Pharmacogenetics of hypersensitivity to abacavir: from PGx hypothesis to confirmation to clinical utility.

The hypersensitivity (HSR) to abacavir (ABC) pharmacogenetics (PGx) program represents the progression from an exploratory discovery to a validated biomarker. Within the program, two retrospective PGx studies were conducted to identify HIV-1 patients at increased risk for ABC HSR, a treatment-limiting and potentially life-threatening adverse event. A strong statistical association between the major histocompatibility complex allele, HLA-B*5701, and clinically diagnosed ABC HSR was identified but varied between racial populations. Subsequently, ABC skin patch testing was introduced as a research tool to supplement clinical case ascertainment. In a randomized, prospective study evaluating the clinical utility of HLA-B*5701 screening, avoidance of ABC in HLA-B*5701-positive patients significantly reduced clinically diagnosed ABC HSR and eliminated patch test-positive ABC HSR. Finally, a retrospective PGx study supports the generalizability of the association across races. Prospective HLA-B*5701 screening should greatly reduce the incidence of ABC HSR by identifying patients at high risk for ABC HSR before they are treated.

Oral cefixime is similar to continued intravenous antibiotics in the empirical treatment of febrile neutropenic children with cancer.

Empiric oral antibiotic therapy for febrile neutropenic cancer patients has been suggested as a means to decrease hospitalization, but the safety of this approach has not been adequately studied in children. We compared continued iv antibiotic therapy with switching treatment to orally administered cefixime in a group of selected febrile neutropenic children for whom blood cultures were sterile after 48 h of incubation. Two hundred episodes of febrile neutropenia were studied (156 patients), and 100 episodes were randomized to receive each treatment. Failure to respond to therapy was defined by documented or suspected bacterial infection, recurrent fever, or discontinuation of assigned therapy for any reason before neutropenia resolved. Rates of treatment failure were similar in the oral cefixime group (28%) and in the iv antibiotic group (27%; P=1.0). Results support the safety of oral cefixime therapy for low-risk febrile neutropenic children, a therapeutic approach that would facilitate earlier outpatient management and decrease the costs of treatment.

Inter- and intra-patient sequence diversity among parainfluenza virus-type 1 nucleoprotein genes.

Parainfluenza viruses (PIV) have been categorized into four discrete types (types 1-4), based on antigenic similarities. Here is described an evaluation of nucleoprotein (NP) sequence variability among nine patients infected with the type 1 virus. The examination of short segments of the NP sequence was sufficient to define significant variability both within and between patient samples. These data, in conjunction with previous studies of hemagglutinin-neuraminidase and fusion protein sequences from PIV-infected patient populations suggest a lack of absolute stability among isolates within each virus type. Potentially, antigenic variability exists to the extent that an immune response elicited toward one isolate may not be fully protective against another of the same type. Thus, sequence variability could contribute to natural re-infections with PIV, as well as to previous vaccine failures. Results highlight the importance of analyzing viruses that break through vaccine-induced immunity, in order to measure the influence of virus diversity on PIV vaccine outcome.

Sequence and analysis of the rpoB gene of Mycobacterium smegmatis.

The rpoB gene encodes the beta subunit of the DNA-dependent RNA polymerase of bacteria. Mutations in defined areas result in resistance to rifampin. Mycobacterium smegmatis is naturally resistant to rifampin, but analysis of the rpoB gene revealed no identifiable rifampin resistance mutations. Another mechanism of resistance may be present.

Functional capacities of clonal antibodies to Haemophilus influenzae type b polysaccharide.

