Abstract
Fifty clinical isolates (penicillin MICs, 0.03-8 μg/mL) of
Streptococcus pneumoniae
were randomly selected from hospitals throughout South Africa, together with seven strains isolated in Hungary
(penicillin MICs, 16-32 μg/mL). Penicillin-binding protein (
pbp
)
1b
and
2a
genes were amplified by PCR, and the purified DNA was digested with
Hinf
I,
Sty
I, and
Mse
I
+
Dde
I restriction enzymes. The fragments were radioactively endlabeled and separated on polyacrylamide gels, and the DNA fingerprints were visualized following autoradiography. A collection of isolates
was further selected for sequence analysis of
pbp1b
and
2a
. DNA fingerprint analysis revealed a uniform profile amongst all isolates for both genes. All isolates revealed a maximum of only
seven nucleotide substitutions in their
pbp1b
genes, resulting in a maximum of three amino acid substitutions in PBP 1B. In the case of the
pbp2a
gene, up to 13 nucleotide substitutions were
observed randomly distributed amongst penicillin-susceptible and resistant isolates, revealing a maximum of five amino acid substitutions in PBP 2A. No amino acid substitutions were found to be common amongst
all penicillin-resistant isolates. Transformation experiments with
pbp1b
and
2a
genes isolated from two resistant strains (MICs, 4 and 16 μg/mL) failed to transform pneumococcal strains
to increased levels of penicillin resistance. These results show that the
pbp1b
and
2a
genes examined here do not display the typical mosaic gene patterns observed in the
pbp2x
,
2b
,
and
1a
genes of penicillin-resistant pneumococci. In addition, the transformation studies suggest that PBPs 1B and 2A may not play a role in the development of penicillin resistance in some pneumococci.