Tissue Concentrations of Enrofloxacin and Its Metabolite Ciprofloxacin after a Single Topical Dose in the Coqui Frog (Eleutherodactylus coqui)

The skin of amphibians is highly permeable, allowing for efficient transcutaneous absorption of drugs after topical administration. This transcutaneous absorption offers a unique treatment modality to achieve systemic therapy, as opposed to what may be considered more stressful methods such as via oral or injectable routes. Topical administration can eliminate the need for handling restraint and thus is particularly beneficial for semi- or fully aquatic species, small species, and debilitated individuals. Treatment efficiency can also be improved for species that are communally housed.

Many different pharmaceuticals are administered topically on amphibians. This route is the preferred route for administering tricaine methane sulfate (MS-222) in most species and for fluid and electrolyte treatment with various Ringer solutions (Wright and Whitaker, 2001; Wright, 2004). Antimicrobials are also commonly delivered topically on amphibians, with studies having been completed evaluating gentamicin (Riviere et al., 1979; Menard, 1984); metronidazole (Mombarg et al., 1992); oxytetracycline (Maruska, 1994; Somsiri et al., 1999); and chloramphenicol, tetracycline hydrochloride, nalidixic acid, nitrofurantoin, and trimethoprim with sulfamethoxazole (Menard, 1984). However, full pharmacokinetic data have only been reported for gentamicin, metronidazole, and oxytetracycline. Bath immersion is the recommended route of administration for all of these antibiotics in amphibians, although spot-on topical therapy with metronidazole has been demonstrated to be effective for systemic antiparasitic treatment (Mombarg et al., 1992).

Enrofloxacin is a fluoroquinolone antimicrobial with good activity against most Gram-negative bacteria, especially those of the family Enterobacteriaceae (e.g., Escherichia coli, Salmonella spp., and Proteus spp.), as well as variable coverage against several Gram-positive organisms (Staphylococcus spp. and Acinetobacter spp.) (Taylor et al., 2001; Papich and Riviere, 2009). In most vertebrates, including some amphibian species (Letcher and Papich, 1994; Howard et al., 2010), enrofloxacin can be metabolized at varying levels to the active metabolite ciprofloxacin. Ciprofloxacin has a similar spectrum of activity as enrofloxacin, as well as increased activity against some bacteria (e.g., Pseudomonas spp.). The result is an additive antibacterial effect with both enrofloxacin and ciprofloxacin (Pirro et al., 1997).

There are only two published studies evaluating subcutaneous and intramuscular routes of administration for enrofloxacin in amphibians (Letcher and Papich, 1994; Howard et al., 2010). Both studies demonstrated sustained plasma bactericidal levels of enrofloxacin for at least 24 h, with significant metabolism to ciprofloxacin. Enrofloxacin is also commonly delivered topically on amphibians for empirical systemic therapy but has never been evaluated.

The objectives of this pilot study were to 1) determine whether a single topical dose of enrofloxacin (10 mg/kg) given transcutaneously to coqui frogs (Eleutherodactylus coqui) is absorbed and distributed to the liver and kidneys; and 2) measure whether enrofloxacin is metabolized to ciprofloxacin in this species. The contribution of ciprofloxacin is important to assess because, against some bacteria (e.g., Pseudomonas aeruginosa), ciprofloxacin is more active than enrofloxacin (Rubin et al., 2008). Historical enrofloxacin minimum inhibitory concentrations (MICs) of bacteria isolated from amphibians at the Wildlife Conservation Society (WCS) were also reviewed to determine the clinical relevance of this research to zoological medical practice.

Twelve adult coqui frogs (3–6 g each) of undetermined sex or age were included in this study. The animals were from a group of 48 animals being held at the WCS that were wild caught in Hawaii, USA, where they are an invasive, introduced species. A mortality event caused the majority (75%, n = 36) of these animals to die within 2 months of arrival (McAloose et al., 2008). All animals were polymerase chain reaction (PCR) positive for Batrachochytridium dendrobatidis (chytrid), and several others were PCR positive for ranavirus. The remaining 12 individuals appeared healthy and in good condition and were treated for chytrid infection with itraconazole baths (Nichols et al., 2000). All 12 animals remained in good health with negative chytrid PCR test results 7 months after arrival. Because these animals were exposed to a potentially lethal ranavirus and posed a risk to the collection, they were scheduled for euthanasia and became available for this study.

