MICROBIAL DRUG RESISTANCE

Volume 10, Number 2, 2004

© Mary Ann Liebert, Inc.

Reduced Susceptibility to Ciprofloxacin and gyrA Gene

Mutation in North Indian Strains of Salmonella enterica
Serotype Typhi and Serotype Paratyphi A

K. RENUKA,1 A. KAPIL,1 S.K. KABRA,2 N. WIG,3 B.K. DAS,1 V.V.S.P. PRASAD,1

R. CHAUDHRY,1 and P. SETH1

ABSTRACT

The emergence of reduced susceptibility to ciprofloxacin among Salmonella enterica serotype Typhi and
serotype Paratyphi A leading to clinical failure of treatment poses a great therapeutic challenge. The mech-
anism of fluoroquinolone resistance in clinical isolates of S. Typhi and S. Paratyphi A is not very well docu-
mented. The present study was carried out with the objective of molecular characterization of reduced
quinolone susceptibility amongst the strains of S. Typhi and S. Paratyphi A isolated from the patients with
enteric fever during January, 2000, to April, 2003, in a North Indian hospital. A total of 422 culture-positive
cases of enteric fever were reported to the hospital during the period of study, of which S. Typhi was isolated
from 350 cases and S. Paratyphi A from 72 cases. The antimicrobial susceptibility of these strains was deter-
mined by disk diffusion and agar dilution method according to NCCLS guidelines, and E-test method. A to-
tal of 140 randomly selected strains, isolated during the years 1993–1999, that were available from the labo-
ratory stocks were also studied to compare with the present strains. To study the quinolone susceptibility, the
strains were divided into nalidixic acid sensitive (NAS), nalidixic acid intermediate resistant, (NAI) and
nalidixic acid resistant (NAR) on the basis of susceptibility to nalidixic acid. Clinical history was available
from 174 patients, of which 93 needed hospitalization due to severe disease. Of these, 82 patients were infected
with NAR strains and 22 patients had a documented evidence of clinical failure to ciprofloxacin therapy. The
patients infected with NAR strains were younger and had a significantly longer duration of fever (p value 
0.05) than those infected with NAS strains. It was observed that the proportion of NAR strains increased grad-
ually over the years. These strains had a significantly higher range of MIC of ciprofloxacin (0.023–1.0 g/ml)
as compared to the NAS strains (0.002–0.125 g/ml) (p value  0.05). The sequencing of quinolone resistance-
determining region (QRDR) of the gyrA gene showed the presence of mutation at either Ser 83 or at Asp 87
in all the NAR and NAI strains. None of the NAS strains had a mutation, suggesting that the gyrA gene mu-
tation is sufficient to confer resistance to nalidixic acid and reduced susceptibility to ciprofloxacin. This mu-
tation, although phenotypically expressed as decreased susceptibility to ciprofloxacin, goes undetected by the
disk diffusion method using the present NCCLS guidelines. Hence, it can increase morbidity and mortality
due to delay in appropriate antibiotic treatment.

INTRODUCTION of cases.33 Even after the availability of effective antimicrobial

agents, the treatment of enteric fever continues to pose thera-

E NTERIC FEVER remains a major public health problem in de-

veloping countries, including India, with the Salmonella
enterica serotype Typhi (S. Typhi) responsible for a majority
of the cases, followed by Salmonella enterica serotype Paraty-
phi A (S. Paratyphi A), which causes approximately 20–25%
peutic challenge. With the emergence of chloramphenicol-re-
sistant strains,19 fluoroquinolones such as ciprofloxacin became
the antibiotic of choice for the treatment of enteric fever.2,9,23
However, S. Typhi isolates with reduced susceptibility22,24
or resistance to fluoroquinolones31,36 has been documented for

Departments of 1Microbiology, 2Pediatrics, and 3Medicine, All India Institute of Medical Sciences, New Delhi, India.

