Intermediate Susceptibility to Ciprofloxacin among Salmonella enterica Serovar Typhi Isolates in Lima, Peru

TEXT

Typhoid fever is a systemic disease caused mainly by Salmonella enterica serotype Typhi. Although there is a decreasing incidence, it is still endemic in many countries of Latin America (1). Since multidrug-resistant (MDR) isolates (i.e., resistance to ampicillin, chloramphenicol, and trimethoprim-sulfamethoxazole) have become widespread, fluoroquinolones (e.g., ciprofloxacin) have been the first-line therapy for typhoid fever in adult patients. However, reports of clinical failure in patients treated with ciprofloxacin were described in the southeast of Asia in the 1990s (2). These isolates were resistant to nalidixic acid (NA), showing an increased MIC value of ciprofloxacin (≥0.125 μg/ml) but still within the range for susceptibility as defined for Enterobacteriaceae; this phenomenon was named “decreased ciprofloxacin susceptibility.” The presence of point mutations in the genes encoding the fluoroquinolone targets, DNA gyrase (gyrA and gyrB), and/or topoisomerase IV (parC and parE), was detected among them (3). Several studies subsequently emphasized the need of changing the breakpoints for fluoroquinolones for Salmonella (4); as a consequence, the Clinical and Laboratory Standards Institute changed the breakpoints as follows: (i) susceptible, ≤0.06 μg/ml; (ii) intermediate susceptible, 0.12 to 0.5 μg/ml; (iii) resistant, ≥1 μg/ml (5).

The emergence of S. Typhi showing intermediate susceptibility to ciprofloxacin (ISC) has been almost unstudied in Latin America. Our objective was to assess the susceptibility pattern of S. Typhi from Peruvian patients, to describe the molecular characteristics of resistance to fluoroquinolones, and to analyze their clonal relationship. Blood culture isolates were collected as part of a microbiological surveillance study from 2008 to 2012 through a network of eight public hospitals in Lima, Peru. Blood cultures were drawn during routine patient care. Each hospital followed its own protocol for blood culturing. Of 2,897 Gram-negative isolates, 154 (5.3%) belonged to the genus Salmonella. Only one isolate per patient was considered. Lactose-negative colonies with typical biochemical characteristics of S. Typhi were further confirmed by agglutination with Salmonella O9, Vi-specific and H:d antisera.

Susceptibility testing was performed by disk diffusion for ampicillin, ceftriaxone, chloramphenicol, tetracycline, and trimethoprim-sulfamethoxazole (6). MIC values for NA and ciprofloxacin were determined by agar dilution (5). Escherichia coli strain ATCC 25922 was used as a control for susceptibility testing.

Screening for mutations in the quinolone resistance-determining region (QRDR) was performed by amplification of a fragment of the gyrA, gyrB, and parC genes containing the QRDR as previously described (6), and the fragments were sequenced on an ABI 3130 sequencer. The presence of the plasmid-mediated quinolone resistance qnr genes (qnrA, qnrB, and qnrS) was determined using PCR (7). The alignment was done using the following strains: Salmonella enterica serovar Typhimurium LT2, ST CT18, and ST Ty2.

Clonal relationships were assessed by pulsed-field gel electrophoresis (PFGE) according to the PulseNet Europe protocol (8). Genomic DNA was digested with the XbaI restriction enzymes; Salmonella Braenderup H9812 was used as a size marker. For cluster analysis, BioNumerics 5.1 was used. The cluster analysis settings applied were the Dice similarity coefficient and unweighted-pair group method using average linkages (UPMGA) dendrogram type with the following parameters: (i) optimization of 0.50% and (ii) position tolerance of 1.50%.

A total of 33 isolates were recovered, and 11 cases (33.3%) were children (<14 years). All were fully susceptible to trimethoprim-sulfamethoxazole, chloramphenicol, ceftriaxone, ampicillin, and tetracycline, except two which showed intermediate susceptibility to ampicillin. Six isolates (18.2%) were resistant to NA and eight (24.2%) showed ISC (Table 1). Most previous studies in Latin American countries focused on Salmonella non-Typhi (9). Over the last 20 years, only two studies assessing S. Typhi were found, reporting full susceptibility of 151 S. Typhi isolates from Colombia and Argentina. Of note, disk diffusion was the method used in both studies. Since MIC values were not determined, it is possible that isolates with ISC were missed (10, 11).

A mutation in gyrA, Gly133-Glu, was detected in all ISC isolates when S. Typhimurium LT2 was used for alignment. However, it was not detected when S. Typhi CT18 and S. Typhi Ty2 were used. This is very likely a serotype-specific mutation. It is present in isolates susceptible to NA irrespective of their MIC values for ciprofloxacin. It has been rarely described in Asian countries but was recently found to be the most common mutation in the Democratic Republic of the Congo (5, 12–14). Six isolates had a second mutation in the gyrA gene, Ser83-Phe (n = 3) and Asp87-Asn (n = 3). These mutations are widespread in S. Typhi from Asian countries, including India (4, 12). One study which included 12 NA-resistant S. Typhi isolates from travelers found that seven isolates carried the Ser83-Phe mutation; interestingly, six travelers came from India and one from Peru (13). In addition, the two remaining isolates had a mutation in the gyrB gene, Ser464-Phe (Table 1). This mutation has been associated with the nonclassical fluoroquinolone resistance phenotype, as both isolates were susceptible to NA; as a consequence ISC is not predicted by the detection of resistance to NA. Such isolates are typically non-MDR; they have been described from France, the United States, and England, but they were primarily acquired in India (15). No mutations were found in the parC and qnr genes.

PFGE revealed a total of 28 different pulsotypes among the 33 isolates, with a maximum of two isolates sharing the same pulsotype (Fig. 1). This high diversity has been reported by Salve and Pichel in Argentina and Colombia (11).

This study has several limitations. A small number of isolates were included, but they were obtained during a 5-year multicenter laboratory surveillance. Besides, we did not collect clinical information regarding treatment follow-up.

Among 33 S. Typhi isolates, none displayed MDR, but ISC was demonstrated in almost a quarter of them. All ISC isolates showed a mutation in gyrA (Gly133-Glu), combined with either a mutation in gyrA at codons 83 and 87 or in the gyrB gene at codon 464. A diverse clonal distribution was demonstrated. Since fluoroquinolones are still the cornerstone of the treatment of typhoid fever in Latin American countries, active surveillance is needed in order to monitor resistance trends to these antimicrobials.