Genomic identification of a novel co-trimoxazole resistance genotype and its prevalence amongst Streptococcus pneumoniae in Malawi

Abstract

Objectives: This study aimed to define the molecular basis of co-trimoxazole resistance in Malawian pneumococci under the dual selective pressure of widespread co-trimoxazole and sulfadoxine/pyrimethamine use.

Methods: We measured the trimethoprim and sulfamethoxazole MICs and analysed folA and folP nucleotide and translated amino acid sequences for 143 pneumococci isolated from carriage and invasive disease in Malawi (2002-08).

Results: Pneumococci were highly resistant to both trimethoprim and sulfamethoxazole (96%, 137/143). Sulfamethoxazole-resistant isolates showed a 3 or 6 bp insertion in the sulphonamide-binding site of folP. The trimethoprim-resistant isolates fell into three genotypic groups based on dihydrofolate reductase (encoded by folA) mutations: Ile-100-Leu (10%), the Ile-100-Leu substitution together with a residue 92 substitution (56%) and those with a novel uncharacterized resistance genotype (34%). The nucleotide sequence divergence and dN/dS of folA and folP remained stable from 2004 onwards.

Conclusions: S. pneumoniae exhibit almost universal co-trimoxazole resistance in vitro and in silico that we believe is driven by extensive co-trimoxazole and sulfadoxine/pyrimethamine use. More than one-third of pneumococci employ a novel mechanism of co-trimoxazole resistance. Resistance has now reached a point of stabilizing evolution. The use of co-trimoxazole to prevent pneumococcal infection in HIV/AIDS patients in sub-Saharan Africa should be re-evaluated.

Keywords: pneumococcal disease; prophylaxis; sulfamethoxazole; trimethoprim.

Figure 1.

Diversity of folP from Malawian pneumococci, 2003–08. (a) Annual sequence divergence of folP from pneumococci in relation to the co-trimoxazole-susceptible laboratory reference strain S. pneumoniae R6. (b) Graph showing the annual dN/dS of folP from pneumococci in relation to the co-trimoxazole-susceptible laboratory reference strain S. pneumoniae R6.

Figure 2.

Maximum likelihood phylogenetic tree based on the folP SNPs from 143 Malawian pneumococci, annotated with serotype and ST. STs represented more than twice in the study population that did not cluster on the same clade are highlighted by a symbol: STunknown22, star; ST347, triangle; Stu48, square; STunknown17, circle. Bootstrap values are shown on each branch. The scale bar represents the number of SNPs.

Figure 3.

Diversity of folA from Malawian pneumococci 2003–08. (a) Annual sequence divergence of folA from Malawian pneumococci in relation to the co-trimoxazole-susceptible laboratory reference strain S. pneumoniae R6. (b) Graph showing the annual dN/dS of folA from Malawian pneumococci in relation to the co-trimoxazole-susceptible laboratory reference strain S. pneumoniae R6.

Figure 4.

Annual distribution of three resistance genotypes amongst 137 trimethoprim-resistant Malawian pneumococci.

Figure 5.

Maximum likelihood phylogenetic tree based on the folA SNPs from 143 Malawian pneumococci, annotated with serotype and ST. Trimethoprim-susceptible isolates are highlighted with a star. Bootstrap values are shown on each branch. The scale bar represents the number of SNPs.