This study was performed with the approval of the University of Illinois Institutional Animal Care and Use Committee (No. 22209). Six adult bearded dragons (Pogona vitticeps; four males and two females) maintained as a research colony were enrolled. The bearded dragons, maintained at the same institution for the prior 4 years, were 5 years of age and weighed 0.29–0.40 kg (0.63–0.88 lbs). They were deemed healthy based on known history and evaluation (physical examination, complete blood count, and fecal analysis) performed 14 days prior to initial N. guarroi exposure. Each lizard had a negative N. guarroi qPCR from a cutaneous swab 14 days prior to the study start.

Lizards were individually housed on newspaper in 91.45 × 71.12 × 45.72-cm enclosures located in the same room for the entirety of the study. Enclosures were spot cleaned daily and deep cleaned weekly with warm soapy water. Room ambient temperature was maintained between 23.3°C and 24.4°C (74.0°F and 76.0°F). Each enclosure contained a heat lamp and UVA/UVB-emitting bulb on for 12 h/day, allowing a basking temperature of 35.0°C–37.8°C (95.0°F–100.0°F). Light from the heat lamp and UVA/UVB-emitting bulbs (ZooMed ReptiSun 10.0 High Output UVB Fluorescent bulb) overlapped at the basking site.

An isolate of N. guarroi, obtained from a bearded dragon with naturally occurring dermatomycosis, was utilized in this study (Lab ID: CC1; GenBank No. MT506461). The bearded dragon was euthanized after the development of deep cutaneous lesions histopathologically consistent with cutaneous mycosis. Postmortem fungal culture of cutaneous swabs produced a pure isolate of white fluffy colonies. The initial isolate identity was confirmed molecularly based on prior published methods for other N. guarroi isolates (McEntire et al., 2022). PCR targets for identity confirmation included the D1/D2 domains of the 28S ribosomal DNA (D1/D2 rDNA) (Cano et al., 2004), the 18S–28S ribosomal internal transcribed spacer region including the 5.8S rDNA (ITS) (primers ITS4 and ITS5) (White et al., 1990), the actin gene (Voigt and Wöstemeyer, 2000), and the β-tubulin gene (primers BT2-F and BT2-R) (Gilgado et al., 2005). PCR products were electrophoresed on a 1% agarose gel, then treated with ExoSAP-IT (USB Corporation, Cleveland, OH, USA) and bidirectionally sequenced (ACGT, Wheeling, IL, USA). Chromatograms were visually inspected and trimmed of primer sequences and low-quality base pair calls. The resulting sequences were compared to those in the NCBI GenBank database using BLASTN. The 601-bp ITS product was 100% identical to N. guarroi (OW988362.1, MW617063.1). Similarly, a 573-bp D1/D2 product was 100% identical to N. guarroi (MH874904.1, HF547864.1). The 872-bp actin product was 100% identical to one N. guarroi sequence (HF547897.1). The 429-bp β-tubulin product was 99.75% identical to N. guarroi (HF547900.1, HF547896.1). Multilocus sequencing confirmed the fungal isolate as N. guarroi.

The isolate was propagated on potato dextrose agar (Remel, Thermo Scientific, Lenexa, KS, USA) and incubated under ambient lighting at 23°C (73.4°F) for at least 10 days. A swab of the isolate was suspended in sterilized distilled water. The suspension was vortexed for 2 min and then filtered through glass wool under vacuum. The number of conidia in suspension was estimated by hemocytometer count.

For the initial exposure event, a conidial concentration of 7.0919 × 10⁴ conidia/μL was used, aligning with another N. guarroi experimental infection study (Gentry et al., 2021). Conidial suspension (10 μL) was applied to five unabraded sites along the dorsal midline and allowed to dry completely. After failure of clinical lesion formation within 31 days postinitial exposure (DPIE), as reported by Gentry et al. (2021), repeat exposures were performed on days 36, 64, and 99 DPIE if each lizard had not developed clinical cutaneous lesions. During additional exposure events, the highest concentration that could be harvested from the fungal mats grown in culture was utilized (1.12 × 10⁵ conidia/μL), and 20 μL per site of this conidial suspension was applied to the same five dorsal midline sites.