146
REDUCED SUSCEPTIBILITY TO CIPROFLOXACIN IN S. TYPHI AND S. PARATYPHI A 147

nearly a decade. In the recent years, workers in several coun- Antimicrobial susceptibility testing
tries have reported treatment failure after administration of
Antimicrobial susceptibility testing was performed using the
ciprofloxacin therapy in patients of enteric fever caused due
following methods:
to strains of Salmonella with reduced susceptibility to
ciprofloxacin.20,22,31,35–37 Resistance to quinolones commonly
1. Disk diffusion method according to NCCLS guidelines,25
arises via mutations in the genes encoding DNA gyrase (gyrA
using disks containing chloramphenicol (30 g), amoxicillin
and gyrB) and DNA topoisomerase IV (parC and parE).5 How-
(10 g), cotrimoxazole (1.25/23.75 g), ceftriaxone (30 g),
ever in Salmonella spp., mutations associated with quinolone
nalidixic acid (30 g), and ciprofloxacin (5 g) (HiMedia
resistance are mostly present in the QRDR of the gyrA gene.6,16
Laboratories Limited, Mumbai, India). E. coli ATCC (Amer-
The most commonly reported mutation is at Ser 83, which also
ican Type Culture Collection) strain 25922 was used as the
has a site for HinfI restriction.6
quality control.
A major problem in the identification of S. Typhi strains with
2. MIC of ciprofloxacin was determined by:
reduced susceptibility to ciprofloxacin is that all the S. Typhi
a. Agar dilution method (NCCLS guidelines) and
strains fall within the susceptible range using the present
b. E-test strips (AB Biodisk, Sweden) according to manu-
NCCLS breakpoints for determining susceptibility by disk dif-
facturer’s instructions.
fusion as well as MIC.25 To avoid the delay in appropriate ther-
apy and decrease the associated morbidity and mortality of en-
The strains were divided into the following categories on the
teric fever, there is a need to identify the strains likely to cause
basis of antimicrobial susceptibility for the analysis of the re-
an infection not responding to ciprofloxacin therapy. We had
sults:
earlier studied nalidixic acid susceptibility as a possible marker
for detection of reduced susceptibility to ciprofloxacin where
1. MDR. The strains resistant to chloramphenicol, amoxicillin,
we found that the ciprofloxacin MIC in nalidixic acid-resistant
and cotrimoxazole.
strains (NARST) was about 10-fold higher than that of nalidixic
2. NAS. Nalidixic acid-sensitive strains (inhibition zone diam-
acid-sensitive strains (NASST).21
eter  19 mm).
To investigate the genetic basis of quinolone resistance, we
3. NAI. Nalidixic acid partially sensitive strains (inhibition
carried out a prospective study in which the antimicrobial re-
zone diameter  14–18 mm).
sistance and molecular characteristics of S. Typhi and S. Paraty-
4. NAR. Nalidixic acid-resistant strains (inhibition zone diam-
phi A strains isolated from the patients of enteric fever during
eter  13 mm).
the years 2000–2003 were studied and compared with the ret-
5. Reduced ciprofloxacin susceptibility. Strains with MIC of
rospective study carried out on the S. Typhi and S. paratyphi A
ciprofloxacin  0.19 g/ml.
strains isolated during 1993–1999, which were available from
the laboratory stocks.
Study of gyrA gene mutations
A 630-bp fragment of the gyrA gene containing the QRDR
MATERIALS AND METHODS was amplified using the primers corresponding to nucleotide
positions 1–30 and 601–630 (codons 1–210) of the E. coli gyrA
Bacterial strains gene sequence (Accession no. X06373).
All the bacterial strains (S. Typhi and S. Paratyphi A) stud-
Preparation of template DNA
ied were isolated from the blood culture samples received from
the patients of suspected enteric fever in the clinical bacteriol- A single bacterial colony from an overnight grown culture
ogy laboratory of the Department of Microbiology at the All was suspended in 100 l of sterile distilled water [E Merck (In-
India Institute of Medical Sciences (AIIMS), New Delhi, India. dia), Limited, Mumbai] and boiled at 100°C for 10 min. The
All the isolates were identified by using specific antisera (aH, suspension was centrifuged at 5300  g for 10 min. after adding
dH, O9, and O4) (Murex Diagnostics Ltd, UK) and standard an equal volume of chloroform:isoamyl alcohol (24:1) solution.
biochemical tests27 and were stored in glycerol stocks at 70°C. The supernatant containing bacterial DNA was used as tem-
All the 422 strains of S. Typhi and S. Paratyphi A isolated plate for PCR.
during January, 2000, to April, 2003, from the patients with
suspected enteric fever were included. For comparison, 140 ran- Amplification of gyrA gene
domly selected strains previously isolated during 1993–1999
Template DNA prepared from bacterial strains as described
were included that were available from the laboratory stocks
above was amplified by PCR with use of oligonucleotide pri-
from out of a total of 448 culture-positive cases during these
mers: 5-ATGAGCGACCTTGCGAGAGAAATTACACCG-3
years. It was ensured that an equal proportion for each year was
and 5-TTCCATCAGCCCTTCAATGCTGATGTCTTC-3 ob-
included.
tained from Biobasic (Canada). The PCR was performed by
using Expand Long Template PCR system according to manu-
Patient
facturer’s (Roche Diagnostics GmbH, Germany) instructions in
Of the 422 culture positive cases of enteric fever during Jan- a final reaction volume of 50 l containing 5 l of 10 Poly-
uary, 2000, to April, 2003, clinical history was available for merase buffer 1, 2.65 U of Expand DNA polymerase, 80 µM
174 patients. The remaining patients visited the outpatient de- dNTPs, 5pm of each primer, and 5 l of template DNA. An Am-
partment only once and were lost to follow up. plitron Thermolyne (USA) thermocycler was used for amplifi-
148 RENUKA ET AL.

