SUMMARYThe ability to produce opacity in horse serum is a characteristic of certain M types of group A streptococci. The types that produce the opacity factor (0 factor) are generally those for which it is difficult to produce good anti-M sera. M-positive (M+) and true M-negative (M-) variants of strains that belong to serotypes in which the serum opacity reaction (s.0.r.) is positive both possess the 0 factor, but there is a difference in binding of the factor to other cellular components in the two variants. The 0 factor is closely bound to the wall-membrane fraction of M-cells, whereas in M + cells it is easily extracted by Lancefield's method or I yo sodium deoxycholate.It is detectable in broth culture supernatants, in the cytoplasm and in the areas surrounding colonies in pour plates of M+ but not of M-cultures. The 0 factor is poorly antigenic, but when it is possible to obtain a good antiserum the inhibitory action is type specific. The 0 factor produced by several M types appears to be inhibited by normal rabbit serum. I N T R O D U C T I O NThe production of opacity in horse serum by group A streptococci is believed to be mediated by a lipoproteinase which acts upon the a,-lipoprotein fraction of the serum (Krumwiede, 1954;Rowen & Martin, 1963). Streptococci of groups other than A apparently do not produce opacity in serum, and within group A the serum opacity reaction (s.0.r.) is positive only in certain serotypes. Gooder (1961) and Kohler (1963) grew streptococci in a liquid medium consisting of 3 parts horse serum and I part Hartley digest broth, and recorded the presence of opacity in the culture supernatant as a positive s.0.r. Consistently s.0.r.-positive serotypes were those in which no M antigen was detectable, or for which satisfactory anti-M sera were difficult to prepare; on the other hand, members of a number of easily recognizable M serotypes were almost invariably s.0.r.-negative. Both workers suggested that there was an inverse relationship between a positive s.0.r. and the production of M antigen. Top & Wannamaker (1968a) incubated killed cells or deoxycholate extracts of fractions consisting predominantly of cell walls and cell membranes (' wall-membrane fractions') in horse serum. They confirmed that opacity was frequently produced by members of 'difficult' M types, but could find no true inverse relationship between M antigen and opacity production, except among type 1 2 strains.Most of the evidence concerning the supposed inverse relationship was obtained by examining collections of M-typable and M-untypable strains with common T-antigen

SUiMARYA streptococcal antigen that is closely associated with the M-antigen, but is not type specific can be detected by means of a complement-fixation test in extracts of M-positive, but not of M-negative, variants of group A streptococci. Purification of acid extracts results in a concomitant increase in the purity both of the typespecific M-antigen and of the M-associated protein (MAP). Antibody to MAP is present in the sera of patients who have had streptococcal infection. The highest titres are found in patients with rheumatic fever.

SOME group-A streptococci cause the appearance of opacity in horse serum (Ward and Rudd, 1938). The production of the opacity factor (OF) is confined to members of certain serotypes (Keogh and Simmons, 1940;Gooder, 1961), and the OF of each serotype is antigenically specific (Top and Wannamaker, 1968a). Widdowson, Maxted and Grant (1970) confirmed the serological specificity of the OF and showed that it corresponded exactly to that of the M type of the streptococcus that produces it. The examination of a limited number of strains of each M type (Widdowson et al., 1970; and our unpublished observations) suggested that OF is produced by all members of 16 welldefined M types, and by a number of other types not so far established by international agreement.We have now made a more sytematic examination of the distribution of OF among strains of group-A streptococci sent to us for typing, and have investigated the use of the serum opacity reaction (SOR), and of its neutralisation by specific antisera, in routine typing and as a means of indentifying hitherto unrecognised streptococcal serotypes. MATERIALS AND METHODSStreptococci Strains used to produce antisera to OF in rabbits were the standard type strains used in the Streptococcus Reference Laboratory for the production of M antisera. Other strains sent to us for type identification from numerous laboratories were used in studies of the distribution of OF.Cultural methods Oxoid Todd-Hewitt Broth with the addition of 1 per cent. Neopeptone was seeded with a loopful of growth from solid medium and incubated overnight at 37°C. These cultures were used for the detection both of OF and of M antigen. Typing techniqueThe methods used in the Streptococcus Reference Laboratory for the preparation of typing sera and for M and T typing are those described by Williams (1958). M antigen was ~

S U M M A R YM-protein and the opacity factor (OF) were both extracted by Lancefield's method from certain types of group A streptococci. The two proteins could not be separated by fractional precipitation with ammonium sulphate, column chromatography or polyacrylamide gel electrophoresis. The OF appeared to be closely associated with the high molecular weight fraction of the M-antigen, and there is some evidence that the two may form part of the same molecule.M-protein and the OF were sought in various other extracts and preparations of M-positive and M-negative variants of serotypes which gave a positive serum opacity reaction (SOR). In all preparations from M-positive variants, the two proteins were always present. In M-negative variants, the OF present in the wall-membrane fraction could be extracted only by ' phage-associated ' lysin. All other extracts from M-negative variants were SOR-negative and M-negative. In the SOR-positive extract obtained by phage lysin treatment of M-negative variants, no M-antigen could be detected by precipitin tests, though the extracts contained a protein of similar electrophoretic mobility to the precipitin-positive protein in the extracts of the M-positive variants of the same serotype.

SUMMARYTwo tests are described for detecting antibody to the type-specific opacity factor (OF) of group A streptococci. This antibody was detected among patients convalescent from streptococcal sore throat in two communities in which outbreaks due to opacity factor-producing strains of group A streptococci occurred.In an outbreak due to streptococci of M-type 22 there was a close correspondence between the distribution of anti-OF and of bactericidal M-antibody for the type. In a smaller outbreak due to M-type 58 streptococci, however, M-antibody was detected more often than antibody to OF.