Haemophilus influenzae type b (Hib) is an important pathogen for young children, and children can be protected with antibodies (Abs) to Hib polysaccharide (PS) capsule, a linear polymer of ribosyl ribitol phosphate. The structure of anti-Hib-PS Abs has been well characterized at the molecular level; about two-thirds of anti-Hib-PS Abs use a V kappa gene named A2, and the remaining anti-Hib-PS Abs use one of many other VL genes. In order to understand the structural basis for the variability in the function of these Abs, we prepared 18 clonally pure Abs from adults and studied their affinity, avidity, bactericidal potency in vitro, and ability to reduce bacteremia in newborn rats. Affinities and avidities were determined as the inverse of the concentrations of short (3 repeating units) and long (20 repeating units) ligands which could bind 50% of anti-Hib-PS Ab in solution, respectively. No significant correlations between the protection of newborn rats and affinity (r = 0.02) or avidity (r = 0.16) were observed. The amount of Ab required to kill 50% of bacteria in vitro decreased with avidity (r = -0.32), as expected. However, Abs with high affinity were unexpectedly found to have less bactericidal activity (r = 0.38). This suggests that avidity may be a better predictor of Ab function than affinity. Affinity and avidity results were negatively correlated (r = 0.76, P = 0.0022), and Abs that had A2 V kappa gene products had higher avidity (P < 0.05) and lower affinity (P = 0.06) than Abs that had other VL genes. A possible explanation of these observations is that the epitope for Abs with the A2 gene is within the Hib-PS chain itself, whereas the epitope for Abs with a non-A2 gene is the terminus of Hib-PS.

Molecular evolution of the F glycoprotein of human parainfluenza virus type 1.

Human parainfluenza virus type 1 (hPIV1) is a major cause of upper and lower respiratory tract infections among children. Immunity is mediated at least in part by antibody to the fusion (F) surface glycoprotein. Thus, genetic variation in the F gene could influence host range, virulence, and immunity. To examine the genetic diversity among hPIV1 isolates, the F genes of hPIV1 isolates from a single geographic location were sequenced and compared with the F gene of a strain isolated in 1957. Genetic variation was 2.2%-3.4%, averaging 0.8 amino acid changes per year. Changes were progressive over time, and virus evolution was dominated by a single lineage. Three of 7 isolates tested did not induce syncytium formation in tissue culture. This phenotype could not be ascribed to a single unique mutation in the F gene, but these 3 isolates had mutations in the transmembrane region of the HN gene. It is unlikely that the limited genetic evolution of the F gene will be an obstacle to vaccine development.

Comparative virulence of Staphylococcus epidermidis isolates in a murine catheter model.

Among coagulase-negative staphylococci, Staphylococcus epidermidis is the species most commonly implicated in catheter-related infections. Whether some staphylococcal organisms are inherently more virulent than others, or whether their ability to infect relates more to the sheer numbers of organisms at the catheter site, remains unclear. We therefore compared eight S. epidermidis isolates and two other coagulase-negative staphylococci using a murine model that allowed us to quantify catheter colonization and abscess formation in the same animal. The organisms were isolated from different clinically relevant settings and were classified according to their slime phenotype. The ability to evoke abscesses or colonize catheters in half of the animals (ID50) was assessed. ID50 inoculum titers (log10 data +/- SD) ranged widely, from 8.5 +/- 0.3 to 10.2 +/- 0.2 for abscess formation (p < 0.005) and from 7.5 +/- 0.5 to 10.3 +/- 1.0 for catheter colonization (p < 0.005). ID50 values by statistical criteria suggested variability among organisms in the ability to induce abscess formation. High slime production correlated with both parameters, but not with the clinical source of the isolate. Our findings demonstrate impressive heterogeneity in the ability of a representative group of S. epidermidis isolates to colonize catheters and to evoke abscess formation and implicate slime productivity as a major virulence factor. The murine model used permitted simultaneous analysis of multiple factors involved in pathogenesis and should be useful in establishing the basis of S. epidermidis pathogenicity.

Human parainfluenza virus type 1 evolution combines cocirculation of strains and development of geographically restricted lineages.

The hemagglutinin neuraminidase (HN) glycoprotein of human parainfluenza virus type 1 (HPIV-1) mediates attachment to the host cell and is the target of protective antibody. Since the efficacy of a potential vaccine depends on antigenic constancy, the antigenic and genetic stability of the HPIV-1 HN glycoprotein was examined for 13 isolates obtained between 1981 and 1989. Antigenic analysis with a panel of 11 monoclonal antibodies demonstrated a single change among 3 isolates from 1989 that distinguished them from all other isolates. The HN genes from all 13 isolates and 13 previously published HN gene sequences shared > 95% homology. Evolutionary analysis demonstrated cocirculation of strains, without a dominant lineage. The 1989 isolates and the previously proposed subtype A isolates occupied distinct evolutionary branches, indicating geographically limited evolution. The slow rate of evolution and HN homogeneity may allow development of a single vaccine formulation for the prevention of disease.

