Streptococcal nucleases. I. Further studies on the A, B, and C enzymes.

Streptococcal DNAse C is more resistant to heat inactivation than the A or B enzyme. DNAses A and C are indifferent to the bacterial ribonucleic acid inhibitor whereas the B enzyme is markedly inhibited. Prolonged digestion with relatively large amounts of DNAse B results in chemical and biological destruction of the inhibitor. Ribonuclease as well as deoxyribonuclease activity is associated with the B enzyme. Both activities require divalent cations and both are inhibited by bacterial ribonucleic acid. The ratios of the two activities are constant in various preparations and after partial heat inactivation. Mutual inhibition of the two activities can be demonstrated in mixed substrate systems. The evidence presented is consistent with the view that the B enzyme is a single nuclease which can attack both deoxyribonucleic and ribonucleic acids.

The serum opacity reaction of Streptococcus pyogenes. The demonstration of multiple, strain-specific lipoproteinase antigens.

Investigation into the antigenicity of streptococcal lipoproteinase has revealed the existence of multiple, immunologically distinct enzymes. Each lipoproteinase identified was found to be strain specific in that it was found only in strains of a particular T-agglutination pattern. In some T patterns, all streptococci of that T pattern which were examined shared a common lipoproteinase antigen. In other T patterns, strains which produced antigenically different lipoproteinases were identified. Evidence that the lipoproteinase antigen is distinct from other well-established cellular antigens of Group A streptococci has been presented. The production of this strain-specific enzyme by strains currently difficult to type by the standard M precipitin method may facilitate more precise identification of these strains and a better assessment of their role in the pathogenesis of Group A streptococcal infections and their sequelae.

Bacterial interference in experimental burns.

A standardized, full thickness, dermal burn in rabbits was used to study interference between strains of Staph. aureus inoculated on the wound surface. Several strains appeared equally capable of colonizing lesions and of preventing superinfection by other staphylococci inoculated at a later time. In addition, cross-infection between rabbits colonized by different strains (502A and Q461) and placed together in cages was prevented, presumably by the same mechanism. Interference appeared to be a strictly local phenomenon, since it did not occur when an animal was colonized by strain 502A at one burn site and subsequently challenged with strain Q461 at a separate lesion. For interference to occur, a minimal time interval (9 hr) was required between inoculation of the interfering strain and inoculation of the challenge strain. In vivo growth rates indicated rapid growth in the first 24 hr by the interfering strain but no detectable multiplication by the challenge strain. Heat-killed staphylococci, even in large numbers, were incapable of producing interference. Penicillin treatment of animals colonized by strain 502A (penicillin-sensitive) abolished interference with strain Q461 (penicillin-resistant). These findings indicate that bacterial multiplication by the interfering strain is an essential feature of this phenomenon. The mechanism of interference between strains of Staph. aureus remains obscure. There was no evidence in these studies for direct bacterial antagonism in vitro or in vivo between most of the strains examined; yet, all were capable of producing interference. Attempts to identify antistaphylococcal activity in passively transferred tissue homogenates and serum collected from infected animals were also negative. The ability of large inocula of staphylococci grown in broth to superinfect colonized lesions indicates that the numerical superiority of the interfering strain over the challenge strain is an important aspect of interference. The observation that in vivo-grown organisms may superinfect in significantly smaller quantities is suggestive of a qualitative advantage as well.

Streptococcal nucleases. II. Characterization of DNAse D.

Preparations of streptococcal DNAse D with high specific activity and free of other streptococcal nucleases have been obtained by zone electrophoresis and column chromatography. Antisera prepared by injecting rabbits with such preparations specifically neutralize the activity of this enzyme. As with DNAse B, preparations of DNAse D regularly exhibit ribonuclease activity. For both B and D enzymes, the order of substrate preference is thymus DNA, yeast RNA, bacterial RNA; but the specific activity of the D enzyme is higher than that of the B enzyme with respect to thymus DNA and lower with respect to bacterial RNA. Both the deoxyribonuclease and the ribonuclease activities exhibited by preparations of both enzymes are inhibited by bacterial RNA, but approximately 100-fold greater concentrations of bacterial RNA are required to achieve inhibition of the deoxyribonuclease activity of the D enzyme equivalent to the inhibition of the B enzyme. The deoxyribonuclease activity of the D enzyme is also inhibited by yeast RNA, but even larger amounts are required. These observations indicate that the D enzyme is immunologically distinct from the other streptococcal nucleases and that it differs quantitatively from the B enzyme with respect to relative specific activities on different substrates and behavior in the presence of the bacterial ribonucleic acid inhibitor.