cation. The program for amplification included a step of initial
denaturation at 95°C for 2 min, followed by 35 cycles of 94°C
for 30 sec, 70°C for 60 sec and 72°C for 90 sec and a final ex-
tension step at 72°C for 5 min. The PCR product was loaded in
1% wt/vol agarose gel (LifeTechnologies, GibcoBRL, Scotland)
prepared in Tris-borate-EDTA buffer and detected by ethidium
bromide staining after electrophoresis (BioRad, USA). The PCR
product was precipitated with ethanol and 0.3 M sodium acetate
and stored at 20°C until used for restriction and sequencing.
HinfI restriction
The amplified 630-bp fragment of the gyrA gene has three
HinfI restriction sites, one of which lies at Ser 83. Therefore,
HinfI restriction digestion of the PCR product was done to de-
tect mutation at Ser 83. A volume of 20 l of the precipitated
PCR product was digested with 10 units of HinfI (Promega,
USA) at 37°C. The restriction fragments were run on 2%
agarose gel (LifeTechnologies, GibcoBRL, Scotland) prepared
in Tris-borate-EDTA buffer and detected by ethidium bromide
staining after electrophoresis (BioRad, USA).
DNA sequencing
The gyrA gene amplified by PCR was sequenced for 52 S.
Typhi and 12 S. paratyphi A strains. These were selected on
the basis of ciprofloxacin MIC so that all the MIC levels from
0.003 to 1.0 g/ml were equally represented. Sequencing was
performed with ABI Prism Big Dye Terminator Cycle Se-
quencing Ready Reaction Kit (Perkin-Elmer Applied Biosys-
tems, Foster City, CA) using the same primers as were used for
PCR. Sequencing was performed with both forward and reverse
primers on GeneAmp 2400 DNA Thermal Cycler and analyzed
in an automatic DNA sequencer ABI PRISM 310 Genetic An-
alyzer (Perkin-Elmer Applied Biosystems).
Data analysis
The clinical features and response to ciprofloxacin treatment
in patients infected with NAS or NAR isolates were compared
by the Wilcoxon rank sum test or Fisher’s exact test. The as-
sociation between nalidixic acid susceptibility and MIC of
ciprofloxacin was determined by using independent t-test. The
result was considered significant at 5% significance level, i.e.,
p value  0.05. The statistical package SAS 8.0 was used for
these analyses. For nucleotide sequence analysis, alignment was
done by using ClustalX 1.8134 and the Genedoc Multiple se-
quence alignment editor and shading utility version 2.6.00226
and sequences were compared with the sequence of S. Typhi
strain CT18 (accession no. AL627274). The codon numbers
were designated on the basis of alignment of translated E. coli
gyrA sequence (accession no. X06373) with a portion of trans-
lated Salmonella sequence.

TABLE 1. CLINICAL FEATURES AND RESPONSE TO CIPROFLOXACIN TREATMENT IN

HOSPITALIZED PATIENTS INFECTED WITH NAS OR NAR Salmonella STRAINS

NAS NAR
Variable (n  11) (n  82) p value

Sex ratio (number of males 65 5527

number of females)
Median age in years (IR, AR) 25 (12.5–28.5, 4–60) 13 (7–21, 0.5–58) 0.05
Median duration of fever in days 11 (9.5–13, 5–20) 17.5 (13.25–26, 8–90) 0.05
(IR, AR)
Median duration of hospitalization 12 (7–14.5, 4–28) 10 (7–15.75, 6–65) 0.05
in days (IR, AR)a
Number (%) of patients with 2 (18) 32 (39) 0.05
hepatomegaly
Number (%) of patients with 2 (18) 29 (35) 0.05
splenomegaly
Number (%) of patients with 1 (9) 7 (8) 0.05
encephalopathy
Number (%) of patients with 0 5 (6) 0.05
anemia
Number (%) of patients with 0 4 (5) 0.05
lymphadenopathy
Number (%) of patients with other 3 (27) 18 (22) 0.05
complicationsb
Number (%) of patients infected 10 (91) 69 (84) 0.05
with S. Typhi
Number (%) of patients infected 1 (9) 13 (16) 0.05
with S. Paratyphi A

AR, Absolute range; IR, interquartile range.

aThese analyses do not include the 4 patients who were infected with NAR and died.
bIncluded are acute renal failure, myocarditis, pancreatitis, pleural effusion, peritonitis, oliguria, pneumonia, ascitis, enteric

perforation, polymyositis, gastrointestinal hemorrhage, and deep vein thrombosis.

