Serum antibodies against Staphylococcus aureus antigens in healthy individuals and patients with invasive infections

Abstract

The present work aims to study the antibody responses against Staphylococcus aureus in health and in deep infections and is composed of six papers. We developed a sandwich enzyme immuno assay (EIA) with a detection limit for alpha-toxin at nanograms levels. This method is applicable to measure alpha-toxin in culture supernatants and in human fluids. The assay is simple, reproducible and performed within 6 hours. We determined the presence of alpha-toxin in serum samples and homologous strains from 41 patients with septicaemia due to Staphylococcus aureus. We detected alpha-toxin in 22% (6/27) of the sera taken on admission from patients whith symptoms presented for ~7 days. In all, alpha-toxin was found in 20 % (8/41) serum samples taken on admission. In seven of these eight samples low antibody titres were also noted. 57 % of the patients ( 20/35) showed a positive antibody antibody detection increased the diagnostic sensitivity to 52 % in early serum samples. The subsequent studies are based on 2 prospective studies performed in Sweden with aproximately 10 years in between. We use EIA method to measure the antibody response. The clinical material for the first prospective study was collected during 1988-1991 at the Department of Infectious Diseases, Örebro University Hospital. 240 consecutive samples from 63 patients suffering S. aureus septicaemia were collected.

We studied the antibody response during disease against three antigens: alpha-toxin, teichoic acid, and lipase. A considerable individual variation in the response to the different antigens was noted, as was a good correlation between clinical course of the disease and the antibody response, indicating the possible use of serology for monitoring disease and recovery. Antibody levels against alpha-toxin measured in the EIA correlated with the alpha-toxin neutralizing activity of the patient sera. Patients with initially low antibody levels displayed a poorer antibody response during the disease than those with higher antibody levels; this phenomenon was observed with the three antigens alpha-toxin, teichoic acid, and lipase, but was most pronounced with alpha-toxin (p < 0.001). Furthermore, initially low antibody levels were associated with the development of complicated septicaemia (11 out of 13 patients). These latter observations point at the significance of low antibody levels in staphylococcal disease. We also analyzed levels of antibodies against clumping factor A (Clf-A) and extracellular fibrinogen binding protein (Efb), antigens belonging to the group of fibrinogen binding proteins. No correlation was seen between levels against these antigens indicating that measured antibodies were produced independently. Antibody levels against Efb were significantly lower in the acute sera than in sera from healthy individuals (p = 0.002). This difference may reflect that individuals with lower antibody levels against Efb are more prone to develop deep staphylococcal infections.

Sample collection for the second prospective study was performed from 2003 to 2005 at the Clinic of Infectious Diseases in Skövde, comprised 218 samples from 96 patients, collected at the time of diagnosis, after completed antibiotic treatment and one month after the end of treatment. We investigated the antibody response against the five previously described antigens and furthermore included three more antigens, enterotoxin A (SEA), Toxic Shock Syndrome Toxin 1(TSST), and Scalded Skin Syndrome toxin (SSS). We found a correlation between patient strains agr-group, PFGE-group, alpha-toxin production and patient initial antibody levels. Patients with fatal outcome displayed lower initial antibody levels to all antigens and significantly to teichoic acid, lipase, SEA, and SSS. In episodes with complicated bacteraemia, significantly lower levels to teichoic acid and lipase were registered in the initial sample.

We analysed the antibody levels in 151 healthy individuals with matched ages as to give reference values to sepsis patients. Eleven antigens were analysed including three new antigens: Clumping factor B (Clf-B), bone sialoprotein-binding (Bsp) protein and Extracellular adherence protein (Eap). We found great individual variations in antibody levels, both in young and elder individuals. Occurrence of S. aureus in the nares at time of sampling was correlated to higher antibody levels, specially against the extracellular proteins alpha-toxin, lipase, SEA, TSS-1, and SSS. Elderly individuals over 65 y showed slightly lower antibody levels. Certain individuals were more prone to produce or not to produce antibodies than others, and certain extracellular proteins were more often inducing high IgG levels.

Conclusions. Analysis of the antibody levels in S. aureus septicaemia showed a variation between patients and between different antigens both in disease and in health. Low antibody levels may have a prognostic value. Healthy colonized individuals show higher antibody levels against extracellular proteins, alpha-toxin, lipase, SEA, TSS-1, SSS. The serological diagnosis can be improved through the use of more than one antigen in a combinatory analysis and serological analysis in the future may be important for monitoring the diseases. It may also be crucial for monitoring the use of passive and active immune therapy against S. aureus invasive disease.