Mutations in the quinolone resistance determining
region in Staphylococcus epidermidis recovered from
conjunctiva and their association with susceptibility
to various fluoroquinolones
M Yamada, J Yoshida, S Hatou, T Yoshida, Y Minagawa
Division for Vision Research,
National Institute of Sensory
Organs, National Tokyo Medical
Center, Tokyo, Japan
Dr M Yamada, Division for Vision
Research, National Institute of
Sensory Organs, National Tokyo
Medical Center, 2-5-1
Higashigaoka, Meguro-ku, Tokyo
Accepted 10 April 2008
This paper is freely available
online under the BMJ Journals
unlocked scheme, see http://
Background: Staphylococcus epidermidis is one of the
prominent pathogens in ocular infection. The prevalence
of mutations in the quinolone resistance determining
region (QRDR) area in S epidermidis isolated from the
ocular surface and its association with fluoroquinolone
resistance has not been fully elucidated.
Methods: Mutations in the QRDR of gyrA, gyrB, parC,
and parE genes of 138 isolates of S epidermidis recovered
from the human conjunctival flora were analysed. The
minimal inhibitory concentrations (MICs) of four fluoro-
quinolones (levofloxacin, gatifloxacin, moxifloxacin and
tosufloxacin) against these isolates were also determined
using agar dilution methods.
Results: The MIC90values of levofloxacin, gatifloxacin,
moxifloxacin and tosufloxacin were 3.13, 1.56, 0.78 and
3.13 mg/ml, respectively. The MIC values of all fluoro-
quinolones showed a bimodal distribution (susceptible
strain and less susceptible strain). Mutations with amino
acid substitution in the QRDR were present in 70 (50.7%)
isolates. 19 different combinations of mutations were
detected: 3 isolates (2.2%) had four mutations, 8 (5.8%)
had three mutations, 43 (31.2%) had double mutations
and 16 (11.6%) had single mutations. Isolates with
mutations in the QRDR of both gyrA and parC (n=53)
were less susceptible to fluoroquinolones.
Conclusions: The present findings show that approxi-
mately half the S epidermidis isolates from the normal
human conjunctiva have mutation(s) in the QRDR. The
presence of mutations in both gyrA and parC is strongly
associated with reduced susceptibility to fluoroquino-
Staphylococcus epidermidis is one of the most
prominent causes of conjunctivitis, keratitis and
quency of different organisms as causative agents
in keratitis varies during different periods and in
different geographical regions, S epidermidis is
among the most frequently encountered organisms
in clinical studies conducted in the USA, Germany
and Japan.3–5It is the most common bacterial
isolate in most large studies of acute postoperative
The fluoroquinolones are the newest family of
antibacterial agents used in the treatment of
ocular infections.2 3 9–11In Japan, ofloxacin was
the first fluoroquinolone introduced for topical
ophthalmic use in 1987. Since then, six other
tosufloxacin (TFLX) and moxifloxacin (MFLX)—
have been approved for clinical use as eye drops in
Japan. In addition to these compounds, ciproflox-
acin has been used clinically in other countries.
Their bactericidal activity
frequently observed Gram-positive and Gram-
negative ocular pathogens is generally excellent
and their high potency has made them a common
choice for the treatment and prevention of ocular
However, as with other antibiotic agents,
continued use in a population raises the issue of
emerging resistance.12–14Since the introduction of
fluoroquinolones for ophthalmic use, the reported
incidence of in vitro resistance to fluoroquinolones
in bacteria isolated from cases with bacterial
keratitis and endophthalmitis has been steadily
increasing. A previous study reviewed the database
of bacterial flora cultured from the conjunctival sac
of 1455 Japanese patients scheduled for intraocular
surgeries between 1995 and 2002.14The incidence
of in vitro resistance of bacterial isolates to
ofloxacin increased from 13.5% in 1995 to 32.8%
in 1999. Moreover, when ofloxacin was replaced by
LVFX in 2000, the incidence of resistance to LVFX
gradually increased from 14.5% in 2000 to 20.5% in
The primary targets of fluoroquinolones are two
essential enzymes of bacterial cells, DNA gyrase
and topoisomerase IV.15–17In S epidermidis, DNA
gyrase is composed of the GyrA and GyrB subunits
encoded by the gyrA and gyrB genes, respectively.