REDUCED SUSCEPTIBILITY TO CIPROFLOXACIN IN S. TYPHI AND S. PARATYPHI A 149

TABLE 2. NUMBER OF MDR, NAS, NAI, AND NAR STRAINS between the patients infected with NAS or NAR strains in terms
INEACH YEAR INCLUDED IN THE STUDY of these complications.
Total number of strains
Response to treatment
Year Studied MDR NAS NAI NAR Eighty-seven patients (55 hospitalized and 32 OPD) needed
treatment with i.v. ceftriaxone or oral third-generation cephalo-
1993 19 7 19 0 0
sporins due to the clinical failure of first-line antityphoidal
1994 20 12 16 0 4
1995 11 2 10 0 1 drugs. A history of documented failure to ciprofloxacin ther-
1996 23 7 18 0 5 apy was available in a total of 31 (22 hospitalized and 9 OPD)
1997 16 3 8 0 8 patients, whereas the other patients did not know the details of
1998 18 1 9 0 9 the initial treatment provided by the private practitioner. Four
1999 33 9 15 1 17 of the admitted patients, including 3 children suffering from ty-
2000 113 43 31 3 79 phoid fever and 1 adult suffering from paratyphoid fever, died.
2001 153 76 18 1 134 All of these patients were infected with NAR strains.
2002 111 29 24 0 87
Until April 2003 45 12 5 0 40
Bacterial strains and antimicrobial susceptibility
Total 562 201 173 5 384
Antimicrobial susceptibility was performed on 422 isolates
of S. Typhi and S. Paratyphi A during the period of study
(January, 2000, to April, 2003). Also, 140 strains available
from laboratory stocks of the isolates of the years 1993–1999
RESULTS
were included for comparison. During these years, the clinical
practice and culture methods remainined the same; there was
Clinical data
an overall increase in the number of culture-positive cases
During the period of the study, the total number of blood during January, 2000, to April, 2003 (422), versus January,
cultures performed for suspected enteric fever cases were 1993, to December, 1999 (448). Thus, a total of 562 strains
66,555, of which 422 were positive for enteric fever pathogens. were studied for their antimicrobial susceptibility by the disk
These included 350 S. Typhi and 72 S. Paratyphi A. No S. diffusion method. These included 472 strains of S. Typhi and
Paratyphi B and S. Paratyphi C were isolated. Of these 422 cul- 90 strains of S. Paratyphi A. No S. Paratyphi B and C were
ture-positive patients, 289 were males and 133 were females. isolated.
The age distribution of these patients was: 0–5 years, 81 pa- Of the 472 S. Typhi strains, 156 (33%), 312 (66%), and 4
tients; 6–15 years, 163 patients; 15 years, 178 patients. (1%) strains were NAS, NAR, and NAI, respectively. Of the
The detailed clinical data was available from 175 patients 90 S. Paratyphi A strains, 17 (19%), 72 (80%), and 1 (1%)
only. Of these, 82 patients were treated as outpatients and 93 strains were NAS, NAR, and NAI, respectively. A total of 201
patients needed hospitalization due to severe disease. strains were MDR of which 183 (91%) were S. Typhi and 18
Of the 82 patients treated on an outdoor basis, 60 were ty- (9%) were S. Paratyphi A. Amongst the MDR, 50 (25%), 147
phoid and 22 were paratyphoid cases. Twelve patients were in- (73%), and 4 (2%) strains were NAS, NAR, and NAI, respec-
fected with NAS and 70 patients were infected with NAR tively. The number of MDR, NAS, NAR, and NAI strains in-
strains. None of these patients had a severe disease except for cluded in the study from each year is shown in Table 2.
the 3 patients who had anemia. There was no significant dif- The MIC of ciprofloxacin was determined for a total of 270
ference in the clinical presentation among patients infected with strains by using E-test strips. All the isolates were found to be
NAS or NAR strains. sensitive to ciprofloxacin (zone diameter  21 mm and MIC 
Hospitalization was required in 93 patients, of which 79 con- 1.0 g/ml) according to NCCLS guidelines. The ciprofloxacin
stituted typhoid and 14 paratyphoid cases. Eleven patients were MIC values among NAR strains was about 10-fold higher than
infected with NAS and 82 patients were infected with NAR NAS strains tested, although there was some overlap between
strains. The demographic and clinical features of the hospital- the two groups, and ranged from 0.023 to 1.0 g/ml (Table 3),
ized patients are shown in Table 1. There was a significant dif-
ference in the age of the patients infected with the NAS or NAR
strains, with patients infected with NAR strains being younger TABLE 3. MIC OF CIPROFLOXACIN
(p value  0.05). Also, the median duration of fever among pa- OF NAS AND NAR ISOLATES
tients infected with NAR strains was significantly longer than
those among patients with NAS strains, 17.5 (interquartile MIC (g/ml) values of NAS NAR
range, 13.25–26 days; absolute range 5–90 days) versus 11 days ciprofloxacina (n  92) (n  173)
(interquartile range, 9.5–13 days; absolute range 5–20 days),
Mean SD 0.023 0.032 0.194 0.136
respectively (p value  0.05). Complications such as severe
Median 0.008 0.19
anemia, encephalopathy, acute renal failure, myocarditis, peri- Absolute range 0.002–0.125 0.023–1.0
tonitis, ascitis, oliguria, deep vein thrombosis, pleural effusion, Interquartile range 0.006–0.023 0.125–0.25
and enteric perforation were found in 30 of the typhoid fever
and 6 of the paratyphoid fever cases. There was no difference ap value  0.05.
150 RENUKA ET AL.

TABLE 4A. MIC OF CIPROFLOXACIN AND MUTATIONS IN THE GYRA GENE OF Salmonella enterica SEROTYPE TYPHI

Isolate and strain number/

year of isolation MIC of ciprofloxacin Mutation Codon change

NAR (n  28)
2508/1994 0.125 Asp 87 → Asn GAC → AAC
8299/1994 0.19 Ser 83 → Phe TCC → TTC
8603/1995 0.094 Ser 83 → Phe TCC → TTC
5531/1997 0.125 Asp 87 → Asn GAC → AAC
8320/1997 0.25 Asp 87 → Tyr GAC → TAC
6943/1998 0.19 Ser 83 → Tyr TCC → TAC
6911/1998 0.094 Ser 83 → Phe TCC → TTC
7437/1999 0.125 Ser 83 → Phe TCC → TTC
7810/1999 0.047 Ser 83 → Phe TCC → TTC
13429/1999 0.064 Ser 83 → Phe TCC → TTC
15283/1999 0.5 Ser 83 → Phe TCC → TTC
6832/2000 0.032 Ser 83 → Phe TCC → TTC
7139/2000 0.094 Ser 83 → Phe TCC → TTC
10672/2000 0.19 Ser 83 → Phe TCC → TTC
4647/2001 0.125 Ser 83 → Phe TCC → TTC
6142/2001 0.094 Asp 87 → Asn GAC → AAC
6487/2001 0.19 Ser 83 → Phe TCC → TTC
8695/2001 0.25 Ser 83 → Phe TCC → TTC
9788/2001 0.19 Ser 83 → Tyr TCC → TAC
10240/2001 0.19 Ser 83 → Phe TCC → TTC
18196/2001 0.125 Ser 83 → Tyr TCC → TAC
17728/2001 0.19 Ser 83 → Tyr TCC → TAC
18327/2001 0.75 Ser 83 → Tyr TCC → TAC
18473/2001 1.0 Ser 83 → Tyr TCC → TAC
1843/2002 0.38 Ser 83 → Tyr TCC → TTC
18830/2002 1.0 Ser 83 → Phe TCC → TTC
10097/2002 1.0 Ser 83 → Phe TCC → TTC
18564/2002 1.0 Ser 83 → Phe TCC → TTC
NAI (n  4)
13241/1999 0.047 Ser 83 → Tyr TCC → TAC
7875/2000 0.047 Asp 87 → Asn GAC → AAC
16213/2000 0.064 Asp 87 → Asn GAC → AAC
1237/2000 0.064 Asp 87 → Asn GAC → AAC
NAS (n  20)
1918/1993 0.094 Nil —
4388/1993 0.004 Nil —
7389/1994 0.023 Nil —
10804/1995 0.064 Nil —
3238/1996 0.008 Nil —
6649/1996 0.012 Nil —
4288/1998 0.008 Nil —
7661/1998 0.064 Nil —
13082/1999 0.003 Nil —
16108/1999 0.125 Nil —
6994/2000 0.023 Nil —
9618/2000 0.004 Nil —
2052/2001 0.006 Nil —
2349/2001 0.016 Nil —
6878/2001 0.008 Nil —
7025/2001 0.006 Nil —
12956/2001 0.125 Nil —
17816/2001 0.125 Nil —
7616/2002 0.016 Nil —
8731/2002 0.032 Nil —
REDUCED SUSCEPTIBILITY TO CIPROFLOXACIN IN S. TYPHI AND S. PARATYPHI A 151

