinic, Bronx, New York, nine HIV+ adults with a history of IPD (bacteremia) followed at the HIV-Star Clinic at Jacobi Medical Center, Bronx, New York, and 14 HIV- healthy Hispanic adult volunteers. All children acquired HIV infection perinatally, and the majority of our pediatric HIV+ patients are of Hispanic descent. When available, the age, sex, pneumococcal vaccination, and history of IPD of every HIV+ child and adult was noted. The Hispanic population we describe, both HIV+ and HIV-, is comprised of individuals from Puerto Rico and the Dominican Republic. IPD was defined when S. pneumoniae was isolated from blood or CSF. The children and adults were identified retrospectively except for one child who developed IPD during the study. CD4+ counts were documented in all HIV+ patients. In the children with IPD the CD4+ counts reported were determined within 3 mo of the pneumococcal infection. All specimens were obtained after informed consent and according to the guidelines of the Institutional Review Boards of the Albert Einstein College of Medicine and Bronx-Lebanon Hospital Center.

Genomic DNA. High molecular weight genomic DNA was purified from whole blood samples or from expectorated buccal cells with the Rapid Prep™ DNA isolation kit (Pharmacia Biotech, Piscataway, NJ) DNA from the cell lines K562 and U937, previously determined to be R/H-131 and R/R-131, respectively(14), were used to standardize the assay. Genomic DNA samples from 42 HIV- normal Hispanic volunteers were kindly provided by Dr. Anne Davidson (Albert Einstein College of Medicine, Bronx, New York).

PCR amplification and dot blot analysis. The human FcγRIIa locus was amplified from genomic DNA by PCR using published primers and methods(14). The oligonucleotides specific for ASO-R131 and ASO-H131 differ by only one nucleotide: 5′-A TTC TCC C(G/A)T TTG GAT C. ASO hybridization conditions were such that ASO-R131 would bind only to PCR products with at least one R131 allele, and ASO-H131 would bind only to PCR products with at least one H131 allele(14). Oligonucleotides for PCR amplification and ASO hybridization were synthesized by the Oligonucleotide Synthesis Facility of the Cancer Center at the Albert Einstein College of Medicine.

ELISA for IgG subclasses. The IgG1 and IgG2 subclasses of serum pneumococcal polysaccharide binding antibodies were determined by ELISA. Polystyrene ELISA plates (Corning Costar, Oneonta, NY) were coated with 10 μg/mL solutions of each of the four following pneumococcal capsular polysaccharides: 4, 14, 19F, and 23F (American Type Culture Collection, Rockville, MD). The plates were blocked with PBS/1% BSA, washed, and incubated with sera from seven HIV+ children with IPD. All sera were heat-inactivated, diluted 1:10, and absorbed with purified pneumococcal cell wall polysaccharide to eliminate cell wall polysaccharide binding antibodies. The sera were applied to the plates in quadruplicate, serially titered, and incubated at 37 °C for 1 h. After washing, duplicate rows were reincubated with each of the two murine monoclonal subclass-specific antibodies against human IgG1 (HP6069, Calbiochem, La Jolla, CA) and IgG2 (HP6002, Calbiochem, La Jolla, CA). The plates were developed with p- nitrophenyl phosphate (Sigma Chemical Co., St. Louis, MO) in diethanolamine buffer (pH 9.8), and ODs were determined using an ELISA reader Ceres 900 (Bio-tek Instruments, Winooski, VT) at 405 nm. Each ELISA plate included the same high reactivity reference serum kindly provided by Dr. María C. Rodriguez-Barradas (Department of Veterans Affairs Medical center, Houston, TX).

Statistical analysis. Distributions of FcγRIIa-H131/R131 alleles were compared using χ2 and Fisher's exact tests. Continuous variables, including CD4 counts, which had a normal distribution in our cohort, were compared by t test. Wilcoxon matched pair test was used to compare the IgG1 and IgG2 (ODs) reactivity of individual serum with each serotype. p values <0.05 were significant.

RESULTS

FcγRIIa polymorphism. The FcγRIIa allelic frequencies of 56 HIV- Hispanic adults were 11 H/H (19.6%), 26 H/R(46.4%), and 19 R/R (34%). In 40 HIV+ children without IPD, the distribution was 7 H/H (17.5%), 23 H/R (57.5%), and 10 R/R (25%). For 8 HIV+ children and nine HIV+ adults with IPD, the genotypes were 1 H/H, 5 H/R, and 2 R/R, and 2 H/H, 4 H/R, and 3 R/R, respectively(Table 1). There were no significant statistical differences between HIV- Hispanics, HIV+ children with and without IPD, and HIV+ adults with IPD.

Table 1 Clinical and laboratory characteristics of HIV+ children with IPD
Full size table

ELISA of IgG subclasses. The IgG1 and IgG2 subclasses of serum pneumococcal polysaccharide-binding antibodies were determined in seven HIV+ children with IPD (one was excluded because of monthly i.v. immune globulin therapy). For each of the four serotypes tested, antibodies of the IgG1 subclass predominated. Little to no polysaccharide reactive IgG2 was detected at a serum dilution of 1:10(Table 2).

