Genetic Diversity of Salmonella enteric serovar Typhi and Paratyphi in Shenzhen, China from 2002 through 2007

Background Typhoid and paratyphoid fever are endemic in China. The objective of this investigation was to determine the molecular features of nalidixic acid-resistant Salmonella enteric serovar Typhi (S. typhi) and Paratyphi (S. paratyphi) from blood isolates in Shenzhen, China. Results Twenty-five S. typhi and 66 S. paratyphi were isolated from 91 bacteriemic patients between 2002 and 2007 at a hospital in Shenzhen, Southern China. Fifty-two percent (13/25) of S. typhi and 95.3% (61/64) of S. paratyphi A were resistant to nalidixic acid. Sixty-seven isolates of nalidixic acid-resistant Salmonella (NARS) showed decreased susceptibility to ciprofloxacin (MICs of 0.125-1 μg/mL). All 75 NARS isolates had a single substitution in the quinolone resistance-determining region (QRDR) of GyrA (Ser83→Phe/Pro/Tyr, or Asp87→Gly/Asn), and 90.7% of these isolates carried the substitution Ser83Phe in GyrA. No mutation was found in the QRDR of gyrB, parC, or parE. Plasmid mediated quinolone resistance genes including qnr and aac(6')-Ib-cr were not detected in any isolate. Twenty-two distinct pulsed field gel electrophoresis (PFGE) patterns were observed among S. typhi. Sixty-four isolates of S. paratyphi A belonged to one clone. Eighty-seven investigated inpatients were infected in the community. Six patients infected by S. paratyphi A had a travel history before infection. Conclusions Nalidixic acid-resistant S. typhi and S. paratyphi A blood isolates were highly prevalent in Shenzhen, China. PFGE showed the variable genetic diversity of nalidixic acid-resistant S. typhi and limited genetic diversity of nalidixic acid -resistant S. paratyphi A.


Background
Typhoid and paratyphoid fever, due to infection with Salmonella enteric serovar Typhi (S. typhi) and Paratyphi (S. paratyphi), are major global problems. Nalidixic acid-resistant (NAR) S. typhi and S. paratyphi are endemic to many Asian countries [1]. NAR isolates have reduced susceptibility to fluoroquinolones, which is associated with higher rates of morbidity and mortality, particularly prolonged fever clearance time and increased need for retreatment [2]. Quinolone resistance in Salmonella is usually associated with mutations of the target site, DNA gyrase, most commonly in the quinolone resistance-determining region (QRDR) of the A subunit. Plasmid mediated quinolone resistance genes of qnr (qnrA, qnrB, qnrS, and qnrD) and aac(6')-Ib-cr has also been described in quinolone-resistant non-Typhi Salmonella [3,4]. In China, the molecular characterization and antimicrobial susceptibility of non-Typhi serotypes of Salmonella enterica have been recently reported [4], however the molecular mechanism of resistance and epidemiology of NAR S. typhi and S. paratyphi is not available. In this study we investigated the molecular basis of resistance and the epidemiology of 25 S. typhi and 66 S. paratyphi blood isolates that were recovered from hospitalized patients in Shenzhen City, Southern China over 6-year period. The cases were retrospectively examined for epidemiologic analysis.

Methods
The study site was the Shenzhen People's Hospital, a 1090-bed medical center for patients who reside in Shenzhen, Guangdong Province of Southern China, with an estimated population of 12 million people. This study has been performed with the approval of Ethics committee of Shenzhen People's Hospital (Shenzhen, China).  [5] and were interpreted according to CLSI performance standard M100-S17 [6]. The antimicrobials were supplied and stored according to the manufacturer's instructions. Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as quality control strains for susceptibility testing. Multidrug-resistant strains were defined as those resistant to ampicillin, chloramphenicol, and trimethoprim/ sulfamethoxazole (TMP-SMZ) [7].

