358
w. B. REDMOND et al.
Of some importance was their observation that GS4E, BKI, and B1 gave
nearly identical lysis patterns except for two strains which showed an
intermediate degree of susceptibility to GS4E and BK1, but were highly
susceptible to phage B1. They reported additional results of considerable
significance ; the phage patterns of the drug-resistant mutants were similar
to those of the original drug-susceptible strains. Of seven strains isolated in
India, three were lower in virulence for guinea-pigs than the remaining
four, but there was no correlation between virulence and phage type in this
small sample. These workers included, among the 37 strains tested, five
obtained from infants suspected of having been infected from a single
source in an institution. All five showed similar lysis patterns. I n contrast,
six strains obtained from children whose sources of infection were thought
to be different showed different lytic patterns in five cases.
Subsequently, Bates and Mitchison (1969) used phage typing of tubercle
bacilli as a means to compare this type of marker with biochemical or
virulence characteristics of strains collected from widely separated locations
throughout the world. Previous epidemiological studies had shown that
tubercle bacilli isolated from patients in southern India were generally less
virulent than organisms isolated in Britain (Frimodt-Moller, 1975 ;
Mitchison et al., 1960; Bhatia et al., 1961) and were more susceptible to the
bactericidal action of hydrogen peroxide (Subbaiah et al., 1960) and more
resistant to thiacetazone and para-aminosalicylic acid (Selkon et al., 1960;
Joseph et al., 1964) than were British strains. Strains from Hong Kong
were more resistant to thiacetazone than British strains, but were similar to
British strains in their virulence for guinea-pigs (Dickinson et al., 1963;
Mitchison and Lloyd, 1964). T o search for a possible relationship between
phage type and any of these phenotypic characteristics, 225 strains of
tubercle bacilli obtained from patients in Britain, Hong Kong, Rhodesia,
and southern India were phage typed. The strains were isolated before drug
therapy had been given, but from 28 of these same patients additional
isolates were collected at intervals during treatment. The latter strains had
become drug resistant during treatment and this change allowed for a
study of phage type and drug resistance emerging in vivo.
Several items of epidemiological importance emerged from this study.
First, an unexpected phage type designated as “intermediate” was observed
to be harboured by patients from southern India. The term “intermediate”
was given for these strains because they showed a pattern intermediate
between types A and B in regard to their susceptibility to lysis by the
phages GS4E, BG1, and BK1. Previously all strains tested from other
geographic areas had shown a consistent pattern of either total susceptibility
or resistance to these three phages. Many of the strains from southern
India did not show this homogeneity and therefore were not clearly type A
w. B. REDMOND et al.
Of some importance was their observation that GS4E, BKI, and B1 gave
nearly identical lysis patterns except for two strains which showed an
intermediate degree of susceptibility to GS4E and BK1, but were highly
susceptible to phage B1. They reported additional results of considerable
significance ; the phage patterns of the drug-resistant mutants were similar
to those of the original drug-susceptible strains. Of seven strains isolated in
India, three were lower in virulence for guinea-pigs than the remaining
four, but there was no correlation between virulence and phage type in this
small sample. These workers included, among the 37 strains tested, five
obtained from infants suspected of having been infected from a single
source in an institution. All five showed similar lysis patterns. I n contrast,
six strains obtained from children whose sources of infection were thought
to be different showed different lytic patterns in five cases.
Subsequently, Bates and Mitchison (1969) used phage typing of tubercle
bacilli as a means to compare this type of marker with biochemical or
virulence characteristics of strains collected from widely separated locations
throughout the world. Previous epidemiological studies had shown that
tubercle bacilli isolated from patients in southern India were generally less
virulent than organisms isolated in Britain (Frimodt-Moller, 1975 ;
Mitchison et al., 1960; Bhatia et al., 1961) and were more susceptible to the
bactericidal action of hydrogen peroxide (Subbaiah et al., 1960) and more
resistant to thiacetazone and para-aminosalicylic acid (Selkon et al., 1960;
Joseph et al., 1964) than were British strains. Strains from Hong Kong
were more resistant to thiacetazone than British strains, but were similar to
British strains in their virulence for guinea-pigs (Dickinson et al., 1963;
Mitchison and Lloyd, 1964). T o search for a possible relationship between
phage type and any of these phenotypic characteristics, 225 strains of
tubercle bacilli obtained from patients in Britain, Hong Kong, Rhodesia,
and southern India were phage typed. The strains were isolated before drug
therapy had been given, but from 28 of these same patients additional
isolates were collected at intervals during treatment. The latter strains had
become drug resistant during treatment and this change allowed for a
study of phage type and drug resistance emerging in vivo.
Several items of epidemiological importance emerged from this study.
First, an unexpected phage type designated as “intermediate” was observed
to be harboured by patients from southern India. The term “intermediate”
was given for these strains because they showed a pattern intermediate
between types A and B in regard to their susceptibility to lysis by the
phages GS4E, BG1, and BK1. Previously all strains tested from other
geographic areas had shown a consistent pattern of either total susceptibility
or resistance to these three phages. Many of the strains from southern
India did not show this homogeneity and therefore were not clearly type A