Frogs were randomly divided into four groups of three animals via simple random sampling. Each group was housed separately in plastic containers measuring 17.8 cm × 11.05 cm × 14.6 cm and covered by a plastic mesh top (Critter-Cages, San Pedro, CA). Temperature was maintained at 24–26°C (74.2–78.8°F) and humidity at 75%. A moist paper towel substrate and a 12 h ambient light cycle were also provided. Animals were maintained on a diet of small crickets and fruit flies.

Enrofloxacin was diluted 1:10 with sterile water for injection (Hospira Inc., Lake Forest, IL) before application to reduce potential skin irritation and ensure accurate dosing. The pH of the solution after dilution was 8.0 and was determined using pH strips (EMD Chemicals Inc., Gibbstown, NJ). Nine animals (three groups) were manually restrained, and a single dose of enrofloxacin (Baytril^®, 10 mg/kg [22.7 mg/mL], Bayer Co., Shawnee Mission, KS; diluted to 2.27 mg/ml in sterile water) was applied over the thoracic dorsum using a micropipetter at time 0. Total volume applied ranged from 0.013 to 0.044 ml. One group (n = 3) was not treated. All animals were weighed before dosing. After treatment, animals were immediately placed back into their enclosures with care taken not to touch the dorsum.

One group was euthanized at time 0 (control), 6, 12, and 24 h after treatment. All animals were euthanized according to American Veterinary Medical Association (2007) guidelines in a solution of tricaine methanosulfate (MS-222, 10 g/l, Argent Chemical Laboratories, Redmond, WA) buffered with sodium bicarbonate (sodium bicarbonate inj, USP strength, 8.4%, Hospira Inc.) to a pH of 7.1–7.2. A gross postmortem necropsy was performed on all animals. The liver and kidneys of each animal were removed, pooled per treatment group, and immediately stored frozen at –86° C until analysis. Samples were analyzed 1 month after collection.

Pooled tissues were weighed and then processed by adding 2.0 ml of 0.1% trifluoroacetic acid solution to each sample as an extraction solvent. The tissue was then lysed with a cell sonicator and homogenized using a hand-held tissue homogenizer with rotating blades (model M133/1281-0 Bio Homogenizer [2 speed], Biospec Products, Inc., Bartlesville, OK). The sample was centrifuged at 1500 × g for 10 min at 25°C (77°F), and 400 μl of the supernatant was removed. The supernatant was then extracted using an HPL solid-phase extraction cartridge (Oasis, Waters Corp., Milford, MA) to elute the drugs. The eluate (in methanol) was then evaporated to dryness under a flow of air at 40°C (104°F) for 20 min, reconstituted with the mobile phase solution (77% deionized water, 23% acetonitrile, and 0.1% frifluoroacetic acid), briefly vortexed, and injected into the high-pressure liquid chromatography system.

Calibration samples were prepared from analytical reference standards. The enrofloxacin reference standard was obtained from the manufacturer (Bayer Health Care, Shawnee, KS), and the ciprofloxacin reference was obtained from the United States Pharmacopeia (Rockville, MD). These reference standards were used to prepare fortified solutions of the extraction solvent. Quality control samples were prepared by fortifying blank (control) tissues from untreated animals at nominal concentrations and processed in a similar manner as the incurred samples. Recovery was measured by fortifying control (blank) tissues with a nominal concentration followed by processing in the same way as for the incurred samples. The concentration of drug was reported as microgram per gram of tissue, wet weight (μg/g).

WCS microbiology reports for bacterial isolates from medical records of amphibian cases collected between 2000 and 2010 at the WCS were reviewed. Isolates were evaluated for bacterial species, source, amphibian species, and enrofloxacin MIC.

