Efficacy of antifungal agents against fungal spores: An in vitro study using microplate laser nephelometry and an artificially infected 3D skin model

Abstract Dermal fungal infections seem to have increased over recent years. There is further a shift from anthropophilic dermatophytes to a growing prevalence of zoophilic species and the emergence of resistant strains. New antifungals are needed to combat these fungi and their resting spores. This study aimed to investigate the sporicidal effects of sertaconazole nitrate using microplate laser nephelometry against the microconidia of Trichophyton, chlamydospores of Epidermophyton, blastospores of Candida, and conidia of the mold Scopulariopsis brevicaulis. The results obtained were compared with those from ciclopirox olamine and terbinafine. The sporicidal activity was further determined using infected three‐dimensional full skin models to determine the antifungal effects in the presence of human cells. Sertaconazole nitrate inhibited the growth of dermatophytes, molds, and yeasts. Ciclopirox olamine also had good antifungal activity, although higher concentrations were needed compared to sertaconazole nitrate. Terbinafine was highly effective against most dermatophytes, but higher concentrations were required to kill the resistant strain Trichophyton indotineae. Sertaconazole nitrate, ciclopirox olamine, and terbinafine had no negative effects on full skin models. Sertaconazole nitrate reduced the growth of fungal and yeast spores over 72 h. Ciclopirox olamine and terbinafine also inhibited the growth of dermatophytes and molds but had significantly lower effects on the yeast. Sertaconazole nitrate might have advantages over the commonly used antifungals ciclopirox olamine and terbinafine in combating resting spores, which persist in the tissues, and thus in the therapy of recurring dermatomycoses.


| INTRODUCTION
Dermatomycoses are mostly superficial fungal infections of the skin, hair, and nails caused by dermatophytes, yeasts, and more rarely, molds. Dermatomycoses are one of the most common dermatological conditions, with more than 25% of people affected worldwide, particularly in tropical and subtropical regions (Havlickova et al., 2008).
Although dermatomycoses are not life-threatening, they are considered a public health problem as they affect the quality of life of individuals (Silva et al., 2014) and account for at least half a billion dollars in health costs annually (White et al., 2014). A wide variety of topical and systemic antimycotic agents are available for the treatment of dermatomycoses (Nenoff et al., 2015), including allylamines, such as terbinafine; triazoles, such as fluconazole, itraconazole, voriconazole, and efinaconazole; imidazoles, such as ketoconazole, sertaconazole, and luliconazole; ciclopirox olamine; tavaborole; amorolfine hydrochloride and griseofulvin. The choice of agent depends on the site and the extent of the infection, as well as the causative organism (Gupta et al., 2005;Meis & Verweij, 2001). Topical antimycotics are frequently characterized by a broad therapeutic spectrum that includes dermatophytes, yeasts, and molds, as well as some Gram-positive bacteria (Nenoff et al., 2015). Superficial mycoses appear to have increased during recent years, although epidemiological data for Germany are scarce as there is no obligation to report these infections and no population-based epidemiological studies have been carried out. However, from 2007 to 2018, there was a noticeable increase in the number of samples processed in the mycological laboratory of the Department of Dermatology at Jena University Hospital . The turnover and the number of packages of terbinafine and ciclopirox olamine containing prescription and over-the-counter medications sold declined significantly in the German market from 2018 to 2020. However, the number of packages containing sertaconazole increased between 2018 and 2019 and declined more moderately than the other two between 2018 and 2020 (Insight Health Database, 2021). Not only the number of fungal infections is steadily rising but there is also a shift from anthropophilic dermatophytes, such as Trichophyton rubrum and Trichophyton interdigitale, to an increasing prevalence of zoophilic species, like Trichophyton benhamiae and Trichophyton mentagrophytes (Nenoff et al., 2014). Recently, T. interdigitale has been reclassified as an anthropophilic species, and the related zoophilic T. interdigitale strain (previously called T. interdigitale var. mentagrophytes) has been renamed T. mentagrophytes (de Hoog et al., 2017). Part of the T. mentagrophytes/interdigitale complex is the new multiresistant species T. indotineae (Tang et al., 2021). Resistant T. indotineae strains have recently emerged, producing new challenges . Consequently, failures of antifungal treatment have been observed, some of which could be related to resistance to the antifungals, such as terbinafine Gupta & Kohli, 2003;Gupta et al., 2021;Mukherjee et al., 2003;Nenoff et al., 2020;Rudramurthy et al., 2018;Salehi et al., 2018;Süß et al., 2019). Point mutations within Erg1, which encodes the squalene epoxidase, are responsible for the resistance to allylamines Ebert et al., 2020;Rudramurthy et al., 2018;Salehi et al., 2018). Interestingly, Ala448Thr Erg1 T. indotineae point mutants show also an often increase in azole resistance whereas the molecular mechanism of this linkage remains unresolved (Burmester et al., 2021Ebert et al., 2020). Azoles interacting with sterol 14-α demethylases and point mutations of the corresponding genes is one mechanism to obtain azole resistance (for review Rosam et al., 2021;Song et al., 2018). Recently, point mutations of sterol 14-α demethylase genes were also identified in azole-resistant T. indotineae strains (Burmester et al., 2021). Another mechanism of azole resistance is the activation of efflux pumps, such as ATP-binding cassette transporters or major facilitator superfamily transporters (for review Pérez-Cantero et al., 2020). Activation of efflux pumps was identified in T. rubrum as the molecular mechanism of azole resistance . The persistence of fungal spores could further underlie the failure of antifungal treatments, as well as recurrent infections and chronic diseases. Spores are dormant entities with a minimal metabolism and heavily reinforced cell walls, which make them less susceptible to many therapies (Seidl et al., 2015).
New antifungal agents are needed to combat the range of microorganisms. The antimicrobial activity and effectiveness of these antifungal agents must be characterized, usually using in vitro tests.
Such tests enable a direct comparison of the effects of the active substances on microorganisms. They are also simple, fast, reproducible, and inexpensive, as well as able to handle a large number of samples (Wiegand et al., 2015a). Microplate laser nephelometry (MLN) is a method for measuring light that is scattered up to 90°by particles suspended in a solution. It has been successfully employed to monitor the growth of bacteria (Wiegand et al., 2015a(Wiegand et al., , 2015bWiegand et al., 2012), yeasts (Finger et al., 2012;Seyfarth et al., 2008), and even dermatophytes  by monitoring the turbidity of the medium and to investigate the effect of different substances on the growth of these microorganisms.
Compared to other methods used (Suller & Russell, 1999;Walsh et al., 2003), MLN enables high-throughput screening, long-term incubation, and in situ monitoring of changes in dose-response curves, as well as the determination of the half-maximal inhibitory concentration (IC 50 ) and minimum inhibitory concentration (MIC).
Active substances are further required to exhibit targeted, antifungal activity in the presence of