The efficacy and pharmacokinetics of terbinafine against the frog-killing fungus (Batrachochytrium dendrobatidis)

The efficacy and pharmacokinetics of terbinafine against the frog-killing fungus... Abstract Captive and wild amphibians are under threat of extinction from the deadly fungal pathogen Batrachochytrium dendrobatidis (Bd). The antifungal drug terbinafine (TBF) is used by pet owners to treat Bd-infected frogs; however, it is not widely used in academic or zoological institutions due to limited veterinary clinical trials. To assess TBF’s efficacy, we undertook treatment trials and pharmacokinetic studies to investigate drug absorption and persistence in frog skin; and then we correlated these data to the minimal lethal concentrations (MLC) against Bd. Despite an initial reduction in zoospore load, the recommended treatment (five daily 5 min 0.01% TBF baths) was unable to cure experimentally infected alpine tree frogs and naturally infected common eastern froglets. In vitro and in vivo pharmacokinetics showed that absorbed TBF accumulates in frog skin with increased exposure, indicating its suitability for treating cutaneous pathogens via direct application. The MLC of TBF for zoosporangia was 100 μg/ml for 2 h, while the minimal inhibitory concentration was 2 μg/ml, suggesting that the drug concentration absorbed during 5 min treatments is not sufficient to cure high Bd burdens. With longer treatments of five daily 30 min baths, Bd clearance improved from 12.5% to 50%. A higher dose of 0.02% TBF resulted in 78% of animals cured; however, clearance was not achieved in all individuals due to low TBF skin persistence, as the half-life was less than 2 h. Therefore, the current TBF regime is not recommended as a universal treatment against Bd until protocols are optimized, such as with increased exposure frequency. antifungal, pharmacokinetics, amphibian declines Introduction The skin disease chytridiomycosis is caused by the fungal pathogens Batrachochytrium dendrobatidis (Bd) and Batrachochytrium salamandrivorans (Bsal).1–3 It is the most devastating disease to vertebrate biodiversity and has been implicated in the extinction of over 90 amphibian species in the past 30 years.4Bd transmits to frogs via an aquatic motile zoospore that, upon encysting on host skin, develops a germination tube that penetrates the stratum corneum and parasitizes cells a few layers deep. The developing zoosporangia mature within epidermal cells and are carried to the surface as the cells keratinize.5,6Bd infection disrupts epidermal electrolyte transport in frogs, resulting in cardiac arrest.7 Topical antifungal treatments for Bd-infected frogs have been developed by adapting mammalian treatments in the absence of amphibian pharmacokinetic data.8 Currently, the most common treatment for Bd infections in captive animals is by itraconazole (Sporanox®) baths.9 However, itraconazole toxicity and side effects, such as skin irritation and osmotic dysfunction have been reported in some species,10–12 although lower doses are now being recommended. In addition, this medicine requires a prescription, which can deter its use by pet owners; and upfront costs are higher for small-scale treatments. Other reported treatments for chytridiomycosis include voriconazole (Vfend®) and chloramphenicol; however, voriconazole is expensive and chloramphenicol requires prolonged baths that are unsuitable for many species.9,13–15 Housing frogs at high temperatures (>28°C) has been successful in clearing Bd in some species16,17 but cannot be used for cold-adapted species, such as alpine frogs. A promising laboratory trial reported an alternative antifungal, terbinafine (TBF), to be 100% effective in clearing Bd infection via five daily 5 min baths at 0.01% (0.1 mg/ml).18 TBF has also been shown to inhibit Bd in vitro, with a minimal inhibitory concentration (MIC) of 63 ng/ml toward zoospores, which was similar to that reported for itraconazole (16–32 ng/ml).19 In another study, a 5 min incubation with 1 mg/ml TBF inhibited zoospore motility, while 5 μg/ml for 5 min was shown to inhibit zoosporangial viability.20 No studies have reported the susceptibility of Bd zoosporangia to TBF at longer incubation times, which is imperative as TBF is likely to remain in the skin for a period after the 5 min drug treatment, as in humans.21 In addition, MIC data for zoosporangia is more likely to be representative of optimal requirements for treating Bd in vivo as it is the parasitic life stage and is more resistant to antimicrobial treatments than zoospores. Few veterinary clinical trials have been undertaken to determine efficacious treatment protocols for Bd-infected frogs, and improvement of antifungal treatment regimes is limited by a lack of pharmacokinetic data on drug bioavailability and persistence.22,23 To address these issues, we tested TBF treatment on infected alpine tree frogs (Litoria verreauxii alpina), and analyzed the pharmacokinetics of TBF penetration, absorption, and persistence in frog skin. Extensive in vitro assays were performed to determine the MIC and minimum lethal concentration (MLC) exposure values of several Bd strains, in both the zoospore and zoosporangial life stages. The pharmacokinetic data were used to optimize the antifungal treatment regime, which was then tested in a clinical trial on naturally infected common eastern froglets (Crinia signifera). This work correlates in vivo concentration and in vitro inhibition data to provide evidence towards developing effective treatment regimes against Bd. Methods Terbinafine clinical treatment trials TBF treatment trials were performed with experimentally infected Litoria verreauxii alpina (alpine tree frogs) and naturally infected Crinia signifera (common eastern froglets). Animals were housed individually in 300 × 195 × 205 mm terraria with gravel and moss substrate, at a room temperature of 20°C. They were fed ad libitum three times weekly with juvenile (10 mm) crickets (Acheta domestica) dusted with amphibian vitamins and mineral supplements (Aristopet). Animals were misted once daily for 60 s with reverse osmosis water. Temperature and humidity were monitored daily. Animal ethics was approved by James Cook University in applications A1897 and A2171. As detailed below, frogs were treated in either (a) a 0.01% w/v terbinafine hydrochloride (TBF.HCl; Fluka, certified reference material, 99.8%)/1% ethanol (Fisher Scientific) bath, or (b) in a commercial formulation of Lamisil® Spray (Novartis Pharmaceuticals, Inc.) that had been diluted 100-fold in distilled water to give a concentration of 0.01% w/v TBF.HCl. Throughout the paper, the formulation of TBF used in each experiment is specified as either “TBF.HCl” indicating the bath was prepared from the salt, or “Lamisil®” where it was prepared by dilution of the commercial formulation. This distinction is important as the latter contains the excipients ethanol, propylene glycol and cetomacrogol 1000.24 In the previous study by Bowerman et al.,18 both formulations were previously used at a maximum concentration of 0.01% w/v TBF.HCl to effectively treat Bd. In this study, no adverse reactions were observed in individuals in any treatment group, indicating the safety of the dosage regimes for these species. Litoria verreauxii alpina Alpine tree frogs (L. v. alpina) are a small frog endemic to the alpine region of southeastern Australia, and are threatened by chytridiomycosis.25 Three year old animals that were in excess to a reintroduction trial were obtained from Taronga Zoo, Sydney, Australia, and were captive raised under strict quarantine protocol. All animals were confirmed as Bd negative prior to the start of the experiment (see methods below). Individuals were experimentally infected with Bd (isolate Waste point- Lverreauxii #2-2013-LB,RW pass 4) harvested from mTGhL agar plates (8 g tryptone (Oxoid), 2 g gelatin hydrolysate (Oxoid), 1 g lactose (Univar) per L of distilled water) after incubation at 22°C for 5 days. Plates were flooded with 3 ml distilled water for 10 min to allow zoospores to be released from zoosporangia. Inoculum was poured off the plates and zoospores were counted using a hemocytometer. Animals were inoculated for 6 h with 3 ml of 1 × 106 zoospore solution, before being transferred back into their individual terraria (n = 14). Bd negative control animals were mock-inoculated using sterile mTGhL agar plates (n = 7). The infected (Bd+) group of L. v. alpina was either bathed in 0.01% TBF.HCl in 1% ethanol (n = 8) or bathed in a control 1% ethanol solution (n = 6) for 5 min for 5 consecutive days. This concentration is the highest reported TBF level used by Bowerman et al. in the previous study.18 The uninfected control L. v. alpina (n = 7) were bathed in 1% ethanol. After each daily treatment, animals were transferred into a new Bd-free enclosure (200 × 150 × 100 mm), lined with damp paper towel substrate. Animals were monitored for infection load via qPCR once a week for 7 weeks (4 weeks after treatment finished). Crinia signifera We undertook a second clinical trial to determine whether longer bath times and higher antifungal concentrations may improve the efficacy of TBF treatments against Bd, as suggested by our in vivo pharmacokinetic data (see methods below). As our initial trial used vulnerable L. v. alpina that were in excess to a reintroduction trial, we were unable to source more individuals of this species for additional treatment experiments. Therefore, our second clinical trial used naturally-infected wild caught C. signifera from the Australian Alps. The Scientific Collection Licence SL101584 was issued by the Office of Environment and Heritage New South Wales. C. signifera is a small ground-dwelling frog widely distributed in Eastern Australia, and can maintain natural Bd infection loads in the wild without mortality.26 Infection was confirmed prior to the start of treatment by quantitative polymerase chain reaction (qPCR) analysis of skin swabs (see methods below). Lamisil® Spray was diluted 100-fold or 50-fold in reverse osmosis water to a concentration of 0.01% or 0.02% w/v TBF.HCl. To test the original recommended treatment regime,18 infected C. signifera were bathed in 0.01% TBF.HCl (n = 8) for 5 min for 5 consecutive days. To test the optimized treatment regimes indicated by pharmacokinetic data, frogs were exposed in daily 30 min baths over 5 consecutive days at either 0.01% TBF.HCl (n = 8), or 0.02% TBF.HCl (n = 9). The no treatment control group was bathed in a 1% ethanol bath for 30 min daily over 5 days (n = 7). After each daily treatment, animals were housed individually in new terraria as above, and tested for Bd presence via qPCR at week 1 and week 4 post treatment (see methods below). Quantification of Bd infection load Bd infection loads were confirmed by quantitative PCR on skin swabs following the method of Boyle et al.27 The extract was analysed using qPCR in singlicate27,28 with a positive and negative control, and a series of zoospore dilution standards. Bd burden was then quantified as “zoospore equivalents” using DNA extracted from standards of known zoospore quantities. We defined “Bd clearance” as a zoospore equivalent of 0, where no Bd DNA was detected from swabbing. In vitro TBF percutaneous absorption via Franz cells Uninfected (n = 8) adult alpine tree frogs (∼40 mm snout-vent length) were euthanized in buffered 0.2% MS-222 solution (tricaine methane sulfonate, Sigma Aldrich). Full thickness ventral skin was removed and the integrity assessed by microscopic examination. Skin samples were mounted in static Franz cells with a 9 mm diameter orifice and 5 ml receptor volume. Due to the size of this species, only one Franz cell experiment was performed per animal. The receptor chamber was filled with 1% ethanol, and the experiment commenced when 1.0 ml of centrifuged donor fluid (0.01% w/v TBF.HCl in 1% ethanol) was added. Four animals were used for testing each condition (5 h or 21 h), with tests performed at 23°C in the dark by covering the Franz cell apparatus in aluminum foil. Upon completion of the experiment, the skin was washed thoroughly with water and frozen at −20°C. TBF was extracted from stored samples by cutting skin into small pieces and sonicating for 5–10 min with 1 to 2 ml of methanol (Fisher Scientific). The supernatant was collected and the cycle repeated a total of five times. Remaining skin was washed with fresh methanol and extracts combined for a final volume of 10 ml. The efficiency of this extraction method was established by high performance liquid chromatography (HPLC). In trial experiments, we confirmed that 98% of extractable TBF was obtained from toad (Rhinella marina) skin within five cycles of extraction. HPLC analysis and terbinafine quantification HPLC was performed on either a Varian ProStar HPLC system with 410 AutoSampler, 240 Solvent Delivery Module and 330 Photodiode Array Detector, or a Shimadzu HPLC system equipped with a SCL-10Avp system controller, a LC-10AT pump, SIL-10A auto sampler and SPD-M10Avp photodiode array detector. HPLC-grade solvents were supplied by Fisher Scientific. Samples were run on a Varian Microsorb-MV 250 × 4.6 mm 100 Å, 5 μm C18 column at 40°C, with an isocratic mobile phase of 95% acetonitrile/0.5% phosphoric acid/4.5% water, at a flow rate of 1.2 ml min−1. HPLC data were collected and analyzed using either Varian Star Chromatography Workstation or Shimadzu Class-VP data collection