Topoisomerase IV is composed of ParC and ParE
subunits encoded by parC and parE genes, respec-
tively. In most bacterial species, mutations occur in
the highly conserved quinolone resistance-deter-
mining regions (QRDR) of the genes that encode
Staphylococcus aureus, several studies have shown
that a combination of mutations in both genes can
cause high-level resistance even to the newer
fluoroquinolones.18–21However, the prevalence of
mutations in the QRDR in S epidermidis isolated
from the ocular surface and its association with
fluoroquinolone resistance have not been fully
investigated.15–17The present study analysed muta-
tions in the QRDR of gyrA, gyrB, parC and parE
genes of 138 isolates of S epidermidis recovered from
conjunctival flora. The susceptibility of these
isolates to LVFX, GFLX, MFLX and TFLX was also
against the most
848Br J Ophthalmol 2008;92:848–851. doi:10.1136/bjo.2007.129858
Bacterial isolates and susceptibility testing
One hundred and thirty-eight isolates of S epidermidis were
collected from the conjunctival sac of 138 eyes of 129 patients
who were scheduled for intraocular surgery at the National
Tokyo Medical Center between November 2004 and June 2005.
The mean (SD) age of the patients was 70.7 (14.9) years (range
6–91 years). The patients had not received either ophthalmic or
systemic antibiotics prior to bacterial sampling.
Scrapes of the inferior conjunctival fornix were taken in the
absence of topical anaesthetic using a sterile cotton swab. The
samples were immediately inoculated into Mueller-Hinton
(MH) agar and incubated at 35uC in air for 16–20 h for the
selection of staphylococci. The MicroScan WalkAway-96
(Baxter Japan, Tokyo) with MicroScan Rapid Pos Combo Panel
(Baxter) was used for the identification of S epidermidis. Positive
cultures were stored at 280uC until the agar dilution testing to
determine the minimum inhibitory concentration (MIC).
MICs for LVFX, GFLX, MFLX and TFLX were determined by
the agar dilution method in accordance with the recommenda-
tions of the Japanese Society of Chemotherapy.22The bacterial
suspensions in saline were inoculated on MH agar plates
supplemented with defined concentrations of drugs. The plates
were incubated at 35uC under aerobic conditions and MICs
were determined after 20–24 h of incubation. Drug concentra-
tions ranged from 0.025 mg/ml to 100 mg/ml in twofold
increments except for TFLX (0.025 mg/ml to 25 mg/ml) because
of its limited solubility.
DNA amplification and sequencing of QRDR
The isolates were suspended in tryptic soy broth and cultured
overnight. Genomic DNA was extracted using the Wizard SV 96
genomic DNA purification system (Promega KK, Japan). One ml
of the genomic DNA solution was applied in 20 ml of
amplification mixture (5 pM each primer, 1.6 ml dNTP mixture,
2 ml Ex Taq buffer and 0.1 ml LA Taq (Takara Bio Inc, Japan)).
Polymerase chain reaction (PCR) amplification was performed
with the primers as shown in table 1. PCR primers were selected
from the published sequences of S epidermidis RP62A. Each
reaction was amplified with the following temperature profiles:
30 cycles at 94uC for 30 s, 55uC for 30 s and 72uC for 1 min. The
amplified DNA products were separated and identified by 2%
agarose gel electrophoresis.
PCR products were purified using ExoSAP according to the
manufacturer’s instructions (GE Healthcare Bio-Sciences KK,
Japan). PCR-amplified DNA was sequenced by the dye
number NC 002976 (S epidermidis RP62A)
Primers used in the study. Nucleotide positions are indicated according to GenBank sequence
Target gene Primer sequence (59 to 39)
284 2 609 699–2 609 724
2 609 441–2 609 460
2 610 508–2 610 528
2 610 278–2 610 297
939 185–939 204
939 361–939 381
938 196–938 219
938 493–938 520
Mutations in the quinolone resistance determining regions of gyrA, parC and parE in 70 strains of
Ser80Tyr + Asp84Val +
Ser80Phe + Asp84Tyr
Ser80Phe + Asp84Asn
Ser80Phe + Asp84Ala
Ser84Tyr + Glu88Lys
Br J Ophthalmol 2008;92:848–851. doi:10.1136/bjo.2007.129858849
terminator method in both the forward and reverse directions.