Physical examination, cutaneous sample collection, lesion photography, and screening with Wood’s lamp fluorescence were performed weekly. On dates corresponding with exposure events, data collection procedures were performed prior to exposure. Bearded dragons were inspected for development of cutaneous lesions consistent with those described in published cases of N. guarroi infection (Gentry et al., 2021) and photographed if present. Given the absence of a case definition for nannizziomycosis, any dermatologic abnormality was documented and imaged. Suspicious lesions were intermittently sampled by swab or impression smear and routinely stained with Wright–Giemsa (HemaTek 3000, Siemens, Tarrytown, NY, USA) for cytologic evaluation (Sharkey et al., 2014). Microscopic cytologic evaluation of these samples was performed by a board-certified veterinary clinical pathologist to assess for the presence of inflammation and/or fungal elements.

Cutaneous swabs were collected for NGS, fungal culture, and qPCR, in that order. During the preclinical phase, when no lesions were present, whole-body samples were collected by vigorously rubbing swabs 20–30 times each along the dorsal and ventral surfaces of the lizard. After the development of suggestive cutaneous lesions, samples were also obtained from the suspect lesions using the same swabs after collecting the whole-body sample.

Samples collected for NGS used a sterile DNA-free swab (HydraFlock^®, Puritan^® Cat. No. 25-3406-H, Guilford, ME, USA) and a sterile sample collection tube prefilled with a DNA/RNA preservative (DNA/RNA Shield™ Zymo Research Corp.; Cat. No. R1108, Irvine, CA, USA). Collection was performed with sterile gloves to limit inadvertent sampling of human microbiota. Samples were shipped in batches to a commercial laboratory (MiDog LLC Science, Irvine, CA, USA), which amplifies fungal DNA through ITS-2 targets as previously described in the literature (Tang et al., 2020). NGS results were reported as positive or negative for the detection of Nannizziopsis spp. DNA.

Samples for fungal culture were collected using sterile cotton-tipped applicators (CTAs; Fisherbrand™ Plastic Handled Cotton Swabs and Applicators, Fisher Scientific, Waltham, MA, USA) that were stored in sterile bags without media. Within 2 h of collection, swabs were directly streaked onto a modified agar composed of potato dextrose agar (Remel, Thermo Scientific, Lenexa, KS, USA), penicillin–streptomycin (60,000/60 U/L; Fisher Scientific), gentamicin sulfate (50 mg/L; Sigma-Aldrich, St. Louis, MO, USA), and cycloheximide (400 mg/L; Sigma-Aldrich, St Louis, MO, USA) to reduce bacterial and saprophytic fungal growth, then incubated under ambient lighting at 23°C (73.4°F) for 10–14 days. Fungal culture results were recorded as positive if growth was present that was morphologically consistent with the original pure N. guarroi isolate (white fluffy colonies with microscopic hyphal structures), or negative if no growth consistent with N. guarroi was observed.

Three cutaneous swabs for qPCR testing were collected with sterile CTAs and stored at − 80°C (− 112°F) until collective batched analysis, when an arbitrarily chosen single swab from each time point was analyzed. A commercial nucleic acid purification kit (QIAamp DNA Mini-Kit, Qiagen Inc., Valencia, CA, USA) was used for DNA extraction; the manufacturer’s recommendations were followed, with the addition of a 1-h incubation at 37°C (98.6°F) with 300 U of lyticase (Sigma-Aldrich, St. Louis, MO, USA) prior to the lysis step. Extracted DNA samples were assessed for concentration (measured in nanograms per microliter) and quality (using the ratio of absorbance at 260 to 280 nm) using spectrophotometry (Nanodrop1000, ThermoFisher Scientific, Wilmington, DE, USA), then stored at – 20°C prior to analysis.

A TaqMan qPCR assay (NGEF1_234), targeting a 57-base pair segment of the N. guarroi elongation factor 1 gene (Keller et al., 2022), was performed on extracted DNA. Real-time qPCR was performed using a real-time PCR thermocycler (QuantStudio 3 real-time PCR system; Applied Biosystems, Foster City, CA, USA), and data were analyzed using QuantStudio Design and Analysis Software v.1.5.2 (Applied Biosystems). Each qPCR reaction contained 12.5 μL of 20× TaqMan Platinum PCR Supermix-UDG with ROX (TaqMan Platinum PCR Supermix-UDG with ROX, Invitrogen, Carlsbad, CA, USA), 1.25 μL of 20× TaqMan primer-probe (final concentrations in each PCR reaction: 900 nM forward primer, 900 nM reverse primer, and 250 nM probe), 2.5 μL of plasmid dilution (or nontemplate control: sterile deionized water), and water to a final volume of 25 μL. Cycling parameters were as follows: one cycle at 50°C for 2 min followed by 95°C for 10 min, 40 cycles at 95°C for 15 s and 60°C for 60 s, and a final cycle of 72°C for 10 min. Samples, standard curves, and nontemplate controls were assayed using three technical replicates unless otherwise specified. Positive controls (plasmid-derived standard curves from 1 × 10⁷ to 1 × 10⁰ target copies per reaction) and nontemplate controls were included in all qPCR runs. Mean fungal quantities (measured as copies per ng DNA) were standardized to the total quantity of DNA in each sample by dividing the mean copies/μL of each sample by the DNA concentration. Samples were considered positive if at least two of the three technical replicates amplified with a CT value within 0.3 units of the other positive replicate(s) (Laing et al., 2019). Samples with CT values greater than the CT value for the 1 × 10⁰ standard dilution were considered negative.

