Resistance to fluoroquinolones and methicillin in ophthalmic isolates of Staphylococcus pseudintermedius from companion animals

Abstract

Resistance to fluoroquinolones and methicillin was determined for 49 ophthalmic isolates of Staphylococcus pseudintermedius from dogs with and without ophthalmic disease. Resistance was observed for ciprofloxacin (40.8%), ofloxacin (38.8%), enrofloxacin (38.8%), levofloxacin (34.7%), and moxifloxacin (4.1%). Eighteen isolates, 16 of which were resistant to oxacillin, were mecA-positive. Nine of the 16 oxacillin-resistant mecA-positive S. pseudintermedius isolates were resistant to more than one fluoroquinolone and 2 isolates were resistant to 5 fluoroquinolones. The frequency of mecA gene occurrence and fluoroquinolone resistance was twice as high among S. pseudintermedius isolates derived from dogs with ophthalmic disease compared with isolates for dogs without ophthalmic disease. The high prevalence of methicillin and fluoroquinolone resistance in S. pseudintermedius from dogs with ophthalmic disease is a concern.

Introduction

Members of the Staphylococcus intermedius group (S. intermedius, S. pseudintermedius, and S. delphini) have been associated with various diseases in different animal species (1). Staphylococcus intermedius was reclassified as S. pseudintermedius and described as the most common bacterial pathogen isolated from skin and ear infections in dogs and cats (2–4). It is also the most common commensal isolate from healthy skin and ears (5). In 2006, infection with S. pseudintermedius after a surgical procedure was first reported in humans (6) and human infection acquired from dogs has been reported more recently (7). However, transmission of S. pseudintermedius between humans and animals is uncommon.

Fluoroquinolones have been successfully used as bactericidal broad-spectrum antibiotics against serious bacterial infections. The quinolones inhibit nucleic acid synthesis by blocking DNA gyrase encoded by gyrA and gyrB, and topoisomerase IV, encoded by grlA and grlB (8). Bacterial resistance to fluoroquinolones is associated with alterations in target enzymes (chromosomal mutations causing reduced affinity of DNA gyrase and topoisomerase IV) or alterations in drug permeation (overexpression of drug efflux pumps) (9). One point mutation in topoisomerase II and IV conferred resistance to S. pseudintermedius (10,11). The frequency of methicillin-resistant S. pseudintermedius (MRSP) has increased and is an emerging problem in veterinary medicine (4,12,13). Many MRSP strains have blaZ and mecA genes and display multi-drug resistance (MDR). The MRSP strains with the mecA gene tend to possess resistance to many classes of antibiotics including fluoroquinolones and represent a challenge for therapy (10,14,15). Transfer of multiple resistance genes to canine pathogenic staphylococci from other animal or human staphylococci is a serious concern (16,17).

The most common route of administration for ocular therapy is topical. The concentration, kinetics and elimination of drugs within the conjunctival sac and corneal permeability affect the penetration of topical medications (18). Tear flow, space within the conjunctival fornix, and pharmaceutical characteristics of the drugs also have a dynamic effect on topical ophthalmic medications (19). Because there are no separate standards for guiding topical antibiotic therapy, the frequency of resistant bacteria appears to be considerable on the basis of in vitro breakpoints applicable to systemic use (20,21).

The aim of this study was to examine fluoroquinolone- and methicillin-resistance of S. pseudintermedius isolates from canine eyes. Furthermore, we carried out polymerase chain reaction (PCR)-based mecA gene detection and investigated resistance to 5 fluoroquinolones among MRSP to compare the findings between isolates from dogs with ophthalmic disease and dogs without ophthalmic disease.