Monoclonal antibodies against a 97-kilodalton antigen from Aspergillus flavus.

We prepared a panel of five monoclonal antibodies (MAbs) directed against Aspergillus flavus that all reacted against one 97-kDa antigen by western blot (immunoblot). Flow cytometry demonstrated that these antibodies bound (in increasing degrees) to all morphologic stages of A. flavus growth: conidia, swollen conidia, and hyphae. Cross-reactivity among species was examined by enzyme-linked immunosorbent assay of fungal culture filtrates. Four MAbs reacted with 10 of 11 A. flavus isolates, and the fifth one reacted with 9 of them. One MAb also reacted with A. fumigatus, two reacted with A. niger, A. wentii, and A. nidulans, and all five reacted with A. ochraceus. None reacted with A. terreus, A. glaucus, A. versicolor, or a Penicillium species. Each MAb bound to A. flavus hyphae in formalin-fixed paraffin sections of a muscle biopsy from a confirmed human case of invasive aspergillosis. In summary, these MAbs identified a 97-kDa antigen found on A. flavus that is both surface bound and an exoantigen. Either the same or a cross-reacting antigen is present in A. fumigatus and other Aspergillus species.

Outer membrane protein binding sites of complement component 3 during opsonization of Haemophilus influenzae.

Complement component 3 (C3) binding to Haemophilus influenzae type b (Hib) is an important step in host defense against invasive disease, but the details of this process remain poorly understood. We have shown that the P1 and P2 outer membrane proteins (OMPs) serve as binding sites for C3 on serum-opsonized Hib. Whole-cell lysates of opsonized Hib were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and the resolved proteins were transferred to nitrocellulose. Immunoblot analysis with monoclonal antibodies (MAbs) to the 49-kDa P1 and 39-kDa P2 OMPs demonstrated high-molecular-weight bands that were not present when the bacteria were opsonized with heat-inactivated or methylamine-treated serum. Immunoblot analysis with MAbs to the 98- or 16-kDa (P6) OMPs did not reveal additional bands. An unencapsulated Hib mutant still lacked C3 bound to the 98-kDa or P6 OMP, indicating that the absence of C3 binding to these proteins was not the result of epitope masking by the capsule. Studies with MAbs to C3 fragments confirmed that the anti-P1- and anti-P2-reactive bands were C3 fragments bound to these OMPs. The molecular weights of proteins reactive to anti-OMP and anti-C3 antibodies indicated that multiple C3 fragments may be bound to P1 or that C3 may be bound to P2 multimers. Finally, the presence of other anti-C3-reactive proteins indicated that several other proteins serve as C3 targets during the opsonization of Hib.

Complement binding to Aspergillus conidia correlates with pathogenicity.

Complement has significant effects on the phagocytosis of Aspergillus organisms. We examined the amount and type of complement component C3 bound to the resting conidia of 29 isolates from nine Aspergillus species. The highly pathogenic species A. fumigatus and A. flavus bound fewer C3 molecules per unit of conidial surface area than did the less pathogenic species A. glaucus, A. nidulans, A. niger, A. ochraceus, A. terreus, A. versicolor, and A. wentii, as determined by quantitative flow cytometry. Immunoblot analysis of C3 fragments bound to conidia demonstrated that for all species most C3b was apparently converted to iC3b. For seven species, iC3b was clearly the major C3 product recognized by immunoblotting. However, A. niger and A. nidulans appear to promote further breakdown of opsonic C3 fragments to C3dg. We found significant variations in size and C3 binding among isolates within the same species. Intraspecies variation may contribute to seemingly discrepant results obtained in studies of Aspergillus phagocytosis.

Specificity of three anti-complement factor 3 monoclonal antibodies.