Evaluation of long-wave UV fluorescence using a Wood’s lamp (UV502 UV Light With Magnifier, Burton Medical, Addison, IL, USA) was performed weekly. The bearded dragons were manually restrained for dermal illumination in a dark room, and photographs of their dorsal and ventral surfaces were obtained. UV fluorescence of cutaneous lesions was recorded as positive or negative by a single investigator.

Four days after clinical disease manifested in the final lizard, animals were euthanized on 152 DPIE with pentobarbital (500 mg/kg, transmucosal; EUTHASOL^®, Virbac AH Inc., Carros, France) (Wong et al., 2024). The next day, up to three cutaneous lesions were arbitrarily identified and sampled from each lizard, representing the most severe lesions that were previously cytologically confirmed and/or suspected to be dermatomycosis lesions. Cutaneous samples for fungal culture, qPCR, and cytology were collected from each lesion, and UV fluorescence was assessed. The ventral coelom of each lizard was entered, and coelomic organs were evaluated for pathologic changes before the entire carcass was immersed in 10% neutral-buffered formalin for fixation and submitted for evaluation by a board-certified veterinary anatomic pathologist.

A time zero was assigned to each bearded dragon on the date the first lesion grossly consistent with nannizziomycosis (Gentry et al., 2021) was noted, and the presence of fungal elements was confirmed cytologically. After cytologic confirmation, each lesion was tracked retroactively using weekly images previously taken to assess for prior presence, and an earlier date of lesion development was assigned if appropriate. The date of the first positive diagnostic result was then assigned for fungal culture, qPCR, NGS, and UV fluorescence in relation to time zero of the date of lesion development.

Although samples were collected weekly, the unexpectedly prolonged time to clinical infection compared to the literature (Gentry et al., 2021) necessitated a reevaluation of data analysis. Thus, samples representing monthly time points throughout the study were submitted for NGS and qPCR to better represent the longitudinal nature of the dataset. Cutaneous samples for qPCR were analyzed from preexposure and at 8, 36, 64, 93, 120, and 148 DPIE, while samples for NGS were run from preexposure and at 8, 36, 64, 93, and 120 DPIE. Fungal cultures and UV fluorescence were performed weekly. Antemortem and postmortem diagnostic testing results are presented descriptively. All descriptive analyses were performed using commercial software (Prism, GraphPad version 10.0.0 [RRID: SCR_002798] or Excel, Microsoft for Mac version 16.86 [RRID: SCR_016137]).

All six bearded dragons required re-inoculation of N. guarroi conidial suspension on 36 and 64 DPIE, and four bearded dragons required re-inoculation on 99 DPIE. Cutaneous lesions consistent with N. guarroi infection developed in all lizards, with the first lesion appearing at a median of 99 DPIE (range: 78–120 DPIE). Three of six lizards (Animal IDs: C, E, F) had cytologic confirmation of the presence of fungal elements (hyphae and/or conidia) in a cutaneous lesion at the time of lesion detection, while three (Animal IDs: B, G, J) had cytologic evidence of fungal elements lagging by 1, 2, or 3 weeks after the lesion became clinically apparent. In this latter cohort, the date of disease was assigned as the first date that a lesion was grossly apparent retrospectively on photographs. Cytologically, Wright–Giemsa stain identified fungal elements that appeared as either septate, branching hyphae or solitary conidia, accompanied by variable inflammatory cells and necrotic debris ( Figure 1 ), consistent with nannizziomycosis (Le Donne et al., 2016).