Materials and methods

Bacterial isolation and identification

Forty-nine Staphylococcus pseudintermedius isolates were obtained from clinical and non-clinical ophthalmic samples (66 individual canine eyes) during the period 2005–2007 at the Veterinary Medical Teaching Hospital of Konkuk University, Seoul, Korea. Among the 49 ophthalmic S. pseudintermedius isolates, 24 were recovered from dogs with ophthalmic disease and 25 were cultured from dogs without ophthalmic disease. The diseases of the dogs in the study are listed in Table 1. Seventy-nine percent (19/24) of dogs with ophthalmic disease and 40% (10/25) of dogs with non-ophthalmic disease had been exposed to topical and/or systemic antibiotics. Eleven of the dogs with ophthalmic disease had been treated with fluoroquinolone; information on antibiotic exposure of the dogs with non-ophthalmic disease was not available. Preliminarily screening of staphylococci and differentiation of staphylococcal species were made on the basis of standard microbiological procedures including phenotypical tests, Gram staining, and catalase test (14,22).

Table 1.

Distribution of canine ophthalmic and non-ophthalmic diseases in this study

	Number (%)
Ophthalmic disease (n = 24)
Corneal ulcer	8 (33.3)
Conjunctivitis	8 (33.3)
Keratoconjunctivitis sicca	6 (25)
Blepharitis	1 (4.2)
Corneal degeneration	1 (4.2)
Non-ophthalmic disease (n = 25)
Kidney disease	11 (44.0)
Neurologic disease	3 (12.0)
Liver disease	2 (8.0)
Neoplasia	2 (8.0)
Urologic disease	2 (8.0)
Skin disease	2 (8.0)
Heart disease	1 (4.0)
Respiratory disease	1 (4.0)
Immune-mediated disease	1 (4.0)

Extraction of bacterial DNA was performed with a commercial kit (Optima Scientific, Tokyo, Japan). Confirmation of Staphylococcus intermedius group (SIG) was carried out by PCR with species-specific primers targeting the 16S rDNA and the SIG nuclease (nuc) gene (23). A PCR-restriction fragment length polymorphism (RFLP) analysis of the pta gene was also performed for S. pseudintermedius identification (24).

Antimicrobial susceptibility testing

Determination of antimicrobial resistance as based on the guidelines of the Clinical and Laboratory Standards Institute (CLSI) (20). Five fluoroquinolones (ciprofloxacin, ofloxacin, enrofloxacin, levofloxacin, and moxifloxacin) were tested with Sensi-Disc™ disks (5 μg each; Becton Dickinson BBL, Sparks, Maryland, USA). Second to fourth generation quinolones were selected. Enrofloxacin is the most commonly used fluoroquino-lone in veterinary medicine and the other 4 fluoroquinolones are used in human medicine.

Detection of mecA and blaZ genes

The mecA and blaZ genes in S. pseudintermedius were deteced with specific PCR reactions (25). The primers used were: mecA (amplicon size: 532 bp), 5μ-AAAATCGATGGTAAAGGTTGGC-3′ (forward) and 5′-AGTTCTGCAGTACCGGATTTGC-3′ (reverse); blaZ (amplicon size: 700 bp), 5′-TGACCACTTTTATCAGCAAC-3′ (forward) and 5′-GCCATTTCAACACCTTCTTTC-3′ (reverse). The PCR conditions consisted of heating at 95°C for 4 min, 30 cycles of denaturation at 95°C for 1 min, annealing at 58°C for 1 min, and elongation at 72°C for 1 min, postextension at 72°C for 7 min, and storage at 4°C. The PCR products were identified by electrophoresis in 2% agarose gel stained with ethidium bromide.

Statistical analysis

Statistical significance was determined with SPSS for Windows version 20.0 (SPSS Inc., Chicago, Illinois, USA). The isolates were divided into 2 groups, according to the presence or absence of ophthalmic disease in the dogs. The percent of resistant isolates among fluoroquinolones for the ophthalmic conditions was compared by linear-by-linear association. The prevalence of mecA for the ophthalmic conditions was compared using chi-square calculated by a 2 × 2 contingency table. Statistical significance was accepted at P < 0.05.