package. The TBF peak was confirmed by the retention time (8 min) and characteristic absorbance spectrum maxima of 223 nm and 280 nm. A linear standard curve of TBF.HCl (0.075 to 25 μg/ml) was used to quantify TBF in the samples, using the peak area at 223 nm. All Franz cell standards and samples were centrifuged at 15°C and 10 μl triplicate injections were performed. A 50 ng/ml standard was also injected, providing a limit of detection for the HPLC method. Specificity was demonstrated by spiking samples, and by ultraviolet (UV) spectral comparison of sample and standard peaks using PolyView2000 software. Pharmacokinetics of terbinafine absorption and persistence in vivo L. v. alpina were used for pharmacokinetic studies, as the toes of C. signifera were too small to permit additional time course analyses. Uninfected frogs were treated in either a 0.01% w/v TBF.HCl bath (n = 12), as above, or in a commercial formulation of Lamisil® Spray diluted 100-fold in distilled water to 0.01% w/v TBF.HCl (n = 12). Frogs were bathed for 5, 30, 120, or 240 min, and toes were removed at the second phalange from triplicate animals immediately after treatment ended (to quantify how much TBF was absorbed in the skin during treatment), and at 2, 4, and 6 h post-treatment (to determine “persistence” and half-life). We defined “persistence” as the amount of TBF that remained in the skin at a specific time, after the end of treatment. After washing in 1 ml triple distilled water for 1 min to remove residual unbound TBF, toes were photographed under an inverted microscope to determine the 2D surface area using ImageJ, and frozen at −80°C until extraction. TBF from each toe was extracted via bead beating for 1 min in 1 ml methanol with two 3.2 mm stainless steel beads. After centrifugation for 4 min at 12000g, the supernatant was removed and the extraction was repeated with fresh methanol a total of 5 times. Supernatants were combined and evaporated under vacuum until dry. Samples were then resuspended in 200 μl methanol and filtered through a 0.22 μm regenerated cellulose filter. TBF content was determined via HPLC, as above, with analysis of triplicate 20 μl injections. TBF concentrations were reported as μg TBF/mm2 frog skin. To compare in vivo pharmacokinetic data (apparent Cmin) to in vitro MLC values, HPLC data (μg/mm2) were converted to mg/ml concentrations (see details of calculations in supplementary information). Terbinafine minimal inhibitory and lethal exposure concentrations The minimal inhibitory concentration (MIC) of TBF toward Bd zoosporangia was determined via the broth dilution method as previously reported.19 Three fungal isolates from Australia (Waste point- Lverreauxii #2-2013-LB, RW pass 4, Paluma-Lgenimaculata #2-2010-MW pass 10, Couta rocks -Limnodynastes peroni-2009-LB-1 pass 15), and one from the United States (MesquiteWash- Lyavapaiensis-1999-JEL pass 31) were tested for susceptibility to TBF.HCl at 0.0312-25 μg/ml in 1% ethanol, and Lamisil® Spray at a concentration of 2–25 μg TBF.HCl/ml diluted in water. Ethanol (1% v/v) was used as a control. The MIC of TBF was also determined for Bd zoospores using three isolates (Waste point, Paluma, and Arizona). TBF (100 μl in 1% ethanol, final concentrations 0.0312–10 μg/ml) was added directly to the inoculated zoospores, and the growth was observed for 7 days. See supplementary information for further details. In minimal lethal concentration (MLC) experiments, the fungicidal effects of TBF were examined through timed exposure experiments.29 Two Bd isolates (Waste Point and Paluma) were grown in 96 well plates for 2 days. After removing the spent media, 200 μl of diluted Lamisil® Spray (50–200 μg TBF.HCl/ml) was added for exposure periods of: 5 min, 30 min, 1 h, 1.5 h, and 2 h, followed by removal of the antifungal drug and replacement with fresh media. Growth was observed over 7 days. The MLC was determined as the lowest concentration and exposure time that was 100% lethal to both Bd isolates, as indicated by the absence of motile zoospores in all eight replicate wells. Histology Standard histological methods were used to determine alpine tree frog skin thickness and characteristics. Adult female animals were euthanized as described previously. Skin from ventral midline and toes was fixed in 10% phosphate buffered formalin (Fisher Scientific), embedded and sectioned in paraffin wax and then stained with hematoxylin and eosin (Sigma Aldrich). Data analyses Data from the treatment trials, Franz cell experiments and pharmacokinetic studies were analyzed using SPSS (v21), R,30 and Microsoft Excel. See supplementary information for detailed information. Results L. v. alpina terbinafine clinical treatment trial To test the efficacy of the recommended TBF treatment protocol,18Bd-infected L. v. alpina were treated for 5 min daily over 5 days in 0.01% TBF.HCl. The treatment was unable to clear animals of Bd infection, and there was no overall difference between the treated and untreated individuals (P = .087, Suppl. Fig. S1). There was a temporary decrease in infection load of the treated individuals 1 week after treatment (P = .012), correlating to a 96% decrease in zoospore load (60% log10 zoospore equivalents, d = −1.067). However, this decrease was not evident after week 1. See supplementary information for details on statistical analyses. In vitro terbinafine percutaneous absorption and histology To understand why the published TBF treatment regime18 failed in L. v. alpina, we undertook an in vitro study using Franz cells to determine whether absorbed TBF accumulates in frog skin during exposure. TBF levels in the skin of uninfected alpine tree frogs averaged 0.22 μg/mm2 after 5 h of contact, and 0.35 μg/mm2 after 21 h (Fig. 1). The amount of drug retained in the skin was significantly higher (P = .029) with longer contact time. Only 17% of the absorbed TBF was detected in the receptor fluid after 5 h and 25% after 21 h, suggesting that during treatments only low levels of TBF would be absorbed systemically. Receptor TBF levels did not significantly increase between 5 and 21 h (P = .057). Similar amounts of TBF were detected in both the skin and the receptor in infected and uninfected frogs (data not shown). Histology revealed that the surface of the alpine tree frog skin possessed rounded papillae, averaging 280 μm across. The dermis thinned greatly between papillae, so that the skin thickness varied between 80 μm to 250 μm (Fig. 2A). The stratum corneum was approximately 2 μm in healthy skin (Fig. 2B), and varied in thickness in infected skin, up to 24 μm (Fig. 2C). Figure 1. View largeDownload slide Levels of terbinafine (TBF) retained in alpine tree frog skin are four times higher than the receptor fluid in in vitro Franz cell experiments. Antifungal levels in the skin increased with longer exposures (5 h vs 21 h) to TBF.HCl (*, p = 0.029). Error bars signify SEM of experimental replicates (n = 4 for each treatment). Figure 1. View largeDownload slide Levels of terbinafine (TBF) retained in alpine tree frog skin are four times higher than the receptor fluid in in vitro Franz cell experiments. Antifungal levels in the skin increased with longer exposures (5 h vs 21 h) to TBF.HCl (*, p = 0.029). Error bars signify SEM of experimental replicates (n = 4 for each treatment). Figure 2. View largeDownload slide Haematoxylin and eosin histological stain of alpine tree frog (L. v. alpina) skin A) Ventral skin of a healthy frog, as used in Franz cell experiments (50x magnification). The epidermis (E) is ∼60 μm thick, and dramatic thinning of the dermis (D) occurs between papillae, resulting in a varying overall skin thickness (80-250 μm). B) Uninfected toe skin, used in in vivo pharmacokinetic experiments, with a stratum corneum (SC) of 2 μm (400x magnification) and C) Infected toe skin with an SC of up to 24 μm due to the accumulation of hyperkeratotic layers with fungal sporangia (400x magnification). This Figure is reproduced in color in the online version of Medical Mycology. Figure 2. View largeDownload slide Haematoxylin and eosin histological stain of alpine tree frog (L. v. alpina) skin A) Ventral skin of a healthy frog, as used in Franz cell experiments (50x magnification). The epidermis (E) is ∼60 μm thick, and dramatic thinning of the dermis (D) occurs between papillae, resulting in a varying overall skin thickness (80-250 μm). B) Uninfected toe skin, used in in vivo pharmacokinetic experiments, with a stratum corneum (SC) of 2 μm (400x magnification) and C) Infected toe skin with an SC of up to 24 μm due to the accumulation of hyperkeratotic layers with fungal sporangia (400x magnification). This Figure is reproduced in color in the online version of Medical Mycology. In vivo terbinafine absorption in skin during treatment To confirm that TBF levels also accumulate in frog skin during in vivo exposure, we performed detailed pharmacokinetic studies. The levels of TBF detected in frog skin increased with increasing bath time (Fig. 3). After one 5 min bath, approximately 0.06 μg TBF/mm2 frog skin was detected, corresponding to an uptake rate of 1.2 μg/mm2/min. The concentration increased fivefold after the 30 min baths, indicating linear uptake rates during the first 30 min (0.9 μg/mm2/min). After longer bath times, the overall amount of TBF detected in the skin increased, but the rate of absorption per min slowed. There was no significant difference in the amount of TBF absorbed in the skin when using either TBF.HCl in 1% ethanol or with the diluted Lamisil® Spray formulation. The maximum TBF absorbed in the skin was approximately 0.1 μg/mm2 after 4 h (Fig. 3), which is consistent with the in vitro Franz cell data (Fig. 1). Longer bath times increased the amount of TBF detected in the skin, with 4 times more antifungal absorbed during the 30 min bath compared with the 5 min bath. Repeated treatments also increased TBF absorption in frog skin, with a 4-fold increase in TBF levels after five daily 5 min baths compared to the single 5 min bath, and a 14-fold increase after four daily 30 min baths compared with the single treatment (Fig. 5A). Figure 3. View largeDownload slide In vivo absorption of terbinafine (TBF) increases in alpine tree frog skin with increased exposure time; TBF.HCl (white circles) and Lamisil® Spray (black circles). There was no significant difference in absorption of TBF in skin during treatment using TBF.HCl in 1% ethanol or diluted Lamisil® Spray formulation (both at the concentration of 0.01% w/v TBF.HCl). Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 3. View largeDownload slide In vivo absorption of terbinafine (TBF) increases in alpine tree frog skin with increased exposure time; TBF.HCl (white circles) and Lamisil® Spray (black circles). There was no significant difference in absorption of TBF in skin during treatment using TBF.HCl in 1% ethanol or diluted Lamisil® Spray formulation (both at the concentration of 0.01% w/v TBF.HCl). Error bars signify SEM of biological replicates (n = 3 for each treatment). Pharmacokinetics of terbinafine persistence in skin post treatment Pharmocokinetics experiments were also performed to determine the persistence (apparent Cmin) and half-life of TBF in frog skin. Between 55 and 87% TBF (average 75%) was lost from the skin within 2 h post treatment (Fig. 4), indicating a drug half-life of <2 h in frog skin. There was no difference in persistence between skin exposed to TBF.HCl solution or Lamisil® Spray. When frogs were treated with four daily 30 min baths rather than a single 30 min bath, the drug levels were 30-fold higher after 2 hours (due to both higher initial absorption and longer persistence, 54% vs 78% loss) (Fig. 5B). Figure 4. View largeDownload slide Terbinafine (TBF) concentrations in alpine tree frog skin after different in vivo treatments (exposure times: 5 min, 30 min, 2 h or 4 h). Samples were taken for HPLC analysis immediately after treatment, and 2, 4 and 6 h post-treatment. There is a rapid initial loss of approximately 75% TBF within the first 2 hours, followed by a more gradual loss of the remaining drug. A) Treatment with TBF.HCl for 30 min (black circles), 2 h (white squares) or 4 h (grey triangles); B) Treatment with Lamisil® Spray for 5 min (grey diamonds), 30 min (black circles) or 2 h (white squares). LOD was 0.003 μg/mm2. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 4. View largeDownload slide Terbinafine (TBF) concentrations in alpine tree frog skin after different in vivo treatments (exposure times: 5 min, 30 min, 2 h or 4 h). Samples were taken for HPLC analysis immediately after treatment, and 2, 4 and 6 h post-treatment. There is a rapid initial loss of approximately 75% TBF within the first 2 hours, followed by a more gradual loss of the remaining drug. A) Treatment with TBF.HCl for 30 min (black circles), 2 h (white squares) or 4 h (grey triangles); B) Treatment with Lamisil® Spray for 5 min (grey diamonds), 30 min (black circles) or 2 h (white squares). LOD was 0.003 μg/mm2. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 5. View largeDownload slide Longer and repeated bath times help improve terbinafine (TBF) absorption and persistence in alpine tree frog skin. A) TBF absorption in the skin after single 5 min or 30 min baths, or multiple daily 5 × 5 min baths or 4 × 30 min baths of TBF.HCl. More TBF was absorbed during the repeated 30 min treatment than the other treatments (*, p < 0.05). B) Persistence of TBF in the skin over time, after five daily 5 min (white circles), one 30 min (grey circles), or four daily 30 min (black squares) TBF treatments. TBF persisted longer after the repeated 30 min treatments than the other treatment regimes. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 5. View largeDownload slide Longer and repeated bath times help improve terbinafine (TBF) absorption and persistence in alpine tree frog skin. A) TBF absorption in the skin after single 5 min or 30 min baths, or multiple daily 5 × 5 min baths or 4 × 30 min baths of TBF.HCl. More TBF was absorbed during the repeated 30 min treatment than the other treatments (*, p < 0.05). B) Persistence of TBF in the skin over time, after five daily 5 min (white circles), one 30 min (grey circles), or four daily 30 min (black squares) TBF treatments. TBF persisted longer after the repeated 30 min treatments than the other treatment regimes. Error bars signify SEM of biological replicates (n = 3 for each treatment). Terbinafine minimal inhibitory and lethal exposure concentrations The MIC of TBF was determined for both Bd zoospores (motile infective stage) and zoosporangia (parasitic life stage). MIC values were consistent across all of the four Bd strains tested, with a MIC for zoospores of 0.063 μg/ml for both Lamisil® Spray and TBF.HCl solution. The MIC for zoosporangia was at least 30-fold higher at 2 μg/ml for Lamisil® Spray and 6 μg/ml for TBF.HCl (in 1% ethanol). No difference in fungal growth was noted between the commercial spray and the TBF.HCl solution, and in control MIC assays there was no effect of ethanol (≤1%) or propylene glycol (≤1%) alone on fungal growth. Together, these results indicate that the additional components (propylene glycol and cetomacrogol 1000) in the spray had no effect at the dilutions employed, and that TBF primarily contributes to the fungicidal activity of Lamisil® against Bd. Exposure experiments showed that at least 100 μg/ml TBF for 2 h is required to kill Bd zoosporangia in vitro. There were some differences in MLC between the strains tested, with the Waste Point Bd strain being slightly more sensitive to TBF exposure than the Paluma strain: at 100 μg/ml, the Waste Point isolate was killed in 1.5 h and Paluma in 2 h. At 200 μg/ml, Waste Point was killed in 1 h and Paluma in 2 h. Calculation of theoretical Cmin values in vivo To determine whether the absorbed TBF levels that accumulated in frog skin correlated to the minimal lethal concentrations required to kill Bd in vitro, we determined the theoretical Cmin values in the skin (mg/ml) from HPLC values (μg/mm2). True Cmin values could not be calculated as the amount of TBF in skin after 24 h was below the limit of detection (<3 ng/mm2). Two hours after the five daily 5 min treatments, only 2.8 mg/ml TBF was detected in the skin (Table 1), assuming a stratum corneum thickness of 2 μm (Fig. 2B). This in vivo concentration (apparent Cmin at 2 h) is about 30 times the minimum lethal dose required in vitro (100 μg/ml for 2 h). Similar 2 h post treatment concentrations were calculated for the single 30 min and 2 h baths. After four daily 30 min treatments, however, the in vivo concentration was calculated at a maximum of 94.5 mg/ml TBF after 2 hours. This concentration is 950 times higher than the in vitro MLC, and 30 times the amount persisting after five 5 min baths, suggesting that longer repeated TBF baths may be more efficacious in treating frogs with Bd. The overall TBF absorption (μg/mm2) was similar between infected and healthy skin (data not shown), although it is possible that a TBF concentration gradient may form across the thickened stratum corneum of diseased animals (up to 24 μm thick in alpine tree frogs), resulting in a lower in vivo concentration in the lower layers of the epidermis. Table 1. Pharmacokinetic parameters for in vivo terbinafine (TBF) absorption and persistencea in Litoria verreauxii alpina skin. Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 <LOD, below limit of detection at <3 ng/mm2. ND, not determined. aPersistence is defined as the apparent Cmin at a specific time point, post-exposure bMaximum theoretical TBF skin concentration calculated from HPLC values (apparent Cmin at 2 h, ng/mm2), assuming all TBF resides in the stratum corneum (∼2 μm in healthy alpine tree frogs). View Large Table 1. Pharmacokinetic parameters for in vivo terbinafine (TBF) absorption and persistencea in Litoria verreauxii alpina skin. Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 <LOD, below limit of detection at <3 ng/mm2. ND, not determined. aPersistence is defined as the apparent Cmin at a specific time point, post-exposure bMaximum theoretical TBF skin concentration calculated from HPLC values (apparent Cmin at 2 h, ng/mm2), assuming all TBF resides in the stratum corneum (∼2 μm in healthy alpine tree frogs). View Large C. signifera terbinafine clinical treatment trial C. signifera were treated with the original antifungal regime of five daily 5 min Lamisil® baths (diluted to 0.01% w/v TBF.HCl),18 and an extended treatment of five daily 30 min Lamisil® baths at 0.01% or 0.02% TBF.HCl. One week after treatment all animals from the 30 min baths (both 0.01% and 0.02% TBF.HCl groups) were negative for Bd (Suppl. Fig. S2). When swabbed again after 4 weeks, 50% of the 0.01% TBF.HCl treatment group and 78% of the 0.02% TBF.HCl treatment group were negative for Bd, compared to 0% for the untreated control group (Table 2). Only 12.5% of frogs from the 5 min 0.01% TBF.HCl baths had cleared Bd infection after 4 weeks. See supplementary information for details on statistical analyses and average infection intensities (Suppl. Fig. S2). Table 2. Proportion of Bd-negative individuals four weeks after daily Lamisil® treatmentsa of Crinia signifera. Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% aLamisil® was diluted in distilled water to the equivalent concentration of 0.01% TBF.HCl or 0.02% TBF.HCl. View Large Table 2. Proportion of Bd-negative individuals four weeks after daily Lamisil® treatmentsa of Crinia signifera. Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% aLamisil® was diluted in distilled water to the equivalent concentration of 0.01% TBF.HCl or 0.02% TBF.HCl. View Large Discussion Lack of treatment success in Australian frogs using the published TBF regime18 lead us to conduct in-depth pharmacokinetic and treatment trials of this antifungal. The recommended treatment of five daily 5 min 0.01% TBF.HCl baths18 was ineffective at clearing Bd infection in both experimentally infected alpine tree frogs (L. v. alpina) and naturally-infected common eastern froglets (C. signifera). These results are in contrast to the only other study on the efficacy of TBF as a treatment for Bd, where the antifungal drug was effective in six amphibian species.18 One explanation for this discrepancy is that the animals in the previous study may have had lower infection loads, as naturally-infected wild caught less susceptible species were used18,31 and the infection load was not quantified.18 Our study involved both a naturally-infected wild caught species (C. signifera) capable of maintaining high infection loads,26 and a susceptible species (L. v. alpina) experimentally-infected with a large zoospore dose. The 96% temporary decrease in zoospore load in L. v. alpina (Suppl. Fig. S1) 1 week after treatment shows that TBF can greatly reduce Bd burdens, which may be sufficient to treat amphibians with low level infections, as indicated by the previous study.18 An additional source of discrepancy is that in the previous study, the control and test group swabs were combined for PCR analysis post treatment,18 which does not ensure that all untreated control animals retained infection. It is unlikely that the lack of success in clearing infection is due to variations in Bd strain sensitivity to TBF, because our in vitro MIC data were similar for four geographically diverse Bd strains. To optimize TBF treatment protocols and improve efficacy, we undertook in vitro and in vivo pharmacokinetic analysis of TBF absorption and persistence in frog skin. The in vitro Franz cell experiments demonstrated that absorbed TBF accumulates in alpine tree frog skin, and that increased absorption occurred after longer contact times. There was limited systemic absorption of TBF, likely due to the drug's affinity for the keratin within the stratum corneum.32,33 It is therefore critical that the entire body surface of infected frogs contacts the bath solution to ensure direct uptake into all skin areas, rather than relying on redistribution of TBF to skin after systemic absorption. In vivo pharmacokinetic experiments then determined how much of the absorbed TBF accumulated in the skin during the recommended 5 min treatments. TBF absorption increased in the skin with contact time; however, 75% of the antifungal was metabolized within 2 h, followed by a more gradual elimination over several hours. This poor drug persistence may explain why the 5 min treatment did not cure alpine tree frogs of Bd, where zoospore loads recovered to control levels after week 1. A similar, although slower, biphasic elimination of TBF occurs in human skin, with 95% loss within 24 h due to oversaturation of drug in the stratum corneum, and complete elimination within approximately 4 days.34 The more rapid loss of TBF from frog skin is likely due to the thinness of the stratum corneum in amphibians, and frequent sloughing every 2–3 days in L. v. alpina.35 In contrast, a recent paper showed that itraconazole levels are maintained for at least 36 h in the skin of Panamanian golden frogs.23 The longer persistence of itraconazole is likely due to high systemic absorption of that drug, which was not indicated for TBF in our Franz cell experiments. Interestingly, TBF formulation did not appear to have an effect on absorption or persistence. There was no significant difference in TBF absorption when frogs were treated with TBF.HCl in 1% ethanol or with a diluted solution of Lamisil® Spray, which is the form used by private hobbyists. Additionally, there was no difference between the two forms of TBF in the MICs towards zoospores or zoosporangia. Thus, at the dilute concentrations used, the other components in the commercial formulation (propylene glycol, ethanol, and cetomacrogol 1000) have a negligible effect on penetration, persistence, and antifungal activity. The MIC of TBF toward zoosporangia was approximately 2 orders of magnitude higher than that reported previously for zoospores19 and is the more therapeutically relevant level to aim for in skin. The MIC of TBF towards the related fungal pathogen of salamanders, Batrachochytrium salamandrivorans (Bsal), is only 0.2 μg/ml, compared with 6 μg/ml for Bd,36 highlighting the difference in fungal species susceptibility. This lower MIC suggests that TBF treatment may be a more effective strategy against Bsal infections in salamanders than for Bd. TBF exposure assays enabled a MLC value to be determined at 100 μg/ml for 2 h. As our pharmacokinetic data indicated that TBF is unlikely to persist at high concentrations for extended periods in frog skin, these fungicidal levels (rather than MIC values) may be indicative of the concentration required to clear Bd in vivo. Two hours after the recommended five daily 5 min baths,18 TBF levels in the skin were only 30 times the in vitro MLC. Therefore, the inability of the 5 min treatments to consistently cure frogs of Bd may be explained by factors that reduce the in vivo TBF bioavailability (e.g. binding to keratin,32 lipids and membranes), and by variations in Bd localization, skin shedding rate,35 and stratum corneum thickness. Consequently, the accumulated drug concentration absorbed after five 5 min treatments may not be 100% effective as a fungicide, as suggested by the recovery of zoospore load after week 1 of the L. v. alpina treatment trial. The rapid decline in TBF concentrations within the skin indicates that more frequent treatments would be required to sustain effective drug levels in vivo. In fact, our pharmacokinetic experiments indicate that repeated baths improve drug absorption and persistence within frog skin. TBF levels were 14 times higher after four repeated 30 min baths, compared with the single 30 min bath. Two hours after the four daily 30 min baths, TBF levels persisting in the skin were 950-fold higher than the in vitro MLC (compared to only 30 times the MLC persisting after five 5 min baths). To test the efficacy of a longer and repeated treatment regime, we treated C. signifera with five daily 30 min Lamisil® baths at 0.01% or 0.02% TBF.HCl. After 4 weeks, 50–78% of individuals within the 30 min treatment groups completely cleared infection, compared to only 12.5% clearance in animals treated with five 5 min exposures. Therefore, longer baths increase the efficacy TBF against Bd; however, the antifungal did not clear all infected frogs of Bd. It is likely that the poor drug persistence of TBF in frog skin limits its ability to completely cure frogs of high Bd loads, allowing fungal levels to recover after treatment ends. Therefore, our results show that the current TBF regimes are ineffective in completely treating frogs of Bd. Clearance rates improved from 50% to 78% with an increased TBF concentration of 0.02%. Ostensibly, an even higher TBF dose or more repeated exposures may prove to be effectual against Bd by maintaining therapeutic drug