TABLE 4B. MIC OF CIPROFLOXACIN AND MUTATIONS IN THE GYRA GENE OF Salmonella enterica SEROTYPE PARATYPHI A
Isolate and strain number/
year of isolation MIC of ciprofloxacin Mutation Codon change

NAR (n  11)
9528/1996 0.38 Asp 87 → Gly GAC → GGC
4704/2000 0.38 Ser 83 → Phe TCC → TTC
1806/2001 0.38 Ser 83 → Phe TCC → TTC
1994/2001 0.25 Ser 83 → Phe TCC → TTC
2542/2001 0.19 Ser 83 → Phe TCC → TTC
4869/2001 0.19 Ser 83 → Phe TCC → TTC
5078/2001 0.25 Ser 83 → Tyr TCC → TAC
6662/2001 0.19 Ser 83 → Phe TCC → TTC
11559/2001 0.25 Asp 87 → Asn GAC → AAC
17842/2001 0.38 Ser 83 → Phe TCC → TTC
3138/2001 1.0 Ser 83 → Phe TCC → TTC
NAS (n  1)
5740/1997 0.047 Nil —

with the higher recorded MIC values detected in NAR strains
isolated after the year 2000. Using our cut off of MIC value 
0.19 g/ml, a total of 91 strains showed reduced susceptibility
to ciprofloxacin, of which 61 were S. Typhi and 30 were S.
Paratyphi A strains. The MIC determined by agar dilution
method correlated with E-test results.
PCR and nucleotide sequence analysis
A 630-bp fragment containing the QRDR of gyrA gene was
obtained after PCR of all the strains. The gyrA gene product
was sequenced from a total of 64 strains, including 52 strains
of S. Typhi and 12 strains of S. Paratyphi A, isolated in dif-
ferent years and representing different MIC levels (Table 4).
These included 39 NAR, 21 NAS, and 4 NAI strains. All the
NAR and NAI strains had a mutation in the QRDR of gyrA,
whereas none of the NAS isolates had a mutation in that
region. Five different kinds of point mutations found were:
Ser 83  Phe in 25 strains, Ser 83  Tyr in 9 strains, Asp
87  Asn in 7 strains, Asp 87  Tyr, and Asp 87  Gly in
1 strain each. The sequence data are shown in Table 4. In a
ddition to these mutations, a silent mutation at Ser 27 (TCT 
TCG) was found in seven of the NAR strains. A single base
difference at the nucleotide position equivalent to 398 of E.
coli gyrA (accession no. X06373) gave rise to codon GAA in
place of codon GGA in five of the strains (including one NAS)
leading to amino acid change Gly 133  Glu. This change
has earlier been reported to be present in S. Typhimurium
NCTC 74.37
HinfI restriction fragment length polymorphism in gyrA gene
product
The enzyme HinfI cuts the 630-bp gyrA gene product at three
sites. Therefore, the gyrA gene product from NAS strains
showed HinfI restriction pattern consisting of four fragments of
sizes 244 bp, 149 bp, 138 bp, and 99 bp. In 33 NAR and 1 NAI
strain, one of the HinfI restriction site at Ser 83 was abolished
due to mutation at this site. Therefore, the gyrA gene product
of these strains was cut into three fragments of sizes 343 bp,
149 bp, and 138 bp. In the remaining 6 NAR and 3 NAI strains
having a mutation at Asp 87, HinfI restriction pattern was same
as that of NAS strains.
Nucleotide sequence accession numbers
The partial DNA sequences of the gyrA gene of S. enterica
serotype Typhi and serotype Paratyphi A tested in this study
were registered in the GenBank nucleotide sequence database
under the accession numbers starting from AY302575 to
AY302589.
DISCUSSION
After the emergence of multidrug-resistant strains of S. Ty-
phi, quinolones became the drug of choice for treatment of en-
teric fever.2,9,23 However, during the last 10 years, several treat-
ment failures with ciprofloxacin were reported, along with
several reports of S. Typhi strains exhibiting decreased sus-
ceptibility to ciprofloxacin.20,22,31,35,36,37 This has been re-
sponsible for the delay in the effective treatment and increased
morbidity and mortality due to enteric fever.
During the period of study, a total of 66,555 blood cultures
were performed for patients presenting with fever, of which 422
were positive for enteric fever pathogens. In our patients, we
found that S. Typhi and S. Paratyphi A were responsible for
83% and 17% of cases, respectively, of enteric fever during
January 2000, to April, 2003. Out of 175 patients where clini-
cal data was available, 93 patients needed hospitalization due
to severe disease, although they were on one or the other ther-
apy at the time of seeking consultation at this tertiary care hos-
pital. Of these, 82 patients were infected with NAR strains and
11 were infected with NAS strains. There was no difference in
the patients infected with NAS or NAR strains in terms of clin-
ical features, except that the patients infected with NAR strains
were younger and had a significantly longer median duration
of fever as compared with patients infected with NAS strains.
The same kind of observation was made in a study carried out