Polymerase chain reaction (PCR) and DNA sequencing
All 91 isolates were screened for the qnr (qnrA, qnrB, and qnrS) genes by multiplex PCR [8] and for aac(6')-Ib by PCR [9]. PCR amplification of the quinolone resistance-determining regions (QRDRs) of gyrA, gyrB, parC, and parE was performed in all isolates as described previously [10]. Mutations in the gyrA, gyrB, parC, and parE genes were identified by DNA sequencing. The PCR products were purified by using a QIAquick PCR purification kit (Qiagen, Hilden, Germany). DNA sequencing of both strands was performed by the direct sequencing method with an ABI Prism 3100 generic analyzer (Applied Biosystems, Foster City, CA), and the DNA sequences of the QRDRs of gyrA, gyrB, parC, and parE were compared with the DNA sequences of the QRDRs of S. typhi, S. paratyphi A, and S. paratyphi B (GenBank: NC_004631, NC_006511, NC_010102). β-lactamase genes were detected by PCR with primers specific for bla CTX-M , bla TEM , bla SHV , bla OXA among isolates resistant to ampicillin as described previously [11][12][13], and PCR products were sequenced as described above. Class 1 intergron as was investigated by PCR. PCR products were sequenced using a pair of specific primers of 5'CS and 3'CS for multidrug-resistant isolates [14].
Pulsed field gel electrophoresis PFGE of XbaI (New England)-digested genomic DNA of all isolates was carried out using the CHEF MAPPER system (Bio-Rad), as described by the standard PulseNet protocol for Salmonella species by the Centers for Disease Control and Prevention [15]. Similarities among macrorestriction patterns were determined both by visual comparison and computer matching with BioNumerics 4.0 software. Dendrograms for similarity were built using the unweighted-pair group method using arithmetic averages. Patterns differing by zero to three fragments are considered to belong to the same PFGE type according to the method of Tenover et al [16].

Case investigation
A case was defined as illness compatible with acute typhoid or paratyphoid fever and isolation of S. typhi or S. paratyphi from a sterile site. A total of 87 cases of acute S. typhi and S. paratyphi A infections were retrospectively examined over a 6-year period; the medical records from 2 outpatients infected by S. paratyphi A were unavailable. Demographic, epidemiologic, and clinical information were recorded on case report forms that included age, sex, habitation, history of travel in the 30 days preceding illness onset, clinical symptoms and signs, laboratory data, and antimicrobial therapy. We did not include data about previous immunization against typhoid fever because it was unavailable for most of patients. Statistical analysis was performed using SPSS for Windows (release 13.0).

PCR and DNA sequencing
All 75 NARS had a single point mutation in the QRDR of gyrA that led to a single-amino-acid substitution at codons 83 or 87 of GyrA (Ser83 Phe, Ser83 Pro, Ser83 Tyr, Asp87 Gly, or Asp87 Asn) (table 3), and 90.7% (68/75) of these isolates carried the substitution Ser83Phe in GyrA. No mutation was found in the QRDR of gyrB, parC, or parE. For all 16 NASS isolates, no point mutation was detected in the QRDR of gyrA/B or parC/E gene. Plasmid-mediated quinolone resistance genes including qnr and aac(6')-Ib-cr were not detected in any isolate. The bla CTX-M-14 gene was detected in the ceftriaxone-resistant isolate of S. paratyphi A, with ISEcp1 located on the upstream of bla CTX-M-14 gene. A 1.9-kb class 1 integron gene cassette dhfrXII-orfF-aadA2 was identified in the multidrug-resistant isolate of S. paratyphi C, in which bla TEM-1 gene was also detected. None of bla CTX-M , bla TEM , bla SHV and bla OXA genes were identified in the ampicillin-resistant isolate of S. typhi.