Transduction of bacteriocin determinants in group A streptococci.

Determinants of streptococcin A-FF22 (SA) production and host cell immunity have been transduced to three serologically distinct Group A streptococci. Streptomycin resistance markers were not cotransducible with bacteriocin determinants. SA+ transductants of strains unrelated to the parent SA+ strain were unstable but SA+ transductants of a spontaneous SA- derivative of the parent appeared to be stable.

Biochemical and immunological characterization of the extracellular nucleases of group B streptococci.

Nearly all group B streptococcal strains representing the five major serotypes were found to produce extracellular nucleases by screening with an agar-well-diffusion technique in DNA-methyl green agar plates. Three different nucleases have been isolated and partially purified by DEAE-and carboxymethyl-cellulose chromatography. They possessed different mobilities on polyacrylamide gel electrophoresis and different molecular weights. These nucleases, designated I, II, and III, are optimally activated by cations of calcium and manganese and exhibited RNase as well as DNase activity. Despite differences in their physical and biochemical properties, nucleases II and III appear antigenically similar, but distinct from nuclease I. These group B streptococcal nucleases are immunologically different from the nucleases of group A streptococci. Neutralizing activity, probably antibody, to nucleases II and III was found in human sera, and was most prevalent in sera of pregnant women colonized with group B streptococci and in their newborn infants.

Suppression of the antistreptolysin O response by cholesterol and by lipid extracts of rabbit skin.

Lipids extracted from rabbit skin block the hemolytic capacity of SO and also suppress the neutralizing antibody response to this streptococcal extracellular antigen in rabbith immunized intravenosly. The modification in antibody response is specific for SO; the antibody responses to streptococcal DNase B and to streptococcal NADase are not affected. Cholesterol, a lipid present in abundance in skin, has a similar specific effect on the antigenicity of SO and may be the component responsible for the demonstrated effects of these lipid extracts of skin. In vitro experiments indicate that lipid extracts of rabbit skin have a greater capacity to block the hemolytic capacity of SO than do lipid extracts of rabbit heart, kidney, lung, liver, or spleen. These data support the view that the feeble ASO response observed in patients with streptococcal pyoderma is a result of the abundance of a local lipid inhibitor, such as cholesterol, in the skin. They may also bear on the pathogenesis of rheumatic fever, a complication which apparently does not occur following group A streptococcal pyoderma. Two possible explanations for this remarkable epidemiologic observation, both related to the presence of a local inhibitor, are considered: (a) suppression of the ASO response, the magnitude of which has been correlated with the risk of developing rheumatic fever after streptococcal infection of the throat, and (b) inhibition of the toxicity of SO, which has been shown to have a direct toxic effect on the mammalian heart and on isolated beating myocytes.

Experimental infection of the skin in the hamster simulating human impetigo. III. Interaction between staphylococci and group A streptococci.

The interaction between staphylococci and Group A beta hemolytic streptococci in mixed lesions was investigated in an experimental impetigo model. A strain of staphylococcus of phage Type 71, which has been shown in vitro to produce a bacteriocin for streptococci and other Gram-positive organisms, eliminates or reduces Group A streptococci in mixed lesions. In contrast, staphylococcal strains of phage Types 75 and 81, which do not produce a demonstrable bacteriocin in vitro, exhibit no such effect. Some variation was noted in the in vivo response of two different streptococcal M Types to the bactericidal effect of phage Type 71 staphylococci. Bacterial antagonism is more pronounced when staphylococci and streptococci are injected simultaneously into animals than when staphylococci are superimposed on preexisting streptococcal lesions. Marked variations were found in the numbers of viable streptococci (colony-forming units) recovered from individual lesions containing identical mixtures of streptococci and phage Type 71 staphylococci. The frequency of a demonstrable bactericidal effect was related to the number of streptococci injected. With small inocula of streptococci, the tendency towards an all-or-none effect was particularly striking. No evidence of selection of streptococcal or staphylococcal mutants which might explain this phenomenon was obtained. These observations suggest that the bactericidal effect of phage Type 71 staphylococci on other Gram-positive organisms, previously demonstrated in vitro, appears to operate also in vivo.

Intergroup phage reactions and transduction between group C and group A streptococci.