No bearded dragon developed lesions at the site of N. guarroi exposure (dorsal midline). The first cutaneous lesions in the four male bearded dragons (Animal IDs: B, C, E, F) were one to three swollen femoral pores per lizard ( Figure 2A ). Throughout the study, affected femoral pores developed soft tissue swelling, erythema, and eventually scabs. Infected femoral pores readily exfoliated as thick, waxy plugs and frequently resulted in open wounds ( Figure 3A ). Most male bearded dragons (three of four, 75%; Animal IDs: B, C, F) also developed yellow, diffuse discoloration on the extremities. Initial lesions in the two female lizards (Animal IDs: G, J) were one to two discolored (yellow or black) lesions lateralized on the dorsal or ventral body surfaces ( Figures 3B, C ). Neither female bearded dragon developed lesions of the femoral pores.

All bearded dragons maintained regular urofecal output and consistency throughout the study. Four lizards (66.67%; Animal IDs: B, F, G, J) had variable appetites that resulted in more than 5% weight loss over two consecutive weeks. Reduced appetite and weight loss were transient and responded to tong feeding and increased diet rations. Body weights at the end of the study ranged from 0.31 to 0.39 kg (0.67–0.85 lbs.). Based on coelomic palpation, one female (Animal ID: G) underwent folliculogenesis after the third exposure event (64 DPIE), then resorbed the follicles after the fourth exposure event (99 DPIE) but before developing cutaneous lesions. During the last month of the study, daily analgesia with meloxicam (Metacam^®, Boehringer Ingelheim, Ingelheim am Rhein, Germany; dose: 0.2–0.4 mg/kg PO q24h) was initiated in one male lizard (Animal ID: B) due to perceived discomfort associated with the infected femoral pores. Perceived pain behaviors (abnormal body posture at rest, lethargy) in this lizard ceased once meloxicam was regularly instituted.

Prior to inoculation, the fungal culture, NGS, and qPCR for each of the bearded dragons was negative. Antemortem, all lizards tested positive for N. guarroi via fungal culture, NGS, and qPCR ( Supplementary Table S1 ). Considering the imbalanced design of antemortem diagnostic testing, with some diagnostics being evaluated weekly (fungal culture and UV fluorescence) and others monthly (qPCR and NGS), paired with intermittent contamination of agar plates that rendered some findings inconclusive, statistical comparisons were not performed. No distinct UV fluorescence was observed at cutaneous lesions or at sites that eventually developed cutaneous lesions via Wood’s lamp examination at any time point. Fluorescence was noted in uninfected male femoral pores, urates, and body sites undergoing ecdysis ( Supplementary Figure S1 ).

Fungal culture was positive for N. guarroi growth before the development of cutaneous lesions in all bearded dragons ( Supplementary Table S1 ). The first positive culture occurred a median of 28 days prior to lesion development, ranging from 50 to 7 days prior. Of the 126 fungal cultures, representing weekly cultures from six lizards, 11 (8.7%) exhibited plate contamination that prevented interpretation. When these contaminated plates are excluded from the dataset, all subsequent weekly cultures were positive for three lizards following their initial positive culture results. One lizard (Animal ID: G) had two intermittent negative cultures following its initial positive, and another (Animal ID: E) had a single negative culture following its initial positive. The remaining individual (Animal ID: C), which had the earliest initial positive culture result (50 days prior to lesion development), exhibited intermittently positive fungal cultures, and for the last five weekly cultures exhibited negative fungal culture growth.

The first positive NGS result occurred before lesion development in five lizards, while it occurred after lesion development in one lizard (Animal ID: E). The first NGS-positive result occurred at a median of 17.5 days prior to lesion development, with a range of 112 days prior to 15 days after lesion development. Five lizards remained consistently NGS positive after their first positive result (100% positive after initial positive). One lizard (Animal ID: G) had a single positive result at 8 DPIE, 112 days prior to developing lesions, then had two negative NGS before testing positive once more, 27 days prior to lesion development. Subsequent NGS testing in this lizard was positive.

The first qPCR-positive result occurred before lesion development in four lizards, on the same date as lesion development in one lizard (Animal ID: C), and after lesion development in another lizard (Animal ID: E). The first qPCR-positive result occurred at a median of 6 days prior to lesion development (range: 27 days preceding to 15 days after lesions developed). Four lizards remained consistently qPCR positive after their first positive. One lizard (Animal ID: F) had a single intermittent negative result following the initial positive, and the final lizard (Animal ID: C) tested qPCR positive only once, then was negative after that time. Of the 15 positive antemortem qPCR samples, copies per reaction ranged from 5.4 to 355.2 with a median of 71.3 and copies per nanogram of DNA ranged from 0.04 to 15 with a median of 0.8 ( Supplementary Figure S2 ).

