The peroxidase-antiperoxidase immunoassay was developed by using selected Salmonella serotypes to evaluate its potential for use in diagnostic bacteriology. S. choleraesuis var. kunzendorf, S. dublin, and S. typhimurium were the test organisms. Strong specific staining with corresponding antiserum was achieved with smears of each Salmonella serotype oh microscope slides from formalinized cell suspensions, live broth cultures of clinical isolates, and tissue suspensions from the livers and spleens of experimentally infected mice. In addition, S. choleraesuis var. kunzendorf was detected in Formalin-fixed and fresh frozen tissues from experimentally infected pigs. The results of this study indicate that the peroxidase-antiperoxidase assay is wellsuited for the rapid identification of Salmonella from pure cultures and that the technique can be useful in research for the detection of this pathogen in histological sections.

C3, C4, factor B, properdin, and C2 binding to serum-sensitive and serum-resistant gonococci was quantitated in C8-deficient and normal human serum by using fluorescein-conjugated antibodies and 3H-labeled components. Organism and serum-specific differences were noted, the most striking of which involved factor B and properdin binding to the serum-sensitive strains in the different sera. C3 binding to these organisms was quantitatively and kinetically equivalent in C8-deficient and normal human serum. In contrast, factor B and properdin binding reached a plateau after 5 min in C8-deficient serum but peaked and fell to control values in normal human serum. Identical results were obtained with normal human serum immunochemically depleted of C8. Between 7 and 15% of the bound C3 participated in formation of the alternative pathway convertase C3bBb/P. Reconstitution of the C cascade by adding purified C8 to C8-deficient serum led to the loss of factor B previously bound to the organisms. Factor B loss occurred coincident with bacterial killing and membrane disruption as observed by electron microscopy. Prevention of membrane disruption by depleting normal human serum of lysozyme had no effect on killing and failed to prevent factor B loss. Stabilization of the C3bBb complex with Ni2+ prevented factor B loss as well as gross membrane disruption but not bacterial killing. C2 (the classical pathway analog of factor B) binding to gonococci was equivalent in C8-deficient and normal human serum peaking within 2.5 min and falling to control values in both sera thereafter. We conclude that the assembly of the membrane attack complex promotes decay of C3bBb/P with release of factor B and properdin but not C3 from the organism surface. Membrane disruption does not appear to be required for this effect. This activity may represent a mechanism to limit continued C consumption.

The eighth component of C (C8) is composed of two subunits C8 beta and C8 alpha-gamma, which are non-covalently bound in a 1/1 ratio in the intact molecule. The genes encoding the polypeptide chains composing the subunits demonstrate close genetic linkage. To assess the functional expression of these genes at the protein level, normal human serum and C8-deficient sera were electrophoresed in native polyacrylamide gels following which C8, C8 beta, and C8 alpha-gamma were detected using hemolytic overlays. These experiments demonstrated that normal sera contained free C8 alpha-gamma in addition to intact C8. Free C8 alpha-gamma was not observed when C8 was reconstituted by mixing C8 beta-deficient serum with C8 alpha-gamma-deficient serum in a ratio optimized for C8 activity, suggesting that the free C8 alpha-gamma observed in normal serum was not due to dissociation of intact C8. Inasmuch as this technique did not adequately separate C8 and C8 beta, sera were also examined by anion exchange chromatography. C8 alpha-gamma-deficient serum contained C8 beta in a single peak in the 1.4 ms/cm fall through. C8 beta-deficient serum contained a major peak of C8 alpha-gamma at 7.1 ms/cm and a lesser peak coeluting with C9 at 9.5 ms/cm. Normal serum contained both intact C8 eluting between 2.4 to 5.5 ms/cm and C8 alpha-gamma eluting at 7.1 ms/cm. Free C8 beta was not detectable in normal serum indicating that free C8 alpha-gamma was not due to C8 dissociation. Mixing aliquots from the chromatographic peak of C8 beta activity with the peaks of C8 alpha-gamma activity in C8 beta-deficient serum or in normal serum generated intact C8 hemolytic activity. Non-reducing SDS-PAGE and Western blotting with anti-C8 confirmed the presence of antigenic material of appropriate m.w. in each peak. These findings demonstrate that serum contains excess C8 alpha-gamma relative to C8 beta, despite the equimolar presence of the subunits in intact C8. Thus the availability of C8 beta determines the quantity of C8 produced. Further, these data suggest the possibility that the C8 structural genes may be differentially expressed despite their close genetic linkage.