levels in the skin. The low apparent Cmin values at 6 h indicate that TBF treatments should be repeated more frequently than every 24 h. However, this optimized regime would require further clinical trials to test efficacy, feasibility, and safety. The results from this study have implications for the protection of wildlife, as optimization of effective Bd treatment protocols requires empirical pharmacological evidence.22 Many endangered frog species rely on captive assurance populations and reintroductions for survival,37 and thus, maintaining disease-free colonies is essential in these zoological institutions. In addition, biosecurity and effective antifungal protocols are critical to preventing the spread of Bd to naive geographical areas, such as Hong Kong38 and Papua New Guinea.39 Finally, antifungal protocols are important tools in front line responses to epidemics, where in situ treatments have been used to cure frogs in the wild.40 Therefore, our nonlethal method of using toe clips to quantify antifungal drug levels could be useful in optimizing other chytridiomycosis treatments and improve the efficacy for endangered frogs; as fewer animals are required than for whole skin analysis,23 and drug concentrations in each individual can be quantified over time. This study highlights the importance of both pharmacokinetic data and controlled veterinary clinical trials in determining the most efficient and successful treatment for disease in amphibians. Pharmacokinetic studies, while common for mammalian treatments, are lacking for amphibians, and as the two skin structures differ,8 treatments need to be tested using an amphibian model. This in-depth series of experiments explored the absorption and persistence of TBF in amphibian skin, and demonstrated that the current recommended dose and treatment time18 is not effective for all amphibian species. While our protocols increased the efficacy of TBF in treating Bd infections, the low persistence of this antifungal in frog skin appears to limit its ability to clear high pathogen loads. Due to TBF’s accessibility without prescription, low up-front costs, and ease of use, it is likely that private hobbyists may continue its use in treating amphibians. The results from our study, however, show that the current TBF protocols are not 100% effective against Bd in all amphibian species. More frequent treatments have the potential to improve clearance rates, but protocols need to be optimized before TBF can be used with confidence by zoological or academic institutions to universally treat frogs with chytridiomycosis. Acknowledgment R. Speare, who initiated our collaboration, tragically died in a car accident in 2016 and his vision in bringing together disparate expertise to better solve real world problems will be greatly missed. We thank P. Harlow and Taronga Zoo for raising the L. v. alpina, B. Scheele for providing the C. signifera, P. Thomas-Hall and V. Llewellyn for technical assistance and helpful discussions, and J. Longcore for providing the Arizona Bd isolate. Thank you to M. Souza and S. Cox for advice on terbinafine inhibitory assays and quantitative analysis. We thank D. Tegtmeier, C. De Jong, J. Hawkes, K. Fossen, S. Percival, M. McWilliams, L. Bertola, M. Stewart, N. Harney, and T. Knavel for data collection and husbandry assistance. Funding This work was supported by a Queensland Government Accelerate Fellowship [14-218 to A.A.R.] with additional funds from the Taronga Conservation Science Initiative and the Queensland Department of Environment and Heritage Protection; the Australian Research Council [FT100100375, LP110200240, DP120100811 to LFS and LB] with additional funds from industry partners Taronga Zoo and New South Wales Office of Environment and Heritage; the Advance Queensland Women's Academic Fund; and a Collaboration Across Boundaries Grant provided by James Cook University. Supplementary material Supplementary data are available at MMYCOL online. Declaration of Interest The authors report no conflicts of interest. The authors alone are responsible for the content and the writing of this paper. References 1. Berger L , Speare R , Daszak P et al. Chytridiomycosis causes amphibian mortality associated with population declines in the rainforests of Australia and Central America . Proc Natl Acad Sci USA . 1998 ; 95 : 9031 – 9036 . Google Scholar CrossRef Search ADS PubMed 2. Skerratt LF , Berger L , Speare R et al. Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs . EcoHealth . 2007 ; 4 : 125 – 134 . Google Scholar CrossRef Search ADS 3. Martel A , Blooi M , Adriaensen C et al. Recent introduction of a chytrid fungus endangers Western Palearctic salamanders . Science . 2014 ; 346 : 630 – 631 . Google Scholar CrossRef Search ADS PubMed 4. Pennisi E . Life and death play out on the skins of frogs . Science . 2009 ; 326 : 507 – 508 . Google Scholar CrossRef Search ADS PubMed 5. Berger L , Hyatt AD , Speare R , Longcore JE . Life cycle stages of the amphibian chytrid Batrachochytrium dendrobatidis . Dis Aquat Org . 2005 ; 68 : 51 – 63 . Google Scholar CrossRef Search ADS PubMed 6. Greenspan SE , Longcore JE , Calhoun AJK . Host invasion by Batrachochytrium dendrobatidis: Fungal and epidermal ultrastructure in model anurans . Dis Aquat Org . 2012 ; 100 : 201 – 210 . Google Scholar CrossRef Search ADS PubMed 7. Voyles J , Young S , Berger L et al. Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines . Science . 2009 ; 326 : 582 – 585 . Google Scholar CrossRef Search ADS PubMed 8. Llewelyn V , Berger L , Glass B . Percutaneous absorption of chemicals: Developing an understanding for the treatment of disease in frogs . J Vet Pharmacol Ther . 2015 ; 39 : 109 – 121 . Google Scholar CrossRef Search ADS PubMed 9. Baitchman EJ , Pessier AP . Pathogenesis, diagnosis, and treatment of amphibian chytridiomycosis . Vet Clin North Am Exot Anim Pract . 2013 ; 16 : 669 – 685 . Google Scholar CrossRef Search ADS PubMed 10. Brannelly LA , Richards-Zawacki CL , Pessier AP . Clinical trials with itraconazole as a treatment for chytrid fungal infections in amphibians . Dis Aquat Org . 2012 ; 101 : 95 – 104 . Google Scholar CrossRef Search ADS PubMed 11. Woodhams DC , Geiger CC , Reinert LK et al. Treatment of amphibians infected with chytrid fungus: Learning from failed trials with antimicrobial peptides, bacteria, and heat therapy . Dis Aquat Org . 2012 ; 98 : 11 – 25 . Google Scholar CrossRef Search ADS PubMed 12. Brannelly LA . Reduced intraconazole concentration and durations are successful in treating Batrachochytrium dendrobatidis infection in amphibians . J Vis Exp . 2014 ; 85 : e51166 . 13. Martel A , van Rooij P , Vercauteren G et al. Developing a safe antifungal treatment protocol to eliminate Batrachochytrium dendrobatidis from amphibians . Med Mycol . 2010 ; 49 : 143 – 149 . Google Scholar CrossRef Search ADS PubMed 14. Holden WM , Ebert AR , Canning PF , Rollins-Smith LA . Evaluation of amphotericin B and chloramphenicol as alternative drugs for treatment of chytridiomycosis and their impacts on innate skin defenses . Appl Environ Microbiol . 2014 ; 80 : 4034 – 4041 . Google Scholar CrossRef Search ADS PubMed 15. Brannelly LA , Skerratt LF , Berger L . Treatment trial of clinically ill corroboree frogs with chytridiomycosis with two triazole antifungals and electrolyte therapy . Vet Res Commun . 2015 ; 39 : 179 – 187 . Google Scholar CrossRef Search ADS PubMed 16. Berger L , Speare R , Hines HB et al. Effect of season and temperature on mortality in amphibians due to chytridiomycosis . Aust Vet J . 2004 ; 82 : 434 – 439 . Google Scholar CrossRef Search ADS PubMed 17. Chatfield MWH , Richards-Zawacki CL . Elevated temperature as a treatment for Batrachochytrium dendrobatidis infection in captive frogs . Dis Aquat Org . 2011 ; 94 : 235 – 238 . Google Scholar CrossRef Search ADS PubMed 18. Bowerman J , Rombough C , Weinstock SR , Padgett-Flohr GE . Terbinafine hydrochloride in ethanol effectively clears Batrachochytrium dendrobatidis in amphibians . J Herpetol Med Surgery . 2010 ; 20 : 24 – 28 . Google Scholar CrossRef Search ADS 19. Woodward A , Berger L , Skerratt LF . In vitro sensitivity of the amphibian pathogen Batrachochytrium dendrobatidis to antifungal therapeutics . Res Vet Sci . 2014 ; 97 : 365 – 367 . Google Scholar CrossRef Search ADS 20. Gold KK , Reed PD , Bemis DA et al. Efficacy of common disinfectants and terbinafine in inactivating the growth of Batrachochytrium dendrobatidis in culture . Dis Aquat Org . 2013 ; 107 : 77 – 81 . Google Scholar CrossRef Search ADS PubMed 21. Hill S , Thomas R , Smith SG , Finlay AY . An investigation of the pharmacokinetics of topical terbinafine (Lamisil) 1% cream . Br J Dermatol . 1992 ; 127 : 396 – 400 . Google Scholar CrossRef Search ADS PubMed 22. Berger L , Speare R , Pessier AP , Voyles J , Skerratt LF . Treatment of chytridiomycosis requires urgent clinical trials . Dis Aquat Org . 2010 ; 92 : 165 – 174 . Google Scholar CrossRef Search ADS PubMed 23. Rifkin A , Visser M , Barrett K , Boothe D , Bronson, E The pharmacokinetics of topical itraconazole in Panamanian golden frogs (Atelopus zeteki) . J Zoo Wildlife Med . 2017 ; 48 : 344 – 351 . Google Scholar CrossRef Search ADS 24. Consumer Information Lamisil . 2016 ; Novartis Pharmaceuticals Inc . URL www.ask.novartispharma.ca/download.htm?res=lamisil_patient_e.pdf&resTitleId=137 . 25. Brannelly LA , Hunter D , Lenger D et al. Dynamics of chytridiomycosis during the breeding season in an Australian Alpine amphibian . PLoS ONE . 2015 ; 10 : e0143629 . Google Scholar CrossRef Search ADS PubMed 26. Scheele BC , Hunter DA , Brannelly LA , Skerratt LF , Driscoll DA . Reservoir-host amplification of disease impact in an endangered amphibian . Conserv. Biol. 2016 ; DOI: https://doi.org/10.1111/cobi.12830 . 27. Boyle DG , Boyle DB , Olsen V , Morgan JAT , Hyatt AD . Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay . Dis Aquat Org . 2004 ; 60 : 141 – 148 . Google Scholar CrossRef Search ADS PubMed 28. Kriger KM , Hero J-M , Ashton KJ . Cost efficiency in the detection of chytridiomycosis using PCR assay . Dis Aquat Org . 2006 ; 71 : 149 – 154 . Google Scholar CrossRef Search ADS PubMed 29. Webb R , Philips A , Speare R , Connolly J , Berger L . Controlling wildlife fungal disease spread: in vitro efficacy of disinfectants against Batrachochytrium dendrobatidis and Mucor amphibiorum . Dis Aquat Org . 2012 ; 99 : 119 – 125 . Google Scholar CrossRef Search ADS PubMed 30. R Core Team . R: A language and environment for statistical computing . R Foundation for Statistical Computing . 2013 ; Vienna, Austria . URL http://www.R-project.org/ . 31. Davidson EW , Parris M , Collins JP et al. Pathogenicity and transmission of chytridioycosis in tiger salamanders (Ambystoma tigrinum) . Copeia . 2003 ; 3 : 601 – 607 . Google Scholar CrossRef Search ADS 32. Tatsumi Y , Yokoo M , Senda H , Kakehi K . Therapeutic efficacy of topically applied KP-103 against experimental tinea unguium in guinea pigs in comparison with amorolfine and terbinafine . Antimicrob Agents Chemother . 2002 ; 46 : 3797 – 3801 . Google Scholar CrossRef Search ADS PubMed 33. Schäfer-Korting M , Schoellmann C , Korting HC . Fungicidal activity plus reservoir effect allow short treatment courses with terbinafine in tinea pedis . Skin Pharmacol Phys . 2008 ; 21 : 203 – 210 . Google Scholar CrossRef Search ADS 34. Faergermann J , Zehender H , Millerioux L . Levels of terbinafine in plasma, stratum corneum, dermis–epidermis (without stratum corneum), sebum, hair and nails during and after 250 mg terbinafine orally once daily for 7 and 14 days . Clin Exp Dermatol . 1994 ; 19 : 121 – 126 . Google Scholar CrossRef Search ADS PubMed 35. Ohmer MEB . Interactions between amphibian skin sloughing and a cutaneous fungal disease: Infection progression, immune defence, and phylogenetic patterns . School of Biological Sciences . 2016 ; PhD diss . 143 . 36. Blooi M , Pasmans F , Rouffaer L et al. Successful treatment of Batrachochytrium salamandrivorans infections in salamanders requires synergy between voriconazole, polymyxin E and temperature . Sci Rep . 2015 ; 5 : 11788 . Google Scholar CrossRef Search ADS PubMed 37. Griffiths RA , Pavajeau L . Captive breeding, reintroduction, and the conservation of amphibians . Conserv Biol . 2008 ; 22 : 852 – 861 . Google Scholar CrossRef Search ADS PubMed 38. Rowley J , Chan S , Tang W et al. Survey for the amphibian chytrid Batrachochytrium dendrobatidis in Hong Kong in native amphibians and in the international amphibian trade . Dis Aquat Org . 2007 ; 78 : 87 – 95 . Google Scholar CrossRef Search ADS PubMed 39. Bower D , Lips K , Schwarzkopf L , Georges A , Clulow S . Amphibians on the brink . Science . 2017 ; 357 : 454 – 455 . Google Scholar CrossRef Search ADS PubMed 40. Hardy B , Pope K , Piovia-Scott J , Brown R , Foley J . Itraconazole treatment reduces Batrachochytrium dendrobatidis prevalence and increases overwinter field survival in juvenile Cascades frogs . Dis Aquat Org . 2015 ; 112 : 243 – 250 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Medical Mycology Oxford University Press

The efficacy and pharmacokinetics of terbinafine against the frog-killing fungus (Batrachochytrium dendrobatidis)

Loading next page...
 
/lp/ou_press/the-efficacy-and-pharmacokinetics-of-terbinafine-against-the-frog-0g9a5ayeOd
Publisher
Taylor & Francis
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1369-3786
eISSN
1460-2709
D.O.I.