Using Phred/Phrap/Polyphred software, the quality score of
each base was calculated. Sample sequences were compared
with a reference sequence and mutations were detected. The
strain S epidermidis ATCC 35984 (RP62A) was used as a
The mutations identified in the QRDR of the gyrA, gyrB, parC
and parE genes are summarised in table 2. Nineteen different
combinations of mutations were identified in 70 isolates,
whereas no mutations were detected in 68 isolates. Three
isolates (mutation profile type 2, 9 and 12) had four amino acid
substitutions, 8 isolates (mutation profile type 4, 5 and 8) had
three amino acid substitutions, 43 isolates (mutation profile
type 1, 3, 6, 7, 10, 11 and 13) had double amino acid
substitutions and 16 isolates (mutation profile type 14–19)
had single amino acid substitutions.
In the gyrA gene, a single-point mutation was found in 53
isolates at codon 84. Double-point mutations in the gyrA gene
were identified in 1 isolate at codons 84 and 88 (mutation
profile type 9). No mutations were found in the QRDR area of
the gyrB gene. In the parC gene, single-point mutations were
found in 51 isolates at codons 69, 80, 81, 84 or 85. Double-point
mutations were identified in 6 isolates at codons 80 and 84
(mutation profile type 4, 5 and 9). Triple-point mutations were
identified in 1 isolate at codons 80, 84 and 85 (mutation profile
type 2). In the parE gene, single-point mutations were found in
16 isolates at codon 404 or 434. Double-point mutations were
identified in 1 isolate at codons 404 and 434.
The MICs of the four tested fluoroquinolones against S
epidermidis are shown in table 3. All four fluoroquinolones had a
bimodal distribution in all isolates (n=138). Isolates with no
mutations in the QRDR (wild type; n=68) were susceptible to
fluoroquinolones. The modes (the number that appears the
most) were 0.2 mg/ml for LVFX, 0.1 mg/ml for GFLX, 0.05 mg/ml
for MFLX, and 0.05 mg/ml for TFLX. Isolates with mutations
restricted in the QRDR of parC and/or parE (n=17) showed
similar susceptibilities to fluoroquinolones as wild type strains
except for one strain with mutation profile type 18. The modes
were 0.2 mg/ml for LVFX, 0.1 mg/ml for GFLX, 0.05 mg/ml for
MFLX and 0.05 mg/ml for TFLX. Isolates with mutations in the
QRDR of both gyrA and parC (n=53) were less susceptible to
fluoroquinolones. The modes were 1.56 mg/ml for LVFX,
0.78 mg/ml for GFLX, 0.39 mg/ml for MFLX and 0.78 mg/ml
for TFLX. Of these 53 isolates, 51 had amino acid substitutions
at GyrA84 and ParC80. One isolate (mutation profile type 9)
with two amino acid substations both in GyrA and ParC had
the highest MICs (25 mg/ml for LVFX, 12.5 mg/ml for GFLX,
MFLX and TFLX, respectively).
The primary targets of fluoroquinolones are two essential
enzymes of bacterial cells, DNA gyrase and topoisomerase IV.18–20
In most bacterial species the mutations in the genes that lead to
fluoroquinolone resistance are limited to a few point mutations at
restricted positions of the genes called QRDR. The present study
revealed that approximately half (50.7%) of S epidermidis isolates
in the human conjunctival flora have mutation(s) in the QRDR
area of gyrA, gyrB, parC and parE genes.