Five of six bearded dragons (Animal IDs: B, C, E, F, J) had lesions on their ventral body surfaces, ranging from one to four cutaneous lesions per lizard. Three of the six bearded dragons (Animal IDs: C, F, G) had lesions on their dorsal body surfaces, ranging from one to three lesions per lizard. All male bearded dragons (Animal IDs: B, C, E, F) had lesions associated with their femoral pores, while no female lizard exhibited a femoral pore lesion. While many of the cutaneous lesions exhibited yellow discoloration of the lesion and the surrounding skin, several lesions were darkly colored and showed no yellow discoloration. Histopathology and additional postmortem diagnostic tests (cytology, culture, qPCR, Wood’s lamp) were performed on three arbitrarily chosen cutaneous lesions in four individuals, and on two cutaneous lesions from two individuals with only two cutaneous lesions, for a total of 16 cutaneous lesions ( Table 1 ).

Lesion cytology was performed on 15/16 (93.75%) lesions. Cytology was not performed on one lesion due to its extremely small size, lack of easily collected exudate, and a desire to preserve tissue architecture for histologic analysis. Cytology agreed with histopathology regarding the presence or absence of fungal elements in eight of 15 (53.3%) of the lesions. Of the seven lesions with cytologically detectable fungi, five had concurrent nonseptic inflammation and one had septic bacterial inflammation. Of the eight cytologically negative lesions, two exhibited nonseptic inflammation, and one showed septic bacterial inflammation.

Fungal culture and qPCR were assessed in all 16 lesions and agreed with histopathology in 15 (93.8%) lesions. Fungal culture and qPCR agreed with each other 100% of the time. qPCR values, expressed as copies per nanogram of DNA, ranged from 0 to 772.4, with a median of 51.9. No lesions exhibited UV fluorescence under Wood’s lamp illumination.

Six of 15 (37.50%) for which all diagnostics were performed showed complete positive agreement among assays in the detection of N. guarroi ( Table 1 ). In the one lesion for which cytology was not performed, fungal culture, qPCR, and histopathology were in positive agreement. Among the 15 histopathologically positive lesions, cytology was the diagnostic test most commonly in disagreement, with eight of 14 (57.14%) cytology results testing negative. This was followed by fungal culture and qPCR, which were both negative for a single lesion (Animal ID: C, lesion 2). This cutaneous lesion, negative on all diagnostic assays outside of histopathology, corresponded to a flat, firm area just caudal to a femoral pore that oozed orange, malodorous, thick, opaque material when sampled. The cutaneous lesion that was histopathologically negative originated from a yellow-discolored region of the dorsal right antebrachium and demonstrated mild regional hyperkeratosis without infectious agents ( Figure 2B ; Animal ID: F, lesion 2). Clinically, this lesion lacked scale and was historically the site of a healed bite wound. The site was sampled due to the yellow discoloration that had developed during the infection trial.

Histopathology of cutaneous lesions, stained with hematoxylin and eosin, identified chronic proliferative or erosive dermatitis. All lesions but one (15/16, 93.75%; Figure 2B ) were confirmed to have intracorneal fungal hyphae and/or conidia, as confirmed by Grocott–Gomori Methenamine Silver (GMS) stain ( Figures 4A, B ) (Grocott, 1955). Seven of the 15 histopathologically positive lesions were of femoral pores, five of which had deeply associated subcutaneous or intramuscular granulomas in addition to fungal structures and dermatitis ( Figures 4C–F ). Subcutaneous or intramuscular granulomas were present in three additional cutaneous lesions from other body sites (ventral hind limb, dorsal interscapular region, and ventral caudal coelom). Staining with GMS identified rare hyphal structures within all deep granulomas ( Figures 4C–F ), except for one granuloma associated with a femoral pore (seven of eight, 87.5%).

Systemic N. guarroi infection was identified in one lizard (one of six, 16.67%; Animal ID: B), which had a pinpoint white lesion grossly apparent in the left cranial lung that corresponded to focal chronic granulomatous pneumonia with intralesional fungal hyphae ( Figures 4G, H ). GMS staining identified fungal structures, and N. guarroi qPCR from this site was positive. Grossly, two bearded dragons (Animal IDs: G, J) had mild to moderate, clear-to-yellow-colored coelomic fluid that was cytologically devoid of infectious agents or neoplastic cells and was of low cellularity. These were both females with resorbing and atretic ovarian follicles.