In a study of intergroup reactions, four virulent Group A streptococcal phages were found to form plaques in high titer on lawns prepared from a number of Group C streptococcal strains. Whether the phages were propagated on the homologous (Group A) strain or a heterologous (Group C) strain did not appear to influence consistently the plaque-forming efficiency on lawns prepared from a homologous (Group A) or a heterologous (Group C) strain or to alter significantly the percent of Group C strains which showed plaque formation. Considerable variability was found in the ability of temperate phages to lyse strains of a heterologous group. A single Group C indicator strain was lysed by a high percentage of freshly induced temperate Group A phages. A single temperate Group C phage lysed a significant proportion of Group A strains when freshly induced or when propagated on a Group A strain. Intragroup transduction of streptomycin resistance was demonstrated between Group C strains. Intergroup transduction of streptomycin resistance and also bacitracin resistance was achieved between Group C and Group A streptococci. These observations provide evidence that Group A streptococci can serve as recipients in intergroup transmission of genetic information. Ultraviolet irradiation of the transducing lysate and lowering the propagation temperature of the transducing lysate increased the frequency of transduction in both the intragroup and intergroup transduction systems.

Immunological properties of hyaluronidases associated with temperate bacteriophages of group A streptococci.

The antigenic relationships of hyaluronidases, bound and free, associated with temperate bacteriophages of group A streptococci were examined with antibody against purified whole phage and with antibody against phage-bound enzyme released by urea and purified to homogeneity. Studies performed by double diffusion in agar (ouchterlony) with antibody against the homologous purified enzyme from a temperate phage of a type 49 streptococcus indicated that the bound and free enzyme gave a single line of identity and that the free hyaluronidase activities in induced lysates of four strains of M type 49 streptococci were immunologically indistinguishable but different from the enzyme in induced lysates of a heterologous type. The four M type 49 strains were from widely different geographical or temporal sources and of different phage subtypes as determined by lyxic patterns. These findings were confirmed in studies that employed a functional assay of enzyme neutralization. An immunoglobulin preparation of antiserum against the purified enzyme as well as one against homologous purified whole phage neutralized the hyaluronidase activity produced by induction of the M type 49 strains and present either phage-bound or soluble in phage-free lysates. These immunoglobulin preparations had little effect on the hyaluronidase activities present in phage-lysates of other M types of group A streptococci. Inhibition of propagation of temperate phages by antibody against the purified phage hyaluronidase paralleled the neutralization of phage-associated enzyme activity by this antibody, indicating that antibody to the purified enzyme can inhibit phage infection. Antibody preparations against the purified phage-bound enzyme or against purified whole phage did not neutralize the extracellular hyaluronidase in the supernate of an uninduced culture of M type 4 streptococci. A human serum strongly inhibitory for the extracellular enzyme of this strain or on the purified phage enzyme from an M type 49 strain. The results support the view that the hyaluronidases associated with the temperate bacteriophages from various M types of group A streptococci do not share common antigenic determinants but that an immunological specificity exists that parallels the serologic specificity of the M protein of the host strains.

Bactericidal substance from Staphylococcus aureus. Biological properties.

A bactericidal substance previously isolated from phage type 71 Slaphylococcus aureus has been further identified and characterized. Staphylococci belonging to phage type 71 produce the substance in higher titers than staphylococci lysed by other phages in group II in addition to phage 71. Other staphylococci do not produce the bactericidal substance. The bactericidal substance shares several of the properties of bacteriocins but differs from this group of antibiotic substances in some respects. A combination of ammonium sulfate fractionation and gel filtration on a Sephadex G-100 column resulted in considerable degree of purification of the bactericidal substance. The substance is a previously unrecognized product of S. aureus and is distinct from other extracellular products of this organism.

Studies of the receptor for phage A25 in group A streptococci: the role of peptidoglycan in reversible adsorption.

Irreversible adsorption of a virulent phage, phage A25, to heat-killed streptococci, groups A, G, and A variant, has been achieved. Adsorption reflected the observed host range for phage A25 in that heat-killed group B cells were not able to inactivate the phage. Broken cells, cell walls, and peptidoglycan prepared from a group A strain K56 failed to adsorb the phage irreversibly, but retained the potential to carry out reversible adsorption. Experimental data including electron microscopy have demonstrated the specificity of reversible adsorption and have identified the peptidoglycan as a necessary cellular component of the receptor. The sensitivity of whole cells and purified peptidoglycan to muralytic enzymes suggests that the cell wall and peptidoglycan must be intact for optimal adsorption. In general the results are explained by postulating that adsorption of A25 phage particles to group A cells occurs by a two-step process; the first step involves recognition and reversible binding of the phage tail to the cell wall peptidoglycan, the second step is an irreversible reaction catalyzed by a yet unidentified cellular component which is destroyed when cells are ruptured.