10.1093/mmy/myy010
Publisher site
See Article on Publisher Site

Abstract

Abstract Captive and wild amphibians are under threat of extinction from the deadly fungal pathogen Batrachochytrium dendrobatidis (Bd). The antifungal drug terbinafine (TBF) is used by pet owners to treat Bd-infected frogs; however, it is not widely used in academic or zoological institutions due to limited veterinary clinical trials. To assess TBF’s efficacy, we undertook treatment trials and pharmacokinetic studies to investigate drug absorption and persistence in frog skin; and then we correlated these data to the minimal lethal concentrations (MLC) against Bd. Despite an initial reduction in zoospore load, the recommended treatment (five daily 5 min 0.01% TBF baths) was unable to cure experimentally infected alpine tree frogs and naturally infected common eastern froglets. In vitro and in vivo pharmacokinetics showed that absorbed TBF accumulates in frog skin with increased exposure, indicating its suitability for treating cutaneous pathogens via direct application. The MLC of TBF for zoosporangia was 100 μg/ml for 2 h, while the minimal inhibitory concentration was 2 μg/ml, suggesting that the drug concentration absorbed during 5 min treatments is not sufficient to cure high Bd burdens. With longer treatments of five daily 30 min baths, Bd clearance improved from 12.5% to 50%. A higher dose of 0.02% TBF resulted in 78% of animals cured; however, clearance was not achieved in all individuals due to low TBF skin persistence, as the half-life was less than 2 h. Therefore, the current TBF regime is not recommended as a universal treatment against Bd until protocols are optimized, such as with increased exposure frequency. antifungal, pharmacokinetics, amphibian declines Introduction The skin disease chytridiomycosis is caused by the fungal pathogens Batrachochytrium dendrobatidis (Bd) and Batrachochytrium salamandrivorans (Bsal).1–3 It is the most devastating disease to vertebrate biodiversity and has been implicated in the extinction of over 90 amphibian species in the past 30 years.4Bd transmits to frogs via an aquatic motile zoospore that, upon encysting on host skin, develops a germination tube that penetrates the stratum corneum and parasitizes cells a few layers deep. The developing zoosporangia mature within epidermal cells and are carried to the surface as the cells keratinize.5,6Bd infection disrupts epidermal electrolyte transport in frogs, resulting in cardiac arrest.7 Topical antifungal treatments for Bd-infected frogs have been developed by adapting mammalian treatments in the absence of amphibian pharmacokinetic data.8 Currently, the most common treatment for Bd infections in captive animals is by itraconazole (Sporanox®) baths.9 However, itraconazole toxicity and side effects, such as skin irritation and osmotic dysfunction have been reported in some species,10–12 although lower doses are now being recommended. In addition, this medicine requires a prescription, which can deter its use by pet owners; and upfront costs are higher for small-scale treatments. Other reported treatments for chytridiomycosis include voriconazole (Vfend®) and chloramphenicol; however, voriconazole is expensive and chloramphenicol requires prolonged baths that are unsuitable for many species.9,13–15 Housing frogs at high temperatures (>28°C) has been successful in clearing Bd in some species16,17 but cannot be used for cold-adapted species, such as alpine frogs. A promising laboratory trial reported an alternative antifungal, terbinafine (TBF), to be 100% effective in clearing Bd infection via five daily 5 min baths at 0.01% (0.1 mg/ml).18 TBF has also been shown to inhibit Bd in vitro, with a minimal inhibitory concentration (MIC) of 63 ng/ml toward zoospores, which was similar to that reported for itraconazole (16–32 ng/ml).19 In another study, a 5 min incubation with 1 mg/ml TBF inhibited zoospore motility, while 5 μg/ml for 5 min was shown to inhibit zoosporangial viability.20 No studies have reported the susceptibility of Bd zoosporangia to TBF at longer incubation times, which is imperative as TBF is likely to remain in the skin for a period after the 5 min drug treatment, as in humans.21 In addition, MIC data for zoosporangia is more likely to be representative of optimal requirements for treating Bd in vivo as it is the parasitic life stage and is more resistant to antimicrobial treatments than zoospores. Few veterinary clinical trials have been undertaken to determine efficacious treatment protocols for Bd-infected frogs, and improvement of antifungal treatment regimes is limited by a lack of pharmacokinetic data on drug bioavailability and persistence.22,23 To address these issues, we tested TBF treatment on infected alpine tree frogs (Litoria verreauxii alpina), and analyzed the pharmacokinetics of TBF penetration, absorption, and persistence in frog skin. Extensive in vitro assays were performed to determine the MIC and minimum lethal concentration (MLC) exposure values of several Bd strains, in both the zoospore and zoosporangial life stages. The pharmacokinetic data were used to optimize the antifungal treatment regime, which was then tested in a clinical trial on naturally infected common eastern froglets (Crinia signifera). This work correlates in vivo concentration and in vitro inhibition data to provide evidence towards developing effective treatment regimes against Bd. Methods Terbinafine clinical treatment trials TBF treatment trials were performed with experimentally infected Litoria verreauxii alpina (alpine tree frogs) and naturally infected Crinia signifera (common eastern froglets). Animals were housed individually in 300 × 195 × 205 mm terraria with gravel and moss substrate, at a room temperature of 20°C. They were fed ad libitum three times weekly with juvenile (10 mm) crickets (Acheta domestica) dusted with amphibian vitamins and mineral supplements (Aristopet). Animals were misted once daily for 60 s with reverse osmosis water. Temperature and humidity were monitored daily. Animal ethics was approved by James Cook University in applications A1897 and A2171. As detailed below, frogs were treated in either (a) a 0.01% w/v terbinafine hydrochloride (TBF.HCl; Fluka, certified reference material, 99.8%)/1% ethanol (Fisher Scientific) bath, or (b) in a commercial formulation of Lamisil® Spray (Novartis Pharmaceuticals, Inc.) that had been diluted 100-fold in distilled water to give a concentration of 0.01% w/v TBF.HCl. Throughout the paper, the formulation of TBF used in each experiment is specified as either “TBF.HCl” indicating the bath was prepared from the salt, or “Lamisil®” where it was prepared by dilution of the commercial formulation. This distinction is important as the latter contains the excipients ethanol, propylene glycol and cetomacrogol 1000.24 In the previous study by Bowerman et al.,18 both formulations were previously used at a maximum concentration of 0.01% w/v TBF.HCl to effectively treat Bd. In this study, no adverse reactions were observed in individuals in any treatment group, indicating the safety of the dosage regimes for these species. Litoria verreauxii alpina Alpine tree frogs (L. v. alpina) are a small frog endemic to the alpine region of southeastern Australia, and are threatened by chytridiomycosis.25 Three year old animals that were in excess to a reintroduction trial were obtained from Taronga Zoo, Sydney, Australia, and were captive raised under strict quarantine protocol. All animals were confirmed as Bd negative prior to the start of the experiment (see methods below). Individuals were experimentally infected with Bd (isolate Waste point- Lverreauxii #2-2013-LB,RW pass 4) harvested from mTGhL agar plates (8 g tryptone (Oxoid), 2 g gelatin hydrolysate (Oxoid), 1 g lactose (Univar) per L of distilled water) after incubation at 22°C for 5 days. Plates were flooded with 3 ml distilled water for 10 min to allow zoospores to be released from zoosporangia. Inoculum was poured off the plates and zoospores were counted using a hemocytometer. Animals were inoculated for 6 h with 3 ml of 1 × 106 zoospore solution, before being transferred back into their individual terraria (n = 14). Bd negative control animals were mock-inoculated using sterile mTGhL agar plates (n = 7). The infected (Bd+) group of L. v. alpina was either bathed in 0.01% TBF.HCl in 1% ethanol (n = 8) or bathed in a control 1% ethanol solution (n = 6) for 5 min for 5 consecutive days. This concentration is the highest reported TBF level used by Bowerman et al. in the previous study.18 The uninfected control L. v. alpina (n = 7) were bathed in 1% ethanol. After each daily treatment, animals were transferred into a new Bd-free enclosure (200 × 150 × 100 mm), lined with damp paper towel substrate. Animals were monitored for infection load via qPCR once a week for 7 weeks (4 weeks after treatment finished). Crinia signifera We undertook a second clinical trial to determine whether longer bath times and higher antifungal concentrations may improve the efficacy of TBF treatments against Bd, as suggested by our in vivo pharmacokinetic data (see methods below). As our initial trial used vulnerable L. v. alpina that were in excess to a reintroduction trial, we were unable to source more individuals of this species for additional treatment experiments. Therefore, our second clinical trial used naturally-infected wild caught C. signifera from the Australian Alps. The Scientific Collection Licence SL101584 was issued by the Office of Environment and Heritage New South Wales. C. signifera is a small ground-dwelling frog widely distributed in Eastern Australia, and can maintain natural Bd infection loads in the wild without mortality.26 Infection was confirmed prior to the start of treatment by quantitative polymerase chain reaction (qPCR) analysis of skin swabs (see methods below). Lamisil® Spray was diluted 100-fold or 50-fold in reverse osmosis water to a concentration of 0.01% or 0.02% w/v TBF.HCl. To test the original recommended treatment regime,18 infected C. signifera were bathed in 0.01% TBF.HCl (n = 8) for 5 min for 5 consecutive days. To test the optimized treatment regimes indicated by pharmacokinetic data, frogs were exposed in daily 30 min baths over 5 consecutive days at either 0.01% TBF.HCl (n = 8), or 0.02% TBF.HCl (n = 9). The no treatment control group was bathed in a 1% ethanol bath for 30 min daily over 5 days (n = 7). After each daily treatment, animals were housed individually in new terraria as above, and tested for Bd presence via qPCR at week 1 and week 4 post treatment (see methods below). Quantification of Bd infection load Bd infection loads were confirmed by quantitative PCR on skin swabs following the method of Boyle et al.27 The extract was analysed using qPCR in singlicate27,28 with a positive and negative control, and a series of zoospore dilution standards. Bd burden was then quantified as “zoospore equivalents” using DNA extracted from standards of known zoospore quantities. We defined “Bd clearance” as a zoospore equivalent of 0, where no Bd DNA was detected from swabbing. In vitro TBF percutaneous absorption via Franz cells Uninfected (n = 8) adult alpine tree frogs (∼40 mm snout-vent length) were euthanized in buffered 0.2% MS-222 solution (tricaine methane sulfonate, Sigma Aldrich). Full thickness ventral skin was removed and the integrity assessed by microscopic examination. Skin samples were mounted in static Franz cells with a 9 mm diameter orifice and 5 ml receptor volume. Due to the size of this species, only one Franz cell experiment was performed per animal. The receptor chamber was filled with 1% ethanol, and the experiment commenced when 1.0 ml of centrifuged donor fluid (0.01% w/v TBF.HCl in 1% ethanol) was added. Four animals were used for testing each condition (5 h or 21 h), with tests performed at 23°C in the dark by covering the Franz cell apparatus in aluminum foil. Upon completion of the experiment, the skin was washed thoroughly with water and frozen at −20°C. TBF was extracted from stored samples by cutting skin into small pieces and sonicating for 5–10 min with 1 to 2 ml of methanol (Fisher Scientific). The supernatant was collected and the cycle repeated a total of five times. Remaining skin was washed with fresh methanol and extracts combined for a final volume of 10 ml. The efficiency of this extraction method was established by high performance liquid chromatography (HPLC). In trial experiments, we confirmed that 98% of extractable TBF was obtained from toad (Rhinella marina) skin within five cycles of extraction. HPLC analysis and terbinafine quantification HPLC was performed on either a Varian ProStar HPLC system with 410 AutoSampler, 240 Solvent Delivery Module and 330 Photodiode Array Detector, or a Shimadzu HPLC system equipped with a SCL-10Avp system controller, a LC-10AT pump, SIL-10A auto sampler and SPD-M10Avp photodiode array detector. HPLC-grade solvents were supplied by Fisher Scientific. Samples were run on a Varian Microsorb-MV 250 × 4.6 mm 100 Å, 5 μm C18 column at 40°C, with an isocratic mobile phase of 95% acetonitrile/0.5% phosphoric acid/4.5% water, at a flow rate of 1.2 ml min−1. HPLC data were collected and analyzed using either Varian Star Chromatography Workstation or Shimadzu Class-VP data collection package. The TBF peak was confirmed by the retention time (8 min) and characteristic absorbance spectrum maxima of 223 nm and 280 nm. A linear standard curve of TBF.HCl (0.075 to 25 μg/ml) was used