No adverse affects were noted in any of the animals during the study. No abnormalities were observed at gross necropsy.

Detectable levels of enrofloxacin and ciprofloxacin were measured in tissues at each posttreatment sample time (Table 1). Enrofloxacin concentration peaked at 6 h, with a decline in concentration observed at successive time points. Ciprofloxacin demonstrated elevated concentrations at 6 and 24 h and a low concentration at 12 h.

Bacterial isolates (n = 201) originated either from animals with clinical disease or from routine quarantine fecal screens, the latter of which included samples collected from pooled specimens. More than 33 amphibian genera (anurans and urodelans) were represented in the review. The most common bacterial isolates were Aeromonas spp., Citrobacter spp., Klebsiella spp., Pseudomonas spp., and Salmonella spp. Enrofloxacin susceptibility patterns for these isolates are shown in Table 2.

For most bacterial species, 90% of the isolates were susceptible to enrofloxacin at a level ≤0.5 μg/ml. The bacteria isolated in this population were common pathogens of amphibians, including Aeromonas sobria, Aeromonas spp., Citrobacter spp., and Klebsiella spp. Several bacterial pathogens showed an intermediate susceptibility to enrofloxacin at 1–2 μg/ml, including P. aeruginosa. Salmonella spp. were frequently found to be resistant to enrofloxacin, with only 82% susceptible at 1 μg/ml.

There was no apparent difference between bacteria isolated and amphibian species, and no evidence of increasing enrofloxacin resistance in any bacterial species over time. There were, however, apparent differences in the location different genera of bacteria were more likely to be isolated, with 63% of Aeromonas spp. cultured from the feces and 17% from coelomic fluid (remaining isolates cultured from the skin or abscesses); 83% of Klebsiella cultured from coelomic fluid; 29% of Pseudomonas spp. cultured from coelomic fluid, 22% from feces, and 19% from skin; and all Salmonella spp. cultured from feces.

This study took the first step toward demonstrating absorption of enrofloxacin after a single topical application at a dose of 10 mg/kg in an amphibian. The dose was selected based on demonstration of serum levels after a single intramuscular administration in the American bullfrog (Rana catesbeiana) (Letcher and Papich, 1994). It is also the recommended high dose for topical administration in several amphibian formularies (Raphael, 1993; Wright and Whitaker, 2001; Wright, 2004; Smith, 2007), and was the dose in a study evaluating intramuscular and subcutaneous injections in African clawed frogs (Xenopus laevis) (Howard et al., 2010).

In domestic dogs and cats, the Clinical Laboratory Standard Institute (2008) breakpoints indicate that bacteria are considered sensitive to enrofloxacin at an MIC level of <0.5 μg/ml, intermediate at an MIC of 1–2 μg/ml, and resistant when the MIC is >4 μg/ml (Papich and Riviere, 2009). No standard breakpoints exist for pathogens from amphibians, but the distributions cited in this study agree with those breakpoints. The 10 yr retrospective review of WCS amphibian bacterial isolates demonstrated that most of the pathogenic bacteria isolated were in the sensitive or intermediate categories. Fluoroquinolone antibiotics such as enrofloxacin act in a concentration-dependent manner, with bactericidal activity evaluated by either the maximum peak concentration (C_MAX) in relation to the bacteria MIC (C_MAX/MIC ratio), or the exposure measured by the area under the curve (AUC) in relation to the MIC (AUC/MIC). Fluoroquinolone efficacy has been associated with a C_MAX/MIC ratio of 8–10, or an AUC/MIC ratio >100 (Papich and Riviere, 2009). Based on our findings, a C_MAX that is a multiple of 1.0 μg/ml should be obtained for effective bactericidal activity.