Group A streptococcal bacteriocin. Production, purification, and mode of action.

A bacteriocin, streptocin A, was isolated from the supernatant fluid of tryptic soy broth cultures of Group A streptococcus strain FF-22. Evidence was obtained which supports the view that the failure to recover active streptocin A after growth of the producer strain in certain fluid media is due to the inactivation of the bacteriocin by concomitantly synthesized streptococcal proteinase. The bacteriocin was purified 139-fold and the active product appeared to be of uniform size, having a molecular weight of approximately 8,000. Streptocin A was bactericidal, but not lytic, for a susceptible Group A streptococcus and the lethal effect was markedly temperature dependent. The bacteriocin inhibited the synthesis of DNA, RNA, and protein, and also prevented the uptake and incorporation of glucose by the sensitive cells. Degradation of RNA occurred, but appeared to be less than that produced by a staphylococcal bacteriocin. This effect may be due to differences in the killing potency of the two bacteriocins in preparations having similar inhibitory activity when measured by lawn culture assays.

M antigens among group A streptococci isolated from skin lesions.

Three new provisional types of Group A streptococci have been identified among strains isolated from skin infections of American Indian children. A previously established type, Type 41, was also recognized among the pyoderma strains. Most of the skin streptococci with T-agglutination pattern 3/13/B3264 could be identified as Type 41 or as one of two new provisional types, Schoenborn (Type 52) and Hanson (proposed Type 53). Strains of the third new provisional type, Kingbird (proposed Type 54), agglutinated as 15/17/19/23/47, a T pattern not associated with skin infections in studies of other populations. Approximately half of the pyoderma strains examined were found to produce an M antigen; the proportion of M-typable strains from skin infections was found to be similar to that currently found in throat cultures of children with pharyngitis. The demonstration of M antigens in streptococci from skin lesions provides evidence of their potential virulence and should facilitate their classification in subsequent epidemiological studies.

Natural history of impetigo. II. Etiologic agents and bacterial interactions.

Intensive observations on 37 children in a population with endemic skin infections provided an opportunity to study the interrelationships between and the significance of the bacterial genera commonly associated with impetigo. Cultures of the respiratory tract, three normal skin sites, and lesions, when present, were taken three times weekly from July to October 1969. Impetigo developed in all 37 children. Group A streptococci alone were recovered from 21% of 361 lesions, Staphylococcus aureus alone from 8%, Staphylococcus epidermidis alone from 5% and mixtures of streptococci and staphylococci from 61%. Vesicular or pustular lesions were more often pure streptococcal than pure staphylococcal. Streptococci alone were more often recovered from early stage lesions rather than from later ones. The pure staphylococcal lesions characteristically occurred early in the season whereas streptococcal or mixed lesions had later peaks.Serial observations on 74 lesions revealed longer persistence of streptococci than staphylococci in mixed lesions. In 85% of the instances the same streptococcal serotype was recovered repeatedly from an individual lesion, whereas staphylococcal types changed in 57% of instances. Phage type 75 accounted for the majority of staphylococcal isolates from all sites, whereas phage type 54 was recovered only from skin lesions. In contrast to streptococci, the site sequence of staphylococcal spread was from the nose to normal skin to skin lesions. These studies reveal important differences in the migration of staphylococci (as compared with streptococci) to various body sites and suggest a subsidiary role for staphylococci in nonbullous impetiginous lesions yielding both organisms.

Natural history of impetigo. I. Site sequence of acquisition and familial patterns of spread of cutaneous streptococci.

The appearance on and spread of Group A streptococci among different body sites in relationship to the development of impetigo were studied prospectively in 31 children in five families. During July and August 1969 intensive clinical, bacteriological, and serological observations were made, including cultures taken at least every other day. In individual children, site sequence of spread of Group A streptococci was from normal skin to lesions and finally to respiratory tract. Streptococci were recovered from normal skin before development of lesions (mean interval of 10 days) in 74% of episodes. Recovery of streptococci from nose and throat followed (by means of 14 and 20 days, respectively) skin acquisition of streptococci (97% of episodes) and lesions (74% of episodes).Distribution of positive normal skin sites among wrist, ankle, and back was similar (28-37%) although 62% of lesions were on the legs. Recovery of a serotype from normal skin was associated with a high risk (76%) of subsequent development of lesions due to that type. New streptococcal serotypes usually entered a family during the peak or decline of a preceding serotoype with a tendency of one to predominate. Among family members the mean interval from index to secondary skin acquisition of streptococci was 4.8 days, but 21 days elapsed from first appearance to last acquisition of skin disease. In the population as a whole, streptococci were recovered in high frequency from normal skin before the increase in prevalence of lesions and also later in the fall when cutaneous infections were absent.