to quantify TBF in the samples, using the peak area at 223 nm. All Franz cell standards and samples were centrifuged at 15°C and 10 μl triplicate injections were performed. A 50 ng/ml standard was also injected, providing a limit of detection for the HPLC method. Specificity was demonstrated by spiking samples, and by ultraviolet (UV) spectral comparison of sample and standard peaks using PolyView2000 software. Pharmacokinetics of terbinafine absorption and persistence in vivo L. v. alpina were used for pharmacokinetic studies, as the toes of C. signifera were too small to permit additional time course analyses. Uninfected frogs were treated in either a 0.01% w/v TBF.HCl bath (n = 12), as above, or in a commercial formulation of Lamisil® Spray diluted 100-fold in distilled water to 0.01% w/v TBF.HCl (n = 12). Frogs were bathed for 5, 30, 120, or 240 min, and toes were removed at the second phalange from triplicate animals immediately after treatment ended (to quantify how much TBF was absorbed in the skin during treatment), and at 2, 4, and 6 h post-treatment (to determine “persistence” and half-life). We defined “persistence” as the amount of TBF that remained in the skin at a specific time, after the end of treatment. After washing in 1 ml triple distilled water for 1 min to remove residual unbound TBF, toes were photographed under an inverted microscope to determine the 2D surface area using ImageJ, and frozen at −80°C until extraction. TBF from each toe was extracted via bead beating for 1 min in 1 ml methanol with two 3.2 mm stainless steel beads. After centrifugation for 4 min at 12000g, the supernatant was removed and the extraction was repeated with fresh methanol a total of 5 times. Supernatants were combined and evaporated under vacuum until dry. Samples were then resuspended in 200 μl methanol and filtered through a 0.22 μm regenerated cellulose filter. TBF content was determined via HPLC, as above, with analysis of triplicate 20 μl injections. TBF concentrations were reported as μg TBF/mm2 frog skin. To compare in vivo pharmacokinetic data (apparent Cmin) to in vitro MLC values, HPLC data (μg/mm2) were converted to mg/ml concentrations (see details of calculations in supplementary information). Terbinafine minimal inhibitory and lethal exposure concentrations The minimal inhibitory concentration (MIC) of TBF toward Bd zoosporangia was determined via the broth dilution method as previously reported.19 Three fungal isolates from Australia (Waste point- Lverreauxii #2-2013-LB, RW pass 4, Paluma-Lgenimaculata #2-2010-MW pass 10, Couta rocks -Limnodynastes peroni-2009-LB-1 pass 15), and one from the United States (MesquiteWash- Lyavapaiensis-1999-JEL pass 31) were tested for susceptibility to TBF.HCl at 0.0312-25 μg/ml in 1% ethanol, and Lamisil® Spray at a concentration of 2–25 μg TBF.HCl/ml diluted in water. Ethanol (1% v/v) was used as a control. The MIC of TBF was also determined for Bd zoospores using three isolates (Waste point, Paluma, and Arizona). TBF (100 μl in 1% ethanol, final concentrations 0.0312–10 μg/ml) was added directly to the inoculated zoospores, and the growth was observed for 7 days. See supplementary information for further details. In minimal lethal concentration (MLC) experiments, the fungicidal effects of TBF were examined through timed exposure experiments.29 Two Bd isolates (Waste Point and Paluma) were grown in 96 well plates for 2 days. After removing the spent media, 200 μl of diluted Lamisil® Spray (50–200 μg TBF.HCl/ml) was added for exposure periods of: 5 min, 30 min, 1 h, 1.5 h, and 2 h, followed by removal of the antifungal drug and replacement with fresh media. Growth was observed over 7 days. The MLC was determined as the lowest concentration and exposure time that was 100% lethal to both Bd isolates, as indicated by the absence of motile zoospores in all eight replicate wells. Histology Standard histological methods were used to determine alpine tree frog skin thickness and characteristics. Adult female animals were euthanized as described previously. Skin from ventral midline and toes was fixed in 10% phosphate buffered formalin (Fisher Scientific), embedded and sectioned in paraffin wax and then stained with hematoxylin and eosin (Sigma Aldrich). Data analyses Data from the treatment trials, Franz cell experiments and pharmacokinetic studies were analyzed using SPSS (v21), R,30 and Microsoft Excel. See supplementary information for detailed information. Results L. v. alpina terbinafine clinical treatment trial To test the efficacy of the recommended TBF treatment protocol,18Bd-infected L. v. alpina were treated for 5 min daily over 5 days in 0.01% TBF.HCl. The treatment was unable to clear animals of Bd infection, and there was no overall difference between the treated and untreated individuals (P = .087, Suppl. Fig. S1). There was a temporary decrease in infection load of the treated individuals 1 week after treatment (P = .012), correlating to a 96% decrease in zoospore load (60% log10 zoospore equivalents, d = −1.067). However, this decrease was not evident after week 1. See supplementary information for details on statistical analyses. In vitro terbinafine percutaneous absorption and histology To understand why the published TBF treatment regime18 failed in L. v. alpina, we undertook an in vitro study using Franz cells to determine whether absorbed TBF accumulates in frog skin during exposure. TBF levels in the skin of uninfected alpine tree frogs averaged 0.22 μg/mm2 after 5 h of contact, and 0.35 μg/mm2 after 21 h (Fig. 1). The amount of drug retained in the skin was significantly higher (P = .029) with longer contact time. Only 17% of the absorbed TBF was detected in the receptor fluid after 5 h and 25% after 21 h, suggesting that during treatments only low levels of TBF would be absorbed systemically. Receptor TBF levels did not significantly increase between 5 and 21 h (P = .057). Similar amounts of TBF were detected in both the skin and the receptor in infected and uninfected frogs (data not shown). Histology revealed that the surface of the alpine tree frog skin possessed rounded papillae, averaging 280 μm across. The dermis thinned greatly between papillae, so that the skin thickness varied between 80 μm to 250 μm (Fig. 2A). The stratum corneum was approximately 2 μm in healthy skin (Fig. 2B), and varied in thickness in infected skin, up to 24 μm (Fig. 2C). Figure 1. View largeDownload slide Levels of terbinafine (TBF) retained in alpine tree frog skin are four times higher than the receptor fluid in in vitro Franz cell experiments. Antifungal levels in the skin increased with longer exposures (5 h vs 21 h) to TBF.HCl (*, p = 0.029). Error bars signify SEM of experimental replicates (n = 4 for each treatment). Figure 1. View largeDownload slide Levels of terbinafine (TBF) retained in alpine tree frog skin are four times higher than the receptor fluid in in vitro Franz cell experiments. Antifungal levels in the skin increased with longer exposures (5 h vs 21 h) to TBF.HCl (*, p = 0.029). Error bars signify SEM of experimental replicates (n = 4 for each treatment). Figure 2. View largeDownload slide Haematoxylin and eosin histological stain of alpine tree frog (L. v. alpina) skin A) Ventral skin of a healthy frog, as used in Franz cell experiments (50x magnification). The epidermis (E) is ∼60 μm thick, and dramatic thinning of the dermis (D) occurs between papillae, resulting in a varying overall skin thickness (80-250 μm). B) Uninfected toe skin, used in in vivo pharmacokinetic experiments, with a stratum corneum (SC) of 2 μm (400x magnification) and C) Infected toe skin with an SC of up to 24 μm due to the accumulation of hyperkeratotic layers with fungal sporangia (400x magnification). This Figure is reproduced in color in the online version of Medical Mycology. Figure 2. View largeDownload slide Haematoxylin and eosin histological stain of alpine tree frog (L. v. alpina) skin A) Ventral skin of a healthy frog, as used in Franz cell experiments (50x magnification). The epidermis (E) is ∼60 μm thick, and dramatic thinning of the dermis (D) occurs between papillae, resulting in a varying overall skin thickness (80-250 μm). B) Uninfected toe skin, used in in vivo pharmacokinetic experiments, with a stratum corneum (SC) of 2 μm (400x magnification) and C) Infected toe skin with an SC of up to 24 μm due to the accumulation of hyperkeratotic layers with fungal sporangia (400x magnification). This Figure is reproduced in color in the online version of Medical Mycology. In vivo terbinafine absorption in skin during treatment To confirm that TBF levels also accumulate in frog skin during in vivo exposure, we performed detailed pharmacokinetic studies. The levels of TBF detected in frog skin increased with increasing bath time (Fig. 3). After one 5 min bath, approximately 0.06 μg TBF/mm2 frog skin was detected, corresponding to an uptake rate of 1.2 μg/mm2/min. The concentration increased fivefold after the 30 min baths, indicating linear uptake rates during the first 30 min (0.9 μg/mm2/min). After longer bath times, the overall amount of TBF detected in the skin increased, but the rate of absorption per min slowed. There was no significant difference in the amount of TBF absorbed in the skin when using either TBF.HCl in 1% ethanol or with the diluted Lamisil® Spray formulation. The maximum TBF absorbed in the skin was approximately 0.1 μg/mm2 after 4 h (Fig. 3), which is consistent with the in vitro Franz cell data (Fig. 1). Longer bath times increased the amount of TBF detected in the skin, with 4 times more antifungal absorbed during the 30 min bath compared with the 5 min bath. Repeated treatments also increased TBF absorption in frog skin, with a 4-fold increase in TBF levels after five daily 5 min baths compared to the single 5 min bath, and a 14-fold increase after four daily 30 min baths compared with the single treatment (Fig. 5A). Figure 3. View largeDownload slide In vivo absorption of terbinafine (TBF) increases in alpine tree frog skin with increased exposure time; TBF.HCl (white circles) and Lamisil® Spray (black circles). There was no significant difference in absorption of TBF in skin during treatment using TBF.HCl in 1% ethanol or diluted Lamisil® Spray formulation (both at the concentration of 0.01% w/v TBF.HCl). Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 3. View largeDownload slide In vivo absorption of terbinafine (TBF) increases in alpine tree frog skin with increased exposure time; TBF.HCl (white circles) and Lamisil® Spray (black circles). There was no significant difference in absorption of TBF in skin during treatment using TBF.HCl in 1% ethanol or diluted Lamisil® Spray formulation (both at the concentration of 0.01% w/v TBF.HCl). Error bars signify SEM of biological replicates (n = 3 for each treatment). Pharmacokinetics of terbinafine persistence in skin post treatment Pharmocokinetics experiments were also performed to determine the persistence (apparent Cmin) and half-life of TBF in frog skin. Between 55 and 87% TBF (average 75%) was lost from the skin within 2 h post treatment (Fig. 4), indicating a drug half-life of <2 h in frog skin. There was no difference in persistence between skin exposed to TBF.HCl solution or Lamisil® Spray. When frogs were treated with four daily 30 min baths rather than a single 30 min bath, the drug levels were 30-fold higher after 2 hours (due to both higher initial absorption and longer persistence, 54% vs 78% loss) (Fig. 5B). Figure 4. View largeDownload slide Terbinafine (TBF) concentrations in alpine tree frog skin after different in vivo treatments (exposure times: 5 min, 30 min, 2 h or 4 h). Samples were taken for HPLC analysis immediately after treatment, and 2, 4 and 6 h post-treatment. There is a rapid initial loss of approximately 75% TBF within the first 2 hours, followed by a more gradual loss of the remaining drug. A) Treatment with TBF.HCl for 30 min (black circles), 2 h (white squares) or 4 h (grey triangles); B) Treatment with Lamisil® Spray for 5 min (grey diamonds), 30 min (black circles) or 2 h (white squares). LOD was 0.003 μg/mm2. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 4. View largeDownload slide Terbinafine (TBF) concentrations in alpine tree frog skin after different in vivo treatments (exposure times: 5 min, 30 min, 2 h or 4 h). Samples were taken for HPLC analysis immediately after treatment, and 2, 4 and 6 h post-treatment. There is a rapid initial loss of approximately 75% TBF within the first 2 hours, followed by a more gradual loss of the remaining drug. A) Treatment with TBF.HCl for 30 min (black circles), 2 h (white squares) or 4 h (grey triangles); B) Treatment with Lamisil® Spray for 5 min (grey diamonds), 30 min (black circles) or 2 h (white squares). LOD was 0.003 μg/mm2. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 5. View largeDownload slide Longer and repeated bath times help improve terbinafine (TBF) absorption and persistence in alpine tree frog skin. A) TBF absorption in the skin after single 5 min or 30 min baths, or multiple daily 5 × 5 min baths or 4 × 30 min baths of TBF.HCl. More TBF was absorbed during the repeated 30 min treatment than the other treatments (*, p < 0.05). B) Persistence of TBF in the skin over time, after five daily 5 min (white circles), one 30 min (grey circles), or four daily 30 min (black squares) TBF treatments. TBF persisted longer after the repeated 30 min treatments than the other treatment regimes. Error bars signify SEM of biological replicates (n = 3 for each treatment). Figure 5. View largeDownload slide Longer and repeated bath times help improve terbinafine (TBF) absorption and persistence in alpine tree frog skin. A) TBF absorption in the skin after single 5 min or 30 min baths, or multiple daily 5 × 5 min baths or 4 × 30 min baths of TBF.HCl. More TBF was absorbed during the repeated 30 min treatment than the other treatments (*, p < 0.05). B) Persistence of TBF in the skin over time, after five daily 5 min (white circles), one 30 min (grey circles), or four daily 30 min (black squares) TBF treatments. TBF persisted longer after the repeated 30 min treatments than the other treatment regimes. Error bars signify SEM of biological replicates (n = 3 for each treatment). Terbinafine minimal inhibitory and lethal exposure concentrations The MIC of TBF was determined for both Bd zoospores (motile infective stage) and zoosporangia (parasitic life stage). MIC values were consistent across all of the four Bd strains tested, with a MIC for zoospores of 0.063 μg/ml for both Lamisil® Spray and TBF.HCl solution. The MIC for zoosporangia was at least 30-fold higher at 2 μg/ml for Lamisil® Spray and 6 μg/ml for TBF.HCl (in 1% ethanol). No difference in fungal growth was noted between the commercial spray and the TBF.HCl solution, and in control MIC assays there was no effect of ethanol (≤1%) or propylene glycol (≤1%) alone on fungal growth. Together, these results indicate that the additional components (propylene glycol and cetomacrogol 1000) in the spray had no effect at the dilutions employed, and that TBF primarily contributes to the fungicidal activity of Lamisil® against Bd. Exposure experiments showed that at least 100 μg/ml TBF for 2 h is required to kill Bd zoosporangia in vitro. There were some differences in MLC between the strains tested, with the Waste Point Bd strain being slightly more sensitive to TBF exposure than the Paluma strain: at 100 μg/ml, the Waste Point isolate was killed in 1.5 h and Paluma in 2 h. At 200 μg/ml, Waste Point was killed in 1 h and Paluma in 2 h. Calculation of theoretical Cmin values in vivo To determine whether the absorbed TBF levels that accumulated in frog skin correlated to the minimal lethal concentrations required to kill Bd in vitro, we determined the theoretical Cmin values in the skin (mg/ml) from HPLC values (μg/mm2). True Cmin values could not be calculated as the amount of TBF in skin after 24 h was below the limit of detection (<3 ng/mm2). Two hours after the five daily 5 min treatments, only 2.8 mg/ml TBF was detected in the skin (Table 1), assuming a stratum corneum thickness of 2 μm (Fig. 2B). This in vivo concentration (apparent Cmin at 2 h) is about 30 times the minimum lethal dose required in vitro (100 μg/ml for 2 h). Similar 2 h post treatment concentrations were calculated for the single 30 min and 2 h baths. After four daily 30 min treatments, however, the in vivo concentration was calculated at a maximum of 94.5 mg/ml TBF after 2 hours. This concentration is 950 times higher than the in vitro MLC, and 30 times the amount persisting after five 5 min baths, suggesting that longer repeated TBF baths may be more efficacious in treating frogs with Bd. The overall TBF absorption (μg/mm2) was similar between infected and healthy skin (data not shown), although it is possible that a TBF concentration gradient may form across the thickened stratum corneum of diseased animals (up to 24 μm thick in alpine tree frogs), resulting in a lower in vivo concentration in the lower layers of the epidermis. Table 1. Pharmacokinetic parameters for in vivo terbinafine (TBF) absorption and persistencea in Litoria verreauxii alpina skin. Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 <LOD, below limit of detection at <3 ng/mm2. ND, not determined. aPersistence is defined as the apparent Cmin at a specific time point, post-exposure bMaximum theoretical TBF skin concentration calculated from HPLC values (apparent Cmin at 2 h, ng/mm2), assuming all TBF resides in the stratum corneum (∼2 μm in healthy alpine tree frogs). View Large Table 1. Pharmacokinetic parameters for in vivo terbinafine (TBF) absorption and persistencea in Litoria verreauxii alpina skin. Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 Exposure time (min) Bath Frequency Cmax (ng/mm2) AUC (ng*h/mm2) Apparent Cmin at 2 h (ng/mm2) Theoretical Cmin at 2 hb (mg/ml) Apparent Cmin at 6 h (ng/mm2) TBF.HCl 5 5 26.8 ± 9.3 26.1 5.7 ± 1.5 2.8 <LOD 30 1 28.7 ± 0.8 41.2 6.3 ± 3.3 3.1 <LOD 30 4 407 ± 108.1 843.9 189 ± 19.3 94.5 ND 120 1 44.1 ± 3.6 58.6 7.2 ± 1.2 3.6 <LOD 240 1 65.5 ± 11.8 148.6 29 ± 10.8 14.5 7.4 ± 2.6 Lamisil® 5 1 6.2 ± 0.8 6.2 <LOD … <LOD 30 1 26.7 ± 3.5 41.2 7.3 ± 1.1 3.6 <LOD 120 1 66.8 ± 10.3 102.1 8.3 ± 2.3 4.1 4.2 ± 0.2 <LOD, below limit of detection at <3 ng/mm2. ND, not determined. aPersistence is defined as the apparent Cmin at a specific time point, post-exposure bMaximum theoretical TBF skin concentration calculated from HPLC values (apparent Cmin at 2 h, ng/mm2), assuming all TBF resides in the stratum corneum (∼2 μm in healthy alpine tree frogs). View Large C. signifera terbinafine clinical treatment trial C. signifera were treated with the original antifungal regime of five daily 5 min Lamisil® baths (diluted to 0.01% w/v TBF.HCl),18 and an extended treatment of five daily 30 min Lamisil® baths at 0.01% or 0.02% TBF.HCl. One week after treatment all animals from the 30 min baths (both 0.01% and 0.02% TBF.HCl groups) were negative for Bd (Suppl. Fig. S2). When swabbed again after 4 weeks, 50% of the 0.01% TBF.HCl treatment group and 78% of the 0.02% TBF.HCl treatment group were negative for Bd, compared to 0% for the untreated control group (Table 2). Only 12.5% of frogs from the 5 min 0.01% TBF.HCl baths had cleared Bd infection after 4 weeks. See supplementary information for details on statistical analyses and average infection intensities (Suppl. Fig. S2). Table 2. Proportion of Bd-negative individuals four weeks after daily Lamisil® treatmentsa of Crinia signifera. Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% aLamisil® was diluted in distilled water to the equivalent concentration of 0.01% TBF.HCl or 0.02% TBF.HCl. View Large Table 2. Proportion of Bd-negative individuals four weeks after daily Lamisil® treatmentsa of Crinia signifera. Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% Bath Exposure time (min) Frequency n Bd-negative at week 4 (%) 1% Ethanol (control) 30 5 7 0% 0.01% TBF 5 5 8 12.5% 0.01% TBF 30 5 8 50% 0.02% TBF 30 5 9 78% aLamisil® was diluted in distilled water to the equivalent concentration of 0.01% TBF.HCl or 0.02% TBF.HCl. View Large Discussion Lack of treatment success in Australian frogs using the published TBF regime18 lead us to conduct in-depth pharmacokinetic and treatment trials of this antifungal. The recommended treatment of five daily 5 min 0.01% TBF.HCl baths18 was ineffective at clearing Bd infection in both experimentally infected alpine tree frogs (L. v. alpina) and naturally-infected common eastern froglets (C. signifera). These results are in contrast to the only other study on the efficacy of TBF as a treatment for Bd, where the antifungal drug was effective in six amphibian species.18 One explanation for this discrepancy is that the animals in the previous study may have had lower infection loads, as naturally-infected wild caught less susceptible species were used18,31 and the infection load was not quantified.18 Our study involved both a naturally-infected wild caught species (C. signifera) capable of maintaining high infection loads,26 and a susceptible species (L. v. alpina) experimentally-infected with a large zoospore dose. The 96% temporary decrease in zoospore load in L. v. alpina (Suppl. Fig. S1) 1 week after treatment shows that TBF can greatly reduce Bd burdens, which may be sufficient to treat amphibians with low level infections, as indicated by the previous study.18 An additional source of discrepancy is that in the previous study, the control and test group swabs were combined for PCR analysis post treatment,18 which does not ensure that all untreated control animals retained infection. It is unlikely that the lack of success in clearing infection is due to variations in Bd strain sensitivity to TBF, because our in vitro MIC data were similar for four geographically diverse Bd strains. To optimize TBF treatment protocols and improve efficacy, we undertook in vitro and in vivo pharmacokinetic analysis of TBF absorption and persistence in frog skin. The in vitro Franz cell experiments demonstrated that absorbed TBF accumulates in alpine tree frog skin, and that increased absorption occurred after longer contact times. There was limited systemic absorption of TBF, likely due to the drug's affinity for the keratin within the stratum corneum.32,33 It is therefore critical that the entire body surface of infected frogs contacts the bath solution to ensure direct uptake into all skin areas, rather than relying on redistribution of TBF to skin after systemic absorption. In vivo pharmacokinetic experiments then determined how much of the absorbed TBF accumulated in the skin during the recommended 5 min treatments. TBF absorption increased in the skin with contact time; however, 75% of the antifungal was metabolized within 2 h, followed by a more gradual elimination over several hours. This poor drug persistence may explain why the 5 min treatment did not cure alpine tree frogs of Bd, where zoospore loads recovered to control levels after week 1. A similar, although slower, biphasic elimination of TBF occurs in human skin, with 95% loss within 24 h due to oversaturation of drug in the stratum corneum, and complete elimination within approximately 4 days.34 The more rapid loss of TBF from frog skin is likely due to the thinness of the stratum corneum in amphibians, and frequent sloughing every 2–3 days in L. v. alpina.35 In contrast, a recent paper showed that itraconazole levels are maintained for at least 36 h in the skin of Panamanian golden frogs.23 The longer persistence of itraconazole is likely due to high systemic absorption of that drug, which was not indicated for TBF in our Franz cell experiments. Interestingly, TBF formulation did not appear to have an effect on absorption or persistence. There was no significant difference in TBF absorption when frogs were treated with TBF.HCl in 1% ethanol or with a diluted solution of Lamisil® Spray, which is the form used by private hobbyists. Additionally, there was no difference between the two forms of TBF in the MICs towards zoospores or zoosporangia. Thus, at the dilute concentrations used, the other components in the commercial formulation (propylene glycol, ethanol, and cetomacrogol 1000) have a negligible effect on penetration, persistence, and antifungal activity. The MIC of TBF toward zoosporangia was approximately 2 orders of magnitude higher than that reported previously for zoospores19 and is the more therapeutically relevant level to aim for in skin. The MIC of TBF towards the related fungal pathogen of salamanders, Batrachochytrium salamandrivorans (Bsal), is only 0.2 μg/ml, compared with 6 μg/ml for Bd,36 highlighting the difference in fungal species susceptibility. This lower MIC suggests that TBF treatment may be a more effective strategy against Bsal infections in salamanders than for Bd. TBF exposure assays enabled a MLC value to be determined at 100 μg/ml for 2 h. As our pharmacokinetic data indicated that TBF is unlikely to persist at high concentrations for extended periods in frog skin, these fungicidal levels (rather than MIC values) may be indicative of the concentration required to clear Bd in vivo. Two hours after the recommended five daily 5 min baths,18 TBF levels in the skin were only 30 times the in vitro MLC. Therefore, the inability of the 5 min treatments to consistently cure frogs of Bd may be explained by factors that reduce the in vivo TBF bioavailability (e.g. binding to keratin,32 lipids and membranes), and by variations in Bd localization, skin shedding rate,35 and stratum corneum thickness. Consequently, the accumulated drug concentration absorbed after five 5 min treatments may not be 100% effective as a fungicide, as suggested by the recovery of zoospore load after week 1 of the L. v. alpina treatment trial. The rapid decline in TBF concentrations within the skin indicates that more frequent treatments would be required to sustain effective drug levels in vivo. In fact, our pharmacokinetic experiments indicate that repeated baths improve drug absorption and persistence within frog skin. TBF levels were 14 times higher after four repeated 30 min baths, compared with the single 30 min bath. Two hours after the four daily 30 min baths, TBF levels persisting in the skin were 950-fold higher than the in vitro MLC (compared to only 30 times the MLC persisting after five 5 min baths). To test the efficacy of a longer and repeated treatment regime, we treated C. signifera with five daily 30 min Lamisil® baths at 0.01% or 0.02% TBF.HCl. After 4 weeks, 50–78% of individuals within the 30 min treatment groups completely cleared infection, compared to only 12.5% clearance in animals treated with five 5 min exposures. Therefore, longer baths increase the efficacy TBF against Bd; however, the antifungal did not clear all infected frogs of Bd. It is likely that the poor drug persistence of TBF in frog skin limits its ability to completely cure frogs of high Bd loads, allowing fungal levels to recover after treatment