PFGE
Overall, 22 different PFGE patterns were observed among 25 isolates of S. typhi from 2002 through 2007 (figure 1); 10 of 22 PFGE patterns were identified among 13 nalidixic acid-resistant isolates. The variable genetic diversity among S. typhi isolates suggested endemic disease from multiple sources. In contrast, among 64 isolates of S. paratyphi A, 41 isolates (including 39 NARS) were assigned to PFGE type A ( figure 2 and 3), 21 isolates (including 20 nalidixic acid-resistant isolates) belonging to subtype A1 (difference by one band of~3 10 kb compared to type A), and 2 nalidixic acid-resistant isolates to subtype A2 (difference by one band of 310 kb and one band of~190 kb compared to type A). The limited genetic diversity (similarity coefficient of 91%) among S. paratyphi A isolates indicated endemic disease from the presence of a single clone over 6-year period.

Case investigation
Infection was acquired in community in 87 patients. All patients were residents of Shenzhen City, and were mostly young or middle age and lived in sanitary environments. Six patients infected by S. paratyphi A had traveled to other cities or regions in the 30 days preceding illness onset, including Shaoguan City in Southern China (n = 1), Chongqing City and Guizhou province in Southwestern China (n = 3), Taiwan (n = 1), and Bangladesh (n = 1). More than 80% of patients (20 S. typhiinfected patients and 52 S. paratyphi A-infected patients, respectively) had received antimicrobials prior to hospital admission. They were primarily hospitalized due to fever for at least 3 days.
Epidemiological, clinical and laboratory features are presented in table 4. Clinical treatment and outcome in 23 nalidixic acid-susceptible Salmonella (NASS) and nalidixic acid-resistant Salmonella (NARS)-infected   When ceftriaxone was combined with TMP-SMZ (0.96 g PO q12h) this was shortened to 6 days during hospitalization; home therapy continued with oral antimicrobials.

Discussion
Nalidixic acid-resistant S. typhi and S. paratyphi are endemic in Vietnam and some other South Asia countries such as India, Pakistan, Bangladesh, and Nepal [17], with a resistance rate range of 38-97%. It has been reported that more than 70% of Salmonella enteric serovar Typhimurium isolates are resistant to ciprofloxacin and some have become multidrug-resistant in regions of China [4].  Table 3 The point mutation in the QRDR of gyrA of nalidixic acid-resistant Salmonella.  noted in the United States, India, Nepal, Pakistan, and Thailand [20]. A report from the United States confirmed that paratyphoid fever most often was caused by nalidixic acid-resistant S. paratyphi A, and like typhoid fever, was usually acquired while traveling internationally. In this observation, infection with S. paratyphi A was associated with travel to South and Southeast Asia, and nalidixic acid-resistant infection was associated with travel to South Asia [20]. PFGE is currently the method for the subtyping of sporadic or epidemic Salmonella isolates. By the use of a standardized PFGE protocol in this study, the Pulse-Net protocol, all isolates of S. paratyphi A were assigned to type A, subtype A1 or A2, which suggests endemic disease from the presence of a single clone over 6-year period. By investigating 62 medical records of inpatients infected by S. paratyphi A, it was confirmed that five patients infected by S. paratyphi A had traveled to other domestic cities or regions, and one had traveled internationally to Bangladesh. Our data also suggests that the same clone of S. paratyphi A was present in China over the study period.  An outbreak of paratyphoid fever associated with S. paratyphi A in New Delhi, India was investigated by PFGE [21]. The five sporadic isolates of S. paratyphi A gave PFGE patterns following XbaI digestion that were distinct, with differences of 8 to 12 bands. In contrast, the 13 outbreak isolates shared only four closely related PFGE patterns differing only in 1 to 6 bands. Similar results were obtained after digestion with a second restriction endonuclease, SpeI. In another study, a total of 39 human isolates of S. paratyphi A from Pakistan, India, Indonesia and Malaysia were typed by PFGE using XbaI restriction digests. This study suggested that a limited number of clones were responsible for paratyphoid fever in those countries [22]. Similarly, the high

Conclusions
Nalidixic acid-resistant S. typhi and S. paratyphi A blood isolates were highly prevalent in Shenzhen, China. PEGF showed the variable genetic diversity of nalidixic acid-resistant S. typhi and limited genetic diversity of nalidixic acid-resistant S. paratyphi A that suggests a clonal expansion of S. paratyphi A infection in the community.