on short course treatment of typhoid fever, where median fever
clearance times among patients infected with NAR strains were
twice as long as those among patients infected with NAS
strains.37
The median duration of hospitalization amongst the patients
infected with NAS or NAR strains was almost the same (12
152
days and 10 days, respectively) as majority of these patients
were put on i.v. ceftriaxone treatment after hospital admission
and were discharged only after completing the treatment. Most
of our patients are illiterate and hence could not give the de-
tails of the antibiotic taken as the first line in the community
but at least 31 patients could give a documented failure to
ciprofloxacin. All these patients were infected with NAR
strains. Four of our hospitalized patients died due to complica-
tions related to enteric fever, although they were treated with
ceftriaxone, which could be due to delay in recognizing non-
response to ciprofloxacin therapy. All of these patients were in-
fected with NAR strains, although the routine disk diffusion
susceptibility test in the clinical microbiology laboratory re-
ported these isolates as ciprofloxacin sensitive. Previous reports
have also described failure to ciprofloxacin therapy in patients
infected with strains exhibiting resistance to nalidixic acid
and/or reduced susceptibility to ciprofloxacin.35,37
We had earlier reported an increase in the incidence of S.
Paratyphi A infection in North India.33 In our study, 80% of S.
Paratyphi A strains were NAR as compared to 66% of S. Ty-
phi strains. Also the ciprofloxacin MIC values were higher in
S. Paratyphi A as compared to S. Typhi. Of the four deaths men-
tioned above, one patient had paratyphoid fever. We also ob-
served that S. Paratyphi A, which has been associated with the
mild illness,30 is also responsible for failure to ciprofloxacin
therapy and severe disease. This observation may be important
in designing vaccine strategies against enteric fever.
In our study, we have found that the proportion of S. Typhi
strains resistant to nalidixic acid has increased over the years
from 1993 to 2002 (Table 2). Also, with the culture methods
and clinical practices remaining the same in the hospital the
number of culture positive cases has also increased (422
during January, 2000, to April, 2003, versus 466 during
1993–1999). It was also noticed that partial sensitivity to
nalidixic acid was found in only five of the isolates, which was
during the intervening period when NAR strains were increas-
ing in number. This points to a possibility of a stepwise increase
in the quinolone resistance under the antibiotic selective pres-
sure. In the Gram-negative organisms, mutations conferring
resistance to quinolones occurs primarily in DNA gyrase in a
stepwise manner.5,7,14,17,29 In Salmonella, the most common
residues associated with mutations leading to quinolone resis-
tance have been Ser 83 and Asp 87 in the gyrA gene, either
alone or together.6,11,16,32
Mutations in both gyrA and parC have been identified in bac-
terial isolates highly resistant to fluoroquinolones.8,10,28 parC
gene mutation have been identified in animal isolates of S. Ty-
phimurium,4 but have not yet been found in the clinical isolates
of S. Typhi or S. Paratyphi A.4
Out of 39 NAR strains sequenced, 33 had a mutation at Ser
83, indicating that it is the most common hotspot for gyrA gene
mutation in Salmonella strains identified in this study that is
analogous to the previous findings.6,16 Ser 83 contains one of
the three HinfI restriction sites located in the gyrA gene. Mu-
tation at a codon analogous to Ser 83 leads to loss of one of
these restriction sites. We also observed that in the absence of
sequencing, HinfI digestion is useful as it is able to detect mu-
tation at Ser 83.
All the NAR S. Typhi and S. Paratyphi A isolates tested had
a single mutation in the gyrA gene, at either Ser 83 or Asp 87.
RENUKA ET AL.
Similar mutations have been reported in a study carried out on
12 S. Typhi isolates from South India.37 Nine of these isolates
had a mutation at Ser 83, two isolates had a mutation at Asp
87, and one isolate had mutation at both Ser 83 and Asp 87, al-
though the MIC of ciprofloxacin of all these isolates was the
same (0.256 g/ml). Double mutations in the gyrA gene have
also been reported in in vitro-selected ciprofloxacin-resistant
mutants of S. Typhi and S. Paratyphi A having MIC of
ciprofloxacin more than or equal to 4 g/ml.16 Therefore, it
seems that double mutations in the gyrA gene can lead to high-
level resistance to quinolones with MIC above the established