Results

Percentages of the 49 S. pseudintermedius isolates that were resistant to ciprofloxacin, ofloxacin, enrofloxacin, levofloxacin, and moxifloxacin were 40.8, 38.8, 38.8, 34.7, and 4.1, respectively (Table 2). The 16 oxacillin-resistant, mecA-positive MRSP isolates appeared to be more resistant to fluoroquinolones than the mecA-negative isolates; 9 of the 16 (56.3%) oxacillin resistant mecA-positive MRSP isolates were resistant to 1 or more fluoroquinolones, including 2 (12.5%) isolates that were resistant to all 5 fluoroquinolones. In contrast, 10 of 31 (32.3%) mecA-negative isolates were resistant to the fluoroquinolones evaluated in this study. Nineteen of 20 ciprofloxacin resistant S. pseudintermedius were also resistant to the other second-generation fluoroquinolones (ofloxacin and enrofloxacin) and the third-generation fluoroquinolone (levofloxacin) (Table 3).

Table 2.

Fluoroquinolone susceptibility of 49 S. pseudintermedius isolates from dogs’ eyes

	Number of S. pseudintermedius isolates (%)
Antimicrobial agents	Susceptible	Intermediate	Resistant
Moxifloxacin	38 (77.6)	9 (18.4)	2 (4.1)
Levofloxacin	31 (63.3)	1 (2.0)	17 (34.7)
Enrofloxacin	29 (59.2)	1 (2.0)	19 (38.8)
Ofloxacin	29 (59.2)	1 (2.0)	19 (38.8)
Ciprofloxacin	29 (59.2)	0 (0.0)	20 (40.8)

Table 3.

Fluroquinolone resistance among the mecA-positive and mecA-negative canine ophthalmic S. pseudintermedius isolates

Pattern of fluoroquinolone resistance	Number of S. pseudintermedius isolates (%)
	mecA PCR-negative (n = 31)	mecA PCR-positive (n = 18)
		Oxa-R (n = 16)	Oxa-Suscb (n = 2)
NR	21 (67.7)	7 (43.8)	1 (50.0)
CIP	1 (3.2)	0 (0.0)	0 (0.0)
CIP + OFX	0 (0.0)	0 (0.0)	0 (0.0)
CIP + OFX + ENR	1 (3.2)	1 (6.3)	0 (0.0)
CIP + OFX + ENR + LVX	8 (25.8)	6 (37.5)	1 (50.0)
CIP + OFX + ENR + LVX + MFX	0 (0.0)	2 (12.5)	0 (0.0)

NR — no fluoroquinolone resistance; CIP — ciprofloxacin; OFX — ofloxacin; ENR — enrofloxacin; LVX — levofloxacin; MFX — moxifloxacin.

Oxa-R — resistant to oxacillin by disk diffusion assay; Oxa-Susc — susceptible or intermediate resistance to oxacillin by disc diffusion assay.

The prevalence of the mecA gene was twice as high in ophthalmic S. pseudintermedius isolated from dogs with ophthalmic disease (12/24, 50%) compared with isolates from dogs with non-ophthalmic disease (6/25, 24%) (Figure 1). However, prevalence of the mecA gene was not correlated with presence of ophthalmic disease (P = 0.059). There was a higher rate of resistance to at least one fluoroquinolone among S. pseudintermedius isolates from dogs with ophthalmic disease (13/24, 54%) compared with isolates from dogs with non-ophthalmic disease (7/25, 28%). The isolates were divided into 3 groups based on resistance to fluoroquinolones: no resistance, resistance to second- and/or third-generation fluoroquinolones, and resistance to all 5 fluoroquinolones (Figure 2). The prevalence of resistant fluoroquinolones was increased in the presence of ophthalmic disease (P = 0.037).

Figure 1.

Correlation between ophthalmic condition and mecA gene prevalence in the 49 S. pseudintermedius isolates.

Figure 2.

Correlation between ophthalmic condition and fluroquinolone resistance in the 49 S. pseudintermedius isolates. Isolates were divided into 3 groups: Group 1 — no resistance to 5 fluoroquinolones, Group 2 — resistance to second-generation (ciprofloxacin, enrofloxacin and ofloxacin) and/or third generation fluoroquinolone (levofloxacin), and Group 3 — resistance to all 5 fluoroquinolnes including marbofloxacin. The percentages of isolates in the 3 groups were: 72%, 28%, and 0%, respectively in the absence of ophthalmic disease and 46%, 46%, and 8%, respectively in the presence of ophthalmic disease.