ends. Therefore, our results show that the current TBF regimes are ineffective in completely treating frogs of Bd. Clearance rates improved from 50% to 78% with an increased TBF concentration of 0.02%. Ostensibly, an even higher TBF dose or more repeated exposures may prove to be effectual against Bd by maintaining therapeutic drug levels in the skin. The low apparent Cmin values at 6 h indicate that TBF treatments should be repeated more frequently than every 24 h. However, this optimized regime would require further clinical trials to test efficacy, feasibility, and safety. The results from this study have implications for the protection of wildlife, as optimization of effective Bd treatment protocols requires empirical pharmacological evidence.22 Many endangered frog species rely on captive assurance populations and reintroductions for survival,37 and thus, maintaining disease-free colonies is essential in these zoological institutions. In addition, biosecurity and effective antifungal protocols are critical to preventing the spread of Bd to naive geographical areas, such as Hong Kong38 and Papua New Guinea.39 Finally, antifungal protocols are important tools in front line responses to epidemics, where in situ treatments have been used to cure frogs in the wild.40 Therefore, our nonlethal method of using toe clips to quantify antifungal drug levels could be useful in optimizing other chytridiomycosis treatments and improve the efficacy for endangered frogs; as fewer animals are required than for whole skin analysis,23 and drug concentrations in each individual can be quantified over time. This study highlights the importance of both pharmacokinetic data and controlled veterinary clinical trials in determining the most efficient and successful treatment for disease in amphibians. Pharmacokinetic studies, while common for mammalian treatments, are lacking for amphibians, and as the two skin structures differ,8 treatments need to be tested using an amphibian model. This in-depth series of experiments explored the absorption and persistence of TBF in amphibian skin, and demonstrated that the current recommended dose and treatment time18 is not effective for all amphibian species. While our protocols increased the efficacy of TBF in treating Bd infections, the low persistence of this antifungal in frog skin appears to limit its ability to clear high pathogen loads. Due to TBF’s accessibility without prescription, low up-front costs, and ease of use, it is likely that private hobbyists may continue its use in treating amphibians. The results from our study, however, show that the current TBF protocols are not 100% effective against Bd in all amphibian species. More frequent treatments have the potential to improve clearance rates, but protocols need to be optimized before TBF can be used with confidence by zoological or academic institutions to universally treat frogs with chytridiomycosis. Acknowledgment R. Speare, who initiated our collaboration, tragically died in a car accident in 2016 and his vision in bringing together disparate expertise to better solve real world problems will be greatly missed. We thank P. Harlow and Taronga Zoo for raising the L. v. alpina, B. Scheele for providing the C. signifera, P. Thomas-Hall and V. Llewellyn for technical assistance and helpful discussions, and J. Longcore for providing the Arizona Bd isolate. Thank you to M. Souza and S. Cox for advice on terbinafine inhibitory assays and quantitative analysis. We thank D. Tegtmeier, C. De Jong, J. Hawkes, K. Fossen, S. Percival, M. McWilliams, L. Bertola, M. Stewart, N. Harney, and T. Knavel for data collection and husbandry assistance. Funding This work was supported by a Queensland Government Accelerate Fellowship [14-218 to A.A.R.] with additional funds from the Taronga Conservation Science Initiative and the Queensland Department of Environment and Heritage Protection; the Australian Research Council [FT100100375, LP110200240, DP120100811 to LFS and LB] with additional funds from industry partners Taronga Zoo and New South Wales Office of Environment and Heritage; the Advance Queensland Women's Academic Fund; and a Collaboration Across Boundaries Grant provided by James Cook University. Supplementary material Supplementary data are available at MMYCOL online. Declaration of Interest The authors report no conflicts of interest. The authors alone are responsible for the content and the writing of this paper. References 1. Berger L , Speare R , Daszak P et al. Chytridiomycosis causes amphibian mortality associated with population declines in the rainforests of Australia and Central America . Proc Natl Acad Sci USA . 1998 ; 95 : 9031 – 9036 . Google Scholar CrossRef Search ADS PubMed 2. Skerratt LF , Berger L , Speare R et al. Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs . EcoHealth . 2007 ; 4 : 125 – 134 . Google Scholar CrossRef Search ADS 3. Martel A , Blooi M , Adriaensen C et al. Recent introduction of a chytrid fungus endangers Western Palearctic salamanders . Science . 2014 ; 346 : 630 – 631 . Google Scholar CrossRef Search ADS PubMed 4. Pennisi E . Life and death play out on the skins of frogs . Science . 2009 ; 326 : 507 – 508 . Google Scholar CrossRef Search ADS PubMed 5. Berger L , Hyatt AD , Speare R , Longcore JE . Life cycle stages of the amphibian chytrid Batrachochytrium dendrobatidis . Dis Aquat Org . 2005 ; 68 : 51 – 63 . Google Scholar CrossRef Search ADS PubMed 6. Greenspan SE , Longcore JE , Calhoun AJK . Host invasion by Batrachochytrium dendrobatidis: Fungal and epidermal ultrastructure in model anurans . Dis Aquat Org . 2012 ; 100 : 201 – 210 . Google Scholar CrossRef Search ADS PubMed 7. Voyles J , Young S , Berger L et al. Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines . Science . 2009 ; 326 : 582 – 585 . Google Scholar CrossRef Search ADS PubMed 8. Llewelyn V , Berger L , Glass B . Percutaneous absorption of chemicals: Developing an understanding for the treatment of disease in frogs . J Vet Pharmacol Ther . 2015 ; 39 : 109 – 121 . Google Scholar CrossRef Search ADS PubMed 9. Baitchman EJ , Pessier AP . Pathogenesis, diagnosis, and treatment of amphibian chytridiomycosis . Vet Clin North Am Exot Anim Pract . 2013 ; 16 : 669 – 685 . Google Scholar CrossRef Search ADS PubMed 10. Brannelly LA , Richards-Zawacki CL , Pessier AP . Clinical trials with itraconazole as a treatment for chytrid fungal infections in amphibians . Dis Aquat Org . 2012 ; 101 : 95 – 104 . Google Scholar CrossRef Search ADS PubMed 11. Woodhams DC , Geiger CC , Reinert LK et al. Treatment of amphibians infected with chytrid fungus: Learning from failed trials with antimicrobial peptides, bacteria, and heat therapy . Dis Aquat Org . 2012 ; 98 : 11 – 25 . Google Scholar CrossRef Search ADS PubMed 12. Brannelly LA . Reduced intraconazole concentration and durations are successful in treating Batrachochytrium dendrobatidis infection in amphibians . J Vis Exp . 2014 ; 85 : e51166 . 13. Martel A , van Rooij P , Vercauteren G et al. Developing a safe antifungal treatment protocol to eliminate Batrachochytrium dendrobatidis from amphibians . Med Mycol . 2010 ; 49 : 143 – 149 . Google Scholar CrossRef Search ADS PubMed 14. Holden WM , Ebert AR , Canning PF , Rollins-Smith LA . Evaluation of amphotericin B and chloramphenicol as alternative drugs for treatment of chytridiomycosis and their impacts on innate skin defenses . Appl Environ Microbiol . 2014 ; 80 : 4034 – 4041 . Google Scholar CrossRef Search ADS PubMed 15. Brannelly LA , Skerratt LF , Berger L . Treatment trial of clinically ill corroboree frogs with chytridiomycosis with two triazole antifungals and electrolyte therapy . Vet Res Commun . 2015 ; 39 : 179 – 187 . Google Scholar CrossRef Search ADS PubMed 16. Berger L , Speare R , Hines HB et al. Effect of season and temperature on mortality in amphibians due to chytridiomycosis . Aust Vet J . 2004 ; 82 : 434 – 439 . Google Scholar CrossRef Search ADS PubMed 17. Chatfield MWH , Richards-Zawacki CL . Elevated temperature as a treatment for Batrachochytrium dendrobatidis infection in captive frogs . Dis Aquat Org . 2011 ; 94 : 235 – 238 . Google Scholar CrossRef Search ADS PubMed 18. Bowerman J , Rombough C , Weinstock SR , Padgett-Flohr GE . Terbinafine hydrochloride in ethanol effectively clears Batrachochytrium dendrobatidis in amphibians . J Herpetol Med Surgery . 2010 ; 20 : 24 – 28 . Google Scholar CrossRef Search ADS 19. Woodward A , Berger L , Skerratt LF . In vitro sensitivity of the amphibian pathogen Batrachochytrium dendrobatidis to antifungal therapeutics . Res Vet Sci . 2014 ; 97 : 365 – 367 . Google Scholar CrossRef Search ADS 20. Gold KK , Reed PD , Bemis DA et al. Efficacy of common disinfectants and terbinafine in inactivating the growth of Batrachochytrium dendrobatidis in culture . Dis Aquat Org . 2013 ; 107 : 77 – 81 . Google Scholar CrossRef Search ADS PubMed 21. Hill S , Thomas R , Smith SG , Finlay AY . An investigation of the pharmacokinetics of topical terbinafine (Lamisil) 1% cream . Br J Dermatol . 1992 ; 127 : 396 – 400 . Google Scholar CrossRef Search ADS PubMed 22. Berger L , Speare R , Pessier AP , Voyles J , Skerratt LF . Treatment of chytridiomycosis requires urgent clinical trials . Dis Aquat Org . 2010 ; 92 : 165 – 174 . Google Scholar CrossRef Search ADS PubMed 23. Rifkin A , Visser M , Barrett K , Boothe D , Bronson, E The pharmacokinetics of topical itraconazole in Panamanian golden frogs (Atelopus zeteki) . J Zoo Wildlife Med . 2017 ; 48 : 344 – 351 . Google Scholar CrossRef Search ADS 24. Consumer Information Lamisil . 2016 ; Novartis Pharmaceuticals Inc . URL www.ask.novartispharma.ca/download.htm?res=lamisil_patient_e.pdf&resTitleId=137 . 25. Brannelly LA , Hunter D , Lenger D et al. Dynamics of chytridiomycosis during the breeding season in an Australian Alpine amphibian . PLoS ONE . 2015 ; 10 : e0143629 . Google Scholar CrossRef Search ADS PubMed 26. Scheele BC , Hunter DA , Brannelly LA , Skerratt LF , Driscoll DA . Reservoir-host amplification of disease impact in an endangered amphibian . Conserv. Biol. 2016 ; DOI: https://doi.org/10.1111/cobi.12830 . 27. Boyle DG , Boyle DB , Olsen V , Morgan JAT , Hyatt AD . Rapid quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay . Dis Aquat Org . 2004 ; 60 : 141 – 148 . Google Scholar CrossRef Search ADS PubMed 28. Kriger KM , Hero J-M , Ashton KJ . Cost efficiency in the detection of chytridiomycosis using PCR assay . Dis Aquat Org . 2006 ; 71 : 149 – 154 . Google Scholar CrossRef Search ADS PubMed 29. Webb R , Philips A , Speare R , Connolly J , Berger L . Controlling wildlife fungal disease spread: in vitro efficacy of disinfectants against Batrachochytrium dendrobatidis and Mucor amphibiorum . Dis Aquat Org . 2012 ; 99 : 119 – 125 . Google Scholar CrossRef Search ADS PubMed 30. R Core Team . R: A language and environment for statistical computing . R Foundation for Statistical Computing . 2013 ; Vienna, Austria . URL http://www.R-project.org/ . 31. Davidson EW , Parris M , Collins JP et al. Pathogenicity and transmission of chytridioycosis in tiger salamanders (Ambystoma tigrinum) . Copeia . 2003 ; 3 : 601 – 607 . Google Scholar CrossRef Search ADS 32. Tatsumi Y , Yokoo M , Senda H , Kakehi K . Therapeutic efficacy of topically applied KP-103 against experimental tinea unguium in guinea pigs in comparison with amorolfine and terbinafine . Antimicrob Agents Chemother . 2002 ; 46 : 3797 – 3801 . Google Scholar CrossRef Search ADS PubMed 33. Schäfer-Korting M , Schoellmann C , Korting HC . Fungicidal activity plus reservoir effect allow short treatment courses with terbinafine in tinea pedis . Skin Pharmacol Phys . 2008 ; 21 : 203 – 210 . Google Scholar CrossRef Search ADS 34. Faergermann J , Zehender H , Millerioux L . Levels of terbinafine in plasma, stratum corneum, dermis–epidermis (without stratum corneum), sebum, hair and nails during and after 250 mg terbinafine orally once daily for 7 and 14 days . Clin Exp Dermatol . 1994 ; 19 : 121 – 126 . Google Scholar CrossRef Search ADS PubMed 35. Ohmer MEB . Interactions between amphibian skin sloughing and a cutaneous fungal disease: Infection progression, immune defence, and phylogenetic patterns . School of Biological Sciences . 2016 ; PhD diss . 143 . 36. Blooi M , Pasmans F , Rouffaer L et al. Successful treatment of Batrachochytrium salamandrivorans infections in salamanders requires synergy between voriconazole, polymyxin E and temperature . Sci Rep . 2015 ; 5 : 11788 . Google Scholar CrossRef Search ADS PubMed 37. Griffiths RA , Pavajeau L . Captive breeding, reintroduction, and the conservation of amphibians . Conserv Biol . 2008 ; 22 : 852 – 861 . Google Scholar CrossRef Search ADS PubMed 38. Rowley J , Chan S , Tang W et al. Survey for the amphibian chytrid Batrachochytrium dendrobatidis in Hong Kong in native amphibians and in the international amphibian trade . Dis Aquat Org . 2007 ; 78 : 87 – 95 . Google Scholar CrossRef Search ADS PubMed 39. Bower D , Lips K , Schwarzkopf L , Georges A , Clulow S . Amphibians on the brink . Science . 2017 ; 357 : 454 – 455 . Google Scholar CrossRef Search ADS PubMed 40. Hardy B , Pope K , Piovia-Scott J , Brown R , Foley J . Itraconazole treatment reduces Batrachochytrium dendrobatidis prevalence and increases overwinter field survival in juvenile Cascades frogs . Dis Aquat Org . 2015 ; 112 : 243 – 250 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Medical MycologyOxford University Press

Published: Mar 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off