Cellular immune responses to extracellular streptococcal products in rheumatic heart disease.

The lymphocyte transformation responses to purified preparations of two extracellular products of group A streptococci (blastogen A and nuclease B), to phytohemagglutinin, and to Candida albicans antigen were measured in tonsillar and peripheral blood lymphocytes from patients with rheumatic heart disease (RHD) and suitably matched nonrheumatic (control) subjects. The mean phytohemagglutinin dose responses of tonsillar and peripheral lymphocytes from RHD patients were essentially indistinguishable from those of controls. In contrast, the responses of tonsillar and peripheral blood lymphocytes to the two extracellular products of group A streptococci were significantly lower in RHD patients than in nonrheumatic control subjects. Candida antigen produced very little stimulation of lymphocytes in any of the subjects. The geometric means of antibody levels against streptolysin O, nuclease B, and nicotinamide adenine dinucleotidase showed no consistent differences between the control group and the group of RHD subjects. Group A streptococci were isolated from the tonsils of approximately 25% of both groups of subjects. The RHD patients clearly had a depressed cellular immune response to the two purified streptococcal extracellular antigens. The equal frequency in recovery of group A streptococci from tonsils and the absence of consistent difference in titers of humoral antibodies to streptococcal extracellular antigens, particularly nuclease B, suggest that this differential response is not due to a lower level of stimulation by repeated exposure to group A streptococcal products.

Attack rates of acute nephritis after type 49 streptococcal infection of the skin and of the respiratory tract.

Prospective studies in a population of American Indian children during an outbreak of acute nephritis associated with the Type 49 Group A streptococcus permitted a comparison of attack rates of renal complications after infection at different sites and at different ages. Acute nephritis or unexplained hematuria developed in 10 of 42 children (23.8%) with Type 49 streptococcal skin infection, in 2 of 44 (4.5%) with Type 49 throat infection, and in 3 of 16 (18.8%) with simultaneous Type 49 infection at both sites. The higher attack rate of nephritis and hematuria in children with pyoderma indicates that skin lesions played a direct and quantitatively greater role than respiratory infection in the pathogenesis of acute nephritis during this outbreak. Skin infections with the Type 49 strain were followed by evidence of renal complications more often in children younger than 6.5 yr (9 of 21 or 43%) than in older children (1 of 21 or 5%). Attack rates of renal complications after Type 49 skin infection were approximately equal in males and females.

The influence of the site of infection on the immune response to group A streptococci.

The immune response after streptococcal infection of the skin and of the upper respiratory tract (URT) was studied prospectively in a group of normal children, ages 3-6 yr. The children were examined and cultures for group A streptococci were obtained weekly from the throat, nose, and skin lesions (when present). Paired sera were collected at the beginning and end of the study, and the changes in antibody titers were measured for three different streptococcal antigens: streptolysin O, deoxyribonuclease B (DNAse B), and nicotinamide adenine dinucleotidase (NADase). The findings suggest that in contrast to infection of the URT antibody response to streptolysin O is relatively feeble after streptococcal infection which is limited to the skin. The response to NADase is also poor after cutaneous infection. Antibody responses to DNAse B are generally good regardless of the site of the infection. These and other studies indicate that anti-DNAse B is the antibody of choice in studying streptococcal infection of the skin and its complications.

C-reactive protein in streptococcal pharyngitis.

This study was designed to explore whether the test for C-reactive protein (CRP) is useful in differentiating bona fide streptococcal infection from the symptomatic carrier at the time of the acute visit to the physician. Serial blood samples from 157 children with symptomatic pharyngitis and a positive culture for group A streptococci were analyzed for the presence or absence of CRP. These data were compared with the patients' antibody responses to two streptococcal extracellular antigens (antistreptolysin O and antistreptococcal deoxyribonuclease B). Seventy-eight percent of patients with serologically confirmed streptococcal pharyngitis had a positive CRP test at the initial visit. Conversely, if the CRP test was negative at the acute visit, only about 25% later showed an antibody response. This latter finding held regardless of the degree of positivity of the initial culture, the presence of exudate or adenitis, or the presence of a temperature greater than 38.3 C (101 F) or coryza. These data suggest that the CRP test may be helpful to the clinician, especially if this abnormal protein is absent at the time of the acute visit.