Table 2 ELISA units* (mean ± SD) of IgG subclasses, IgG1 and IgG2, determined by ELISA in the HIV+children with IPD
Full size table

Clinical features of children with IPD. There were eight HIV+ children with IPD who ranged in age from 8 to 78 mo (mean 32.7 mo). These children had 11 episodes of IPD, and the recurrence rate was 27%(3/11). All IPD episodes were bacteremic. The primary focus of six episodes was pneumonia, one was meningitis, and one was endocarditis. No source was identified for three episodes (Table 1). Serotypes of the isolates were not available. None of the children had sickle cell disease. The children with IPD did not differ with respect to age, male to female ratio, or CD4+ count when compared with the other 40 HIV+ children without IPD (data not shown). The HIV+ adults had a male to female ratio of 1.2:1, ages of 32-53 y (mean 41.25 y), and CD4+ counts of 2-567 cells/mm3 (mean 199.12 cells/mm3).

DISCUSSION

In this work we report that the distribution of FcγRIIa genotypes in HIV+ Hispanic children is statistically comparable to that of HIV- Hispanics. This suggests that FcγRIIa polymorphism is not associated with HIV infection itself, although we did not evaluate rapid progressors. There was no increased incidence of FcγRIIa R/R-131 among the HIV+ children with IPD, and their FcγRIIa genotype distribution was similar to that of HIV+ children without IPD. Our sample size was sufficient to detect major differences in FcγRIIa genotypes between the children with IPD and the control group. However, we cannot rule out statistically significant differences that might have been found with a larger sample. The HIV+ children in our cohort that developed IPD were asymptomatic or mildly symptomatic (CD4+ count mean 802 cells/mm3, range 169-1532 cells/mm3) at the time of pneumococcal infection. They all had uncomplicated clinical courses and were successfully treated.

Several authors have evaluated the association of FcγRIIa genotypes with susceptibility to infections with capsulated pathogens(11, 13, 15, 16). Similar to our results, a recent study showed that the FcγRIIa genotype distribution was the same in a cohort of African American children with sickle cell disease and IPD and an ethnically matched control group without either sickle cell disease or IPD(15). Conversely, FcγRIIa-R/R131 has been found to be a risk factor for recurrent bacterial respiratory infections(11) and fulminant meningococcal shock(13) in cohorts of Caucasian children who were not immunodeficient. Effective host defense against S. pneumoniae is dependent upon both opsonic antibodies and complement(6, 17). In adult pneumococcal polysaccharide vaccinees, phagocytosis of S. pneumoniae is correlated with the titer of capsular IgG2 in heated, but not in complement-containing, sera(17). This suggests that the role of FcγRIIa may be most critical for complement-independent IgG-mediated phagocytosis. Therefore, the lack of correlation that we and others(15) have found between FcγRIIa polymorphism and IPD could be a function of the fact that the importance of FcγRIIa is paramount in hypocomplementemic conditions(16). In this regard, FcγRIIa polymorphism could impact upon the clinical course of pneumococcal infection after capsular polysaccharide-induced complement consumption occurs(6, 18).

In children younger than 2 y there is a physiologic delay in the production of IgG2, and the predominant subclass of polysaccharide antibodies is IgG1(19, 20). Hence, the importance of FcγRIIa for phagocytosis of S. pneumoniae in younger children remains to be determined. In HIV+ children, IgG1 and IgG3 levels are increased, but the physiologic IgG2 deficiency of infancy is prolonged, and IgG2 levels remain decreased(21, 22). Our serologic analysis indicated that HIV+ children with IPD have little to no IgG2 binding to four clinically relevant pneumococcal polysaccharides (Table 2). In one study, children with recurrent pneumococcal otitis manifested predominantly IgG1 pneumococcal polysaccharide antibodies(23), and in another, children with normal serum Ig levels and recurrent bacterial respiratory tract infections had IgG1 levels comparable to adults after pneumococcal polysaccharide vaccination, but lower IgG2(19). It has been suggested that FcγRIIa genotypes may play a role in regulating human IgG subclass production/turnover, because those with the high affinity genotype had lower IgG2 levels(24). We could not evaluate such a trend in our cohort because of the small sample size. To determine whether FcγRIIa polymorphism contributes to susceptibility to pneumococcal infections in high risk pediatric groups, it will be more important to analyze FcγRIIa genotype expression according to the level of serotype-specific IgG2, not total IgG2. Antigen-specific IgG2 has not been evaluated in prior reports of the incidence of infections in HIV+ children with low serum IgG2(22). In the few studies that have measured serotype specific pneumococcal polysaccharide IgG2, low levels were reported in individuals with severely impaired antibody responses to pneumococcal vaccination(25, 26).

Increasing rates of pneumococcal antibiotic resistance(27) and poor pneumococcal polysaccharide vaccine responses in HIV+ individuals(28, 29) are critical problems for the management of pneumococcal infections. T cell-derived factors or complement are not required for FcγRIIa expression, which is constitutive(10). Therefore, in those with defects in expression of inducible phagocytic receptors based on cytokine dysregulation (e.g. HIV infection), antibody-based therapies which provide IgG subclasses that can bind FcγRIIa might be most appropriate for prophylaxis or therapy. The distribution of FcγRIIa polymorphisms in our cohort supports the proposal that antibody therapy could be given to many high risk individuals. The prospect of passive immunotherapy with specific anticapsular antibodies might be a promising alternative for the treatment and/or prevention of IPD in HIV+ children and other immunodeficient groups.