Monoclonal antibodies which recognize specific C3 fragments may be used to distinguish C3 cleavage products bound to organisms. We defined the specificity of three commercially available monoclonal antibodies by Western immunoblot analysis, enzyme-linked immunosorbent assay, and a quantitative flow cytometric technique. Two monoclonal antibodies with specificity for (i) an erythrocyte-bound C3d epitope or (ii) an erythrocyte-bound C3c epitope retained their specificity in all assays. However, the third monoclonal antibody with selectivity for erythrocyte-bound C3bi failed to retain specificity for C3bi bound to non-erythrocyte surfaces in each of our assays; binding instead to all C3 fragments containing the C3g domain. We postulate that erythrocyte C3b-binding surface proteins may alter the availability of certain C3b epitopes and influence observed anti-C3 monoclonal antibody specificity. We conclude that the specificity of monoclonal antibodies for C3 fragments should be confirmed with assays which do not employ erythrocytes or other surfaces bearing C3 receptors. Also, our quantitative flow cytometric technique is a potentially valuable tool for the enumeration of particle-bound C3 fragments.

Role of the Staphylococcus epidermidis slime layer in experimental tunnel tract infections.

An experimental animal model was used to assess the slime layer of Staphylococcus epidermidis as a pathogenic factor in tunnel tract infections. Mice were inoculated with high-slime-producing or non-slime-producing strains of S. epidermidis, either along the length of a subcutaneous catheter or in the area where a catheter had been placed and immediately removed (controls). Among the catheter-bearing mice, the phenotypically distinct staphylococci produced similar, high frequencies of abscess formation (72% [44 of 61] versus 81% [31 of 38]; P = 0.29). In controls, the non-slime-producing organisms were significantly more pathogenic (87% [40 of 46] versus 57% [25 of 44] abscess formation; P = 0.001). No consistent difference was detected between blood isolates obtained from patients with central venous catheter bacteremia and those from neonates with bacteremia in the absence of a prosthetic medical device. Quantitative culture of removed catheters showed greater adherence by the slime-producing isolates (P = 0.014). In this mouse model, slime production by S. epidermidis did not increase the risk of catheter tunnel tract infection, despite the greater catheter adherence of the slime-producing organisms. These findings suggest that traumatized tissue may be a sufficient condition for the development of S. epidermidis catheter-associated infections.

Correlation between antibody affinity and serum bactericidal activity in infants.

Nearly one-half of infants immunized with Haemophilus influenzae b capsular polysaccharide (polyribosylribitol phosphate; PRP)-protein conjugate produce low-affinity antibody. To test the hypothesis that antibody affinity is linked to biologic function, sera were obtained before and 1 month after immunization of 18-month-old infants with PRP-diphtheria toxoid conjugate vaccine. Correlation was attempted of anti-PRP affinity, concentrations of anti-PRP, and anti-outer membrane proteins and of immunoglobulin isotype with bactericidal activity. Nine subjects produced anti-PRP of low affinity (K less than 10(4) l/mol), and 11 had higher affinity antibodies (average K, 2.8 x 10(4) l/mol). By multiple regression analysis, antibody affinity was the only variable significantly related to the bactericidal activity of serum after immunization with the conjugate vaccine (r = .71; P = .04). Thus, serum anti-PRP from a substantial proportion of infants appeared functionally deficient in association with low-affinity antibody.

Complement component 3 binding to Haemophilus influenzae type b in the presence of anticapsular and anti-outer membrane antibodies.

Antibodies directed against the capsular polysaccharide (polyribosyl ribitol phosphate [PRP]) or the outer membrane proteins (OMP) of Haemophilus influenzae type b (Hib) promote bactericidal activity, complement 3 (C3) binding, and ingestion by phagocytic cells. To assess the relative contribution of anti-OMP to host defense against Hib, we compared the opsonic activities of anti-PRP and anti-OMP as reflected by the amounts of C3 bound to the bacterial surface. Immunoglobulin G (IgG) fractions containing either anti-PRP or anti-OMP were incubated with Hib in the presence of a C5-deficient complement source. C3, total IgG, and IgG subclasses bound to the bacteria were quantified by enzyme-linked immunosorbent assay. The maximum amount of C3 which could be bound to Hib was greater in the presence of anti-PRP than in the presence of anti-OMP. Also, except at low IgG concentrations, the rate of increase in bound C3 as a function of increasing IgG concentration was greater for anti-PRP than for anti-OMP. Hib-bound anti-OMP consisted primarily of IgG1 and IgG3, whereas bound anti-PRP was primarily IgG1 and IgG2. Thus, the potential for C3 binding to Hib is greater in the presence of anti-PRP than in the presence of anti-OMP, probably because of the larger number of binding sites available to the former. Nonetheless, OMP appear to provide important targets for opsonic antibody and would be logical components of a PRP-conjugate vaccine or may be efficacious as vaccines against nontypeable H. influenzae.