The current study measured tissue drug concentration not circulating drug concentration. Tissue levels were measured due to the small size of the animals. Liver and kidneys were selected for tissue analysis due to ease of removal, likelihood of similarity between individuals, and overall tissue volume. We acknowledge that pharmacokinetic activity and therapeutic efficacy cannot be predicted by measuring tissue levels alone (Mouton et al., 2008), as distribution among tissues may be tissue dependent. The pharmacokinetic-pharmacodynamic targets such as C_MAX/MIC ratios and AUC/MIC ratios are determined from plasma or serum concentrations, not homogenized tissues. However, it would have been impossible to collect sufficient plasma from these animals for analysis. Fluoroquinolones are relatively uniformly distributed to tissues, with a high volume of distribution observed in other animals (Papich and Riviere, 2009). However, because fluoroquinolones are primarily active in extracellular fluid, and intracellular distribution may occur, we recognize the possibility that these results from tissue concentrations may overestimate plasma drug concentrations (Papich and Riviere, 2009).

Depending on the species, there may be significant conversion of enrofloxacin to its active metabolite ciprofloxacin, which can account for 10–40% of the C_MAX of the total fluoroquinolone concentration (Papich and Riviere, 2009). It has been shown in vitro that the combination of enrofloxacin and ciprofloxacin has additive inhibitory effects for susceptible pathogens (Pirro et al., 1997; Blondeau et al., 2012). Ciprofloxacin is more active against Gram-negative bacteria than enrofloxacin, and it has a significantly lower MIC to P. aeruginosa (0.5 μg/ml) compared with that of enrofloxacin (2.0 μg/ml) (Rubin et al., 2008). In contrast, enrofloxacin is more active against specific Gram-positive bacteria (Staphylococcus spp.) (Papich and Riviere, 2009; Blondeau et al., 2012). It should be noted that some initially sensitive Pseudomonas organisms may develop a high level of resistance to ciprofloxacin due to the single-step mutation seen in this genus (Papich and Riviere, 2009). In this study, significant metabolism of enrofloxacin to ciprofloxacin was demonstrated, with ciprofloxacin levels being greater than enrofloxacin at every time point.

Fluoroquinolone plasma concentrations of 8–10 times the MIC of the targeted bacteria are suggested for best efficacy; and based on the retrospective review of the MIC of bacterial isolates in this study, an MIC of 1.0 μg/ml is a good reference point for comparison. In this study, the C_MAX of tissue levels of both enrofloxacin and ciprofloxacin evaluated separately failed to reach the MIC of 1.0 μg/ml. The combined fluoroquinolone C_MAX of tissue levels range from 0.77 to 1.25 μg/g, which is <2 times a multiple of the MIC. Assuming tissue concentrations of fluoroquinolones are indeed similar to plasma concentrations, then the dose of 10 mg/kg might not be sufficient to achieve therapeutic levels of enrofloxacin after topical application in coqui frogs. However, because this study was not a complete pharmacokinetic study and plasma concentrations were not evaluated, a direct correlation cannot be made regarding dose and therapeutic levels.

This study demonstrated that enrofloxacin may be an ideal medication for topical administration to amphibians. Because it is more lipophilic than other flouroquinolones (e.g., ciprofloxacin and marbofloxacin), it is expected to be more readily absorbed across the skin (Papich and Riviere, 2009). It is likely that ciprofloxacin, being much less lipophilic than enrofloxacin (approximately 100 times less lipophilic), would not penetrate the skin as easily. However, it is also important that conversion to ciprofloxacin occurs because it enhances activity against bacteria known to cause infections in frogs (Blondeau et al., 2012). In this regard, enrofloxacin could be regarded as a prodrug to deliver ciprofloxacin to these animals; and as a result, topical administration in coqui frogs with conversion to ciprofloxacin results in measurable tissue concentrations of both enrofloxacin and ciprofloxacin for at least 24 h. Larger scale pharmacokinetic and efficacy studies, including paired circulating and tissue levels, are required in multiple amphibian species to better evaluate transcutaneous dosing in this taxa.

We thank Lisa Eidlin and the WHC support staff for caring for these frogs. This study was reviewed and approved by the Wildlife Conservation Society Institutional Animal Care and Use Committee. Project 08:01.