breakpoints. The varying MICs of ciprofloxacin in NAR strains
suggest that some other mechanism, such as increased expres-
sion of efflux pump, in addition to gyrA gene mutation may
be involved in high-level resistance to fluoroquinolones. The
detection of strains with reduced susceptibility to fluoro-
quinolones, which may have a single point mutation in the gyrA
gene, is important because these strains may persist and acquire
additional mutations to achieve a higher degree of resistance
when standard doses are used. The problem in the clinical lab-
oratory is failure to identify Salmonella strains with reduced
susceptibility to ciprofloxacin by using the disk diffusion
method, which is the method of performing the antimicrobial
susceptibility testing for the routine diagnostic laboratory. Ac-
cording to the present NCCLS guidelines, these strains are in-
terpreted as sensitive to ciprofloxacin, thus it is possible that
the interpretative breakpoints for Enterobacteriaceae may not
hold true for S. Typhi and S. Paratyphi A and there is a need
to revise the NCCLS breakpoints for testing fluoroquinolone
susceptibility for Salmonella.1 We suggest that until this issue
is resolved, nalidixic acid susceptibility testing combined with
PCR-HinfI digestion provide useful and simple methods to
identify Salmonella strains that can result in an infection not
responding to ciprofloxacin therapy because we found that the
majority of patients with severe disease and documented
ciprofloxacin failure were infected with NAR strains, as has
also been observed by other workers.20,22,31,35,37
ACKNOWLEDGMENT
The work was supported by financial assistance from the De-
partment of Science and Technology, Government of India.
REFERENCES
1. Aarestrup, F. M., C. Wiuff, K. Molbak, and E.J. Threlfall. 2003.
Is it time to change fluoroquinolone breakpoints for Salmonella
spp.? (Letter) Antimicrob. Agents Chemother. 47:827–829.
2. Anand, A.C., V.K. Kataria, W. Singh, and S.K. Chatterjee.
1990. Epidemic multiresistant enteric fever in eastern India. (Let-
ter). Lancet. 335:352.
3. Aoyoma H, K. Sato, T. Kato, K. Hirai, and S. Mitsuhashi.
1987. Norfloxacin resistance in a clinical isolate of Escherichia
coli. Antimicrob. Agents Chemother. 31:1640–1641.
4. Baucheron, S. , H. Imberechts, E. Chaslus-Dancla, and A.
Cloeckaert. 2002. The AcrB multidrug transporter plays a major
role in high-level fluoroquinolone resistance in Salmonella enter-
ica serovar Typhimurium phage type DT204. Microb. Drug Resist.
8:281–289.
REDUCED SUSCEPTIBILITY TO CIPROFLOXACIN IN S. TYPHI AND S. PARATYPHI A
5. Bearden, D.T. , and L.H. Danziger. 2001. Mechanism of action
of and resistance to quinolones. Pharmacotherapy 21:224S–232S.
6. Brown, J.C., P.M.A. Shanahan, M.V. Jesudason, C.J. Thom-
son, and G.B. Amyes. 1996. Mutations responsible for re-
duced susceptibility to 4 quinolones in clinical isolates of multi-
resistant Salmonella typhi in India. J. Antimicrob. Chemother.
37:891–900.
7. Cloeckaert, A., and E. Chaslus-Dancla. 2001. Mechanism of
quinolone resistance in Salmonella. Vet. Res. 32:291–300.
8. Everett, M.J., Yu Fang Jin, V. Ricci, and E.G. Piddock. 1996.
Contributions of individual mechanisms to fluoroquinolone resis-
tance in 36 Escherichia coli strains isolated from humans and an-
imals. Antimicrob. Agents Chemother. 40:2380–2386.
9. Eykyn, S.J., and H. Williams. 1987. Treatment of multiresis-
tant S. Typhi with oral ciprofloxacin. (Letter). Lancet. 2:1407–
1408.
10. Georgiou, M., R. Munoz, F. Roman, R. Canton, Gomez-Lus, J.
Campos., et al. (1996). Ciprofloxacin resistant H. influenzae strains
possess mutations in analogous positions of gyrA and par C. An-
timicrob. Agents Chemother. 40:1741–1744.
11. Giraud, E., A. Brisabois, J-L. Martel, and E. Chaslus-Dancla.
1999. Comparative studies of mutations in animal isolates and
experimental in vitro and in vivo selected mutants of Salmo-
nella spp. suggest a counter selection of highly fluoroquinolone
resistant strains in the field. Antimicrob. Agents Chemother.
43:2131–2137.
12. Giraud E, A. Cloeckaert, D. Kerboeuf, and E. Chaslus-Dancla.
2000. Evidence of active efflux as primary mechanism of resis-
tance to ciprofloxacin in Salmonella enterica serovar Ty-
phimurium. 44:1223–1228.
13. Gutmann L, D. Billot-Klein, R. Williamson, F.W. Goldstein, J.
Mounier, J.F. Acar, and E. Collatz. 1988. Mutation of Salmo-
nella Paratyphi A conferring cross-resistance to several groups of
antibiotics by decreased permeability and loss of invasiveness. An-
timicrob. Agents Chemother. 32:195–201.
14. Heisig, P. 1993. High-level fluoroquinolone resistance in Salmo-