Discussion

Fluoroquinolone resistance among the ophthalmic S. pseudintermedius isolates was higher (40%) than that of skin isolates that showed 18.9% to 27% in Korea (15), while being < 10% elsewhere (26,27), suggesting an association with frequent use of fluoroquinolone in ophthalmic disease. The highest resistance was observed to ciprofloxacin and resistance to other second-generation fluoroquinolones (enrofloxacin and ofloxacin) and third generation fluoroquinolone (levofloxacin), which are not permitted for veterinary use, was similarly high. This resistance pattern is similar to those previously reported (21,28), namely that resistance to one of ciprofloxacin, ofloxacin, or levofloxacin often confers resistance to the other two compounds. Fortunately, resistance to human-licensed moxifloxacin was low (4.1%); this likely reflects the need for 2 mutations, in both DNA gyrase and topoisomerase IV (8). However, the prevalence of intermediate moxifloxacin resistance (9/49; 18.4%) and resistance to more than one fluoroquinolone in ophthalmic S. pseudintermedius was comparatively high. Antibiotic susceptibility tests should be performed before the fluoroquinolones are prescribed.

When ocular antimicrobial agents are administered for infections, consideration should be given to penetration of topically applied drugs to reach the site of action and duration of drug action (29). Physiologically, corneal stroma (hydrophilic), epithelium and endothelium (lipophilic) act as natural barriers in the eye and reduce the bioavailability of drugs. Penetration of second-, third- and fourth-generation fluoroquinolones into aqueous and vitreous humor has been studied in a rabbit endophthalmitis model (30). This study reported that new generation fluoroquinolones had better aqueous and vitreous penetration than did older generation fluoroquinolones and ocular infection/inflammation increased the penetration of the drugs. Another study reported that levofloxacin and moxifloxacin might be considered the penetrating fluoroquinolones into cornea, aqueous humor, anterior vitreous, and that repeated instillation resulted in approximately 2 times greater penetration than did single instillation (31). There is no certain way to determine antibiotic concentration in ocular tissues during topical therapy; many researchers agree that susceptibility of bacterial isolates from the eye should be evaluated using CLSI procedures based on systemic breakpoints (32). However, the concentration of topical antibiotics may exceed the minimum inhibitory concentration (MIC) for ocular isolates and may be rapidly diluted through lacrimation and drainage (33). Due to the dynamics of breakpoint versus antibiotic resistance of ocular isolates, further studies are needed.

Sixteen mecA-positive oxacillin-resistant S. pseudintermedius isolates also displayed about 2 times higher resistance to fluoroquinolones. Sublethal exposure to fluoroquinolones alters the SOS response in methicillin-resistant Staphylococcus aureus (MRSA) and selects for stable mutants with enhanced expression of methicillin resistance (34). The minimum inhibitory concentrations (MIC) of fluoroquinolones for MRSA with the same combinations of mutations in grlA and gyrA gene were diverse due to other mutations or alterations in grlA, gyrA, grlB, gyrB, and the quinolone efflux sytem (35). These results suggest that the detection of MRSP and susceptibility determination using MIC are required before use of fluoroquinolone antibiotics to minimize the spread of antibiotic resistance.

The absence of difference in the prevalence of the blaZ gene (92% both in ophthalmic and non-ophthalmic disease dogs) is consistent with the frequent use of systemic antibiotics. Furthermore, the approximately 2 times higher prevalence of the mecA gene (50% in ophthalmic disease isolates and 24% in non-ophthalmic disease isolates) and multiple resistance to fluoroquinolones (54% in ophthalmic disease isolates and 28% in non-ophthalmic disease isolates) suggest that newer fluoroquinolones are also frequently used in canine ophthalmic medications.

In conclusion, this study demonstrates that methicillin resistance and fluoroquinolone resistance were prevalent in S. pseudintermedius isolates from ophthalmic disease in dogs. The results are consistent with the frequent use of fluoroquinolone in ophthalmic disease. Continuing surveillance and accurate therapeutic decision for antibiotic use in ophthalmic disease in dogs are necessary.