Safety and pharmacokinetics of 566C80, a hydroxynaphthoquinone with anti-Pneumocystis carinii activity: a phase I study in human immunodeficiency virus (HIV)-infected men.

A hydroxynaphthoquinone compound (566C80) has been shown to be effective in the prevention and treatment of murine Pneumocystis carinii pneumonitis. In a phase I study, five cohorts of four human immunodeficiency virus-infected men received 100, 250, 750, 1500, and 3000 mg of the compound orally once daily for 12 days. A sixth cohort received 750 mg three times daily for 5 days, then twice daily for 16 days. Evaluation included clinical, hematologic, and biochemical studies and the pharmacokinetics of 566C80. The only drug-related adverse effect was a maculopapular rash in one patient that resolved without discontinuation of the drug. With the largest dosage tested (3000 mg) the following pharmacokinetic measures were achieved: maximum plasma concentration, 39 micrograms/ml; time to maximum plasma concentration, 8.0 h; area under plasma concentration-time curve at steady state, 1088 h.micrograms/ml; plasma half-life, 51 h; and total plasma clearance, 4.09 l/h. Compound 566C80 offers promise as a new drug class for P. carinii pneumonia.

Antibody affinity in infants after immunization with conjugated capsular polysaccharide from Haemophilus influenzae type b.

The affinities of IgG antibodies to Haemophilus influenzae b capsular polysaccharide (polyribosyl ribitol phosphate [PRP]) elicited 1 month after immunization of 47 infants 2-greater than 18 months of age with a PRP-outer membrane protein conjugate (PRP-OMP) were measured by ELISA. Thirty-four sera had affinities distributed normally about a logarithmic mean of 3.2 x 10(5) l/mol, but 13 samples had undetectable affinities (less than 10(4) l/mol). Median affinities of sera from children 2-6 (1.5 x 10(5) l/mol) and 7-11 months of age (1.6 x 10(5) l/mol) were significantly greater than the median affinities of sera from infants 12-18 (1.8 x 10(4) l/mol) or greater than 18 months of age (4.2 x 10(4) l/mol). Sera from children greater than 18 months of age vaccinated with PRP conjugated to diphtheria toxoid had a median affinity of 6.1 x 10(4) l/mol, equivalent to that of the same age group vaccinated with PRP-OMP. Children vaccinated with PRP conjugate vaccines may produce antibodies of very low affinity, a finding that may have significance for protection from invasive disease.

Bacterial antigens and neutrophil granule proteins in middle ear effusions.

Otitis media with effusion is a significant cause of hearing loss in young children. We hypothesized that persistent bacterial antigens in middle ear effusions (MEEs) might act as chronic inflammatory stimuli causing release of neutrophil proteins. Concentrations of neutrophil lactoferrin and a 37-kd cationic bactericidal protein (CAP 37) were measured in 47 MEEs collected from 27 children at the time of tympanostomy tube placement. Antigens of Streptococcus pneumoniae were detected by latex particle agglutination and those of Haemophilus influenzae by dot-blot assay. Bacterial antigens were detectable in 24 (51%) of MEEs: S pneumoniae in 10 (21%), H influenzae in 12 (26%), and both antigens in 2 (4%). Concentrations of lactoferrin and CAP 37 in H influenzae antigen-positive MEEs were significantly higher than in either S pneumoniae antigen-positive or antigen-negative MEEs. We conclude that H influenzae antigen causes a greater middle-ear inflammatory response, as judged by neutrophil products, than does S pneumoniae antigen.