Fluoroquinolone resistance has been studied intensively in S
aureus.18–21The genes encoding topoisomerase IV in S aureus are
called grlA and grlB, which are analogous to parC and parE in S
epidermidis, respectively. Fluoroquinolone resistance in S aureus
Susceptibility of strains of S epidermidis to four fluoroquinolones
No of isolates with the following MIC (mg/ml)
0.025 0.05 0.1 0.20.390.78 1.563.13 6.2512.525
All isolates (n=138)
Wild type (n=68)
Mutations in parC and/or parE (n=17)
Mutations in both gyrA and ParC (n=53)
GFLX, gatifloxacin; LVFX, levofloxacin; MFLX, moxifloxacin; TFLX, tosufloxacin.
850 Br J Ophthalmol 2008;92:848–851. doi:10.1136/bjo.2007.129858
is generally associated with two single-point mutations in gyrA Download full-text
at codon 84, and in grlA at codon 80 or 84. S aureus isolates with
higher levels of resistance are associated with the second
mutation in grlA at codon 80 or 84, depending on the position
of the first mutation. When the second mutation in gyrA occurs
at codon 85 or 88, in addition to the first mutation at codon 84,
the strain shows the highest fluoroquinolone resistance even to
The present QRDR sequencing results indicate that the major
mechanism of fluoroquinolone resistance in S epidermidis is
analogous to that of S aureus. Isolates with mutations restricted
to the QRDR of parC and/or parE (n=17) in this study were
similarly susceptible to fluoroquinolones as wild type strains.
However, the presence of two mutations (n=53) in both gyrA
gene (located at codon 84) and parC gene (located at codon 80)
have been found to be associated with the development of
fluoroquinolone resistance.15 16
In this study only one isolate (mutation profile type 9), which
was highly resistant to all four fluoroquinolones tested, had two
amino acid substitutions both in GyrA and ParC. Previous
studies have shown that isolates of S epidermidis and S aureus
with two amino acid substitutions both in GyrA and ParC
(GrlA in S aureus) have the highest fluoroquinolone resistance.
The isolates with this mutation type are reported to be
relatively rare in S epidermidis15 16and to account for less than
10% in S aureus.18–20However, a high prevalence (50%) of two
amino acid substitutions in both GyrA and GrlA has recently
been reported.21The empirical use of newer fluoroquinolones
without a proper clinical indication may produce additional
resistant strains of S epidermidis, as has already occurred with S
One possible limitation of the present study was that the
patients were scheduled for intraocular surgery. Bacterial
isolates therefore represent conjunctival flora rather than ocular
pathogens. However, in common ocular infections such as
bacterial conjunctivitis and bacterial keratitis, pathogens are
frequently the normal bacterial flora that reside on the ocular
This istrue even
endophthalmitis, in which S epidermidis is the most common
bacterial isolate from vitreous aspirates.7 8Organisms isolated
from the vitreous were genetically identical to those collected
from the ocular surface in 68–82% of patients with post-
operative endophthalmitis,7suggesting that the study of in vitro
susceptibility to various fluoroquinolones is valid.
Drug resistance is a serious concern in treating ocular
infections. The current study showed that approximately half
the S epidermidis isolates from the conjunctival flora have
mutation(s) in the QRDR. Both gyrA gene and parC gene are
associated with the development of fluoroquinolone resistance.
Funding: Supported in part by a grant from the Ministry of Health, Labour and
Competing interests: None.
Ethics approval: The principles of the World Medical Association Declaration of
Helsinki were followed. Each subject received a thorough explanation of the purpose of
the study and all procedures involved in the study, and provided written informed
consent prior to enrolment. Approval for this investigation was granted by the
Committee for the Protection of Human Subjects at National Tokyo Medical Center.
O’Brien TP. Bacterial keratitis. In: Krachmer JH, Mannis MJ, Holland EJ, eds. Cornea
and external disease: clinical diagnosis and management. Volume II. St Louis: Mosby-
Year Book, 1997:1139–90.
Graves A, Henry M, O’Brien TP, et al. In vitro susceptibilities of bacterial ocular
isolates to fluoroquinolones. Cornea 2001;20:301–5.
O’Brien TP, Maquire MG, Fink NE, et al. Efficacy of ofloxacin versus cefazolin and
tobramycin in the therapy for bacterial keratitis. Report from the Bacterial Keratitis
Study Research Group. Arch Ophthalmol 1995;113:1257–65.