nella typhimurium isolate due to alteration of both gyrA and gyrB
genes. J. Antimicrob. Chemother. 32:367–377.
15. Heisig, P., H. Schidletzky, and H. Falkenstein-Paul. 1993. Mu-
tations in the gyrA gene of a high-level fluoroquinolone resistant
clinical isolate of Escherichia coli. Antimicrob. Agents Chemother.
37:696–701.
16. Hirose, K., A. Hashimoto, K. Tamura, Y. Kawamura, T. Ezaki,
H. Sagara, et al. 2002. DNA sequence analysis of DNA gyrase
and DNA topoisomerase IV quinolone resistance determining re-
gions of Salmonella enterica serovar Typhi and serovar Paratyphi
A. Antimicrob. Agents Chemother. 46:3249–3252.
17. Hooper, D.C. 2001. Emerging mechanisms of fluoroquinolone re-
sistance. Emerg. Infect. Dis. 7:337–341.
18. Ishii, H., K. Sato, K. Hoshino, M. Sato, A. Yamaguchi, T. Sawai,
and Y. Osada. 1991. Active efflux of ofloxacin by a highly
quinolone resistant strain of Proteus vulgaris. J. Antimicrob.
Chemother. 28:827–836.
19. Kapil, A., A. Ayyagiri, A, R.K. Garg, and K.C. Agarwal. 1994.
Salmonella Typhi with transferable chloramphenicol resistance iso-
lated in Chandigarh during 1983–87. Indian J. Pathol. Microbiol.
38:179–183.
20. Kapil, A., S. Sood, N.R. Dash, B.K. Das, and P. Seth. 1999.
Ciprofloxacin in typhoid fever. Lancet 354:164.
21. Kapil, A., Renuka, and B.K. Das. 2002. Nalidixic acid suscepti-
bility test to screen ciprofloxacin resistance in Salmonella Typhi.
Indian J. Med. Res. 115:49–54.
153
22. Larway, O., J.C.N. Van, A. Buu-Hoi, and J.F. Acar. 1997. Ty-
phoid fever due to a Salmonella Typhi strain of reduced suscepti-
bility to fluoroquinolones. Clin. Microbiol. Infect. 3:541–543.
23. Mandal, B.K. 1991. Modern treatment of typhoid fever. J. Infect.
22:1–4.
24. Mehta, A., C. Rodriques, and U.R. Joshi. 1992. Multiresistant
Salmonella organism in India. (Letter). J. Am. Med. Assoc.
267:1614.
25. National Committee for Clinical Laboratory Standards. 2002.
Performance standards for antimicrobial disk susceptibility tests,
seventh edition. Approved standard M2–A7. National Committee
for Clinical Laboratory Standards, Wayne, PA.
26. Nicholas, K.B., H.B. Nicholas, Jr., and D.W. Deerfield II. 1997.
Genedoc: Analysis and Visualization of Genetic Variation, EMB-
NEW. NEWS 4:14, http://www.psc.edu/biomed/genedoc (accessed
April 21, 2003.
27. Old, D.C. 1996. Salmonella. In Collee, J.G., B.P. Marmion, A.G.
Fraser, and A. Simmons (eds.). Mackie & McCartney practical
medical microbiology, 14th ed. Churchill Livingstone, Edinburgh,
pp. 386–403.
28. Pan, X. -S., J. Amblen, S. Mehtar, and L.M. Fisher. 1996. In-
volvement of topoisomerase IV and DNA gyrase as ciprofloxacin
targets in S. pneumoniae. Antimicrob. Agents Chemother.
40:2321–2326
29. Piddock, L.J.V., V. Ricci, I. McLaren, and D.J. Griggs. 1998.
Role of mutation in gyrA and parC genes in nalidixic-acid-resis-
tant Salmonella serotypes isolated from animals in United King-
dom. J. Antimicrob Chemother. 41:635–641.
30. Richens, J. 1996. Typhoid and paratyphoid fevers. In Weatherall,
D.J., J.G.G. Ledingham, and D.A. Warrell, (eds.). Oxford textbook
of medicine, 3rd ed., vol. 1. Oxford University Press, pp. 560–567.
31. Rowe, B., L.R. Ward, and E.J. Threlfall. 1995. Ciprofloxacin re-
sistant Salmonella Typhi in UK. Lancet 346:1302.
32. Ruiz J, D. Castro, P. Goni, J.A. Santamaria, J.J. Borrego, and
J. Vila. 1997. Analysis of mechanism of quinolone resistance in
nalidixic acid resistant clinical isolates of Salmonella serotype Ty-
phimurium. J. Med. Microbiol. 46:623–628.
33. Sood, S., A. Kapil, N. Dash, B.K. Das, V. Goel, and P. Seth.
1999. Paratyphoid fever in India: An emerging problem. Emerg.
Infect. Dis. 5:483–484.
34. Thomson, J.D., T.J. Gibson, F. Plewniak, F. Jeanmougin, and
D.G. Higgins. 1997. The ClustalX windows interface: flexible
strategies for multiple sequence alignment aided by quality analy-
sis tools. Nucleic Acids Res. 24:4876–4882.
35. Threlfall, E.J., and L.R. Ward. 2001. Decreased susceptibility to
ciprofloxacin in Salmonella enterica serotype Typhi, United King-
dom. Emerg. Infect. Dis. 7:448–450.
36. Umashankar S, R.A. Wall, and J. Berger. 1992. A case of
ciprofloxacin resistant typhoid fever. Commun. Dis. Rep. Rev
2:R139–R140.
37. Wain, J., N.T.T. Hoa, N.T. Chinh, H. Vinh, M.J. Everett, T.S.
Diep, et al. 1997. Quinolone resistant Salmonella Typhi in Viet
Nam: Molecular basis of resistance and clinical response to treat-
ment. Clin. Infect. Dis. 25:1404–1410.
Address reprint requests to:
Professor A. Kapil
Department of Microbiology
All India Institute of Medical Sciences
New Delhi, India
E-mail: artikapil@yahoo.com