Study Group of National Surveillance of Infectious Keratitis in Japan.
National surveillance of infectious keratitis in Japan. Current status of isolates, patient
background, and treatment. J Jpn Ophthalmol Soc 2006;110:961–72.
Frohlich SJ, deKaspar HM, Grabson T, et al. Bacterial keratitis in patients with and
without contact lens anamnesis. Klin Monat Augenheilk 1999;214:211–6.
Bourcier T, Thomas F, Borderie V, et al. Bacterial keratitis: predisposing factors,
clinical and microbiological review of 300 cases. Br J Ophthalmol 2003;87:834–8.
Han DP, Wisniewski SR, Wilson LA, et al. Spectrum and susceptibilities of
microbiologic isolates in the Endophthalmitis Vitrectomy Study. Am J Ophthalmol
Sandvig KU, Dannevig L. Postoperative endophthalmitis: establishment and results
of a national registry. J Cataract Refract Surg 2003;29:1273–80.
Neu HC. Microbiologic aspects of fluoroquinolones. Am J Ophthalmol
Leibowitz HM. Clinical evaluation of ciprofloxacin 0.3% ophthalmic solution for
treatment of bacterial keratitis. Am J Ophthalmol 1991;112(Suppl 4):S34–47.
Hyndiuk RA, Eiferman RA, Caldwell DR, et al. Comparison of ciprofloxacin
ophthalmic solution 0.3% to fortified tobramycin-cefazolin in treating bacterial corneal
ulcers. Ophthalmology 1996;103:1854–62.
Chalita MG, Hu ¨fling-Lima AN, Paranhos A, et al. Shifting trends in in vitro antibiotic
susceptibilities for common ocular isolates during period of 15 years. Am J Ophthalmol
Goldstein MH, Kowalski RP, Gordon YJ. Emerging fluoroquinolone resistance in
bacterial keratitis. Ophthalmology 1999;106:1313–8.
Kurokawa N, Hayashi K, Konishi M, et al. Increasing ofloxacin resistance of bacterial
flora from conjunctival sac of preoperative ophthalmic patients in Japan.
Jpn J Ophthalmol 2002;46:586–9.
Dubin DT, Fitzgibbon JE, Nahvi MD, et al. Topoisomerase sequences of coagulase-
negative staphylococcal isolates resistant to ciprofloxacin or trovafloxacin. Antimicrob
Agents Chemother 1999;43:1631–7.
Li Z, Deguchi T, Yasuda M, et al. Alteration in the GyrA subunit of DNA gyrase and
the ParC subunit of DNA topoisomerase IV in quinolone-resistant clinical isolates of
Staphylococcus epidermidis. Antimicrob Agents Chemother 1998;42:3293–5.
Sreedharan S, Peterson LR, Fisher LM. Ciprofloxacin resistance in coagulase-
positive and -negative staphylococci: role of mutations at serine 84 in the DNA gyrase
A protein of Staphylococcus aureus and Staphylococcus epidermidis. Antimicrob
Agents Chemother 1991;35:2151–4.
Wang T, Tanaka M, Sato K. Detection of grlA and gyrA mutations in 344
Staphylococcus aureus strains. Antimicrob Agents Chemother 1998;42:23–40.
Hooper DC. Fluoroquinolone resistance among Gram-positive cocci. Lancet Infect Dis
Horii T, Suzuki Y, Monji A, et al. Detection of mutations in quinolone resistance-
determining regions in levofloxacin- and methicillin-resistant Staphylococcus aureus:
effect of the mutations on fluoroquinolone MICs. Diagn Microbiol Infect Dis
Iihara H, Suzuki T, Kawamura Y, et al. Emerging multiple mutations and high-level
fluoroquinolone resistance in methicillin-resistant Staphylococcus aureus isolated
from ocular infections. Diagn Microbiol Infect Dis 2006;56:297–303.
Japanese Society of Chemotherapy. Revision of methods for determining
minimum inhibitory concentrations. Chemotherapy 1981;29:76–9.
Br J Ophthalmol 2008;92:848–851. doi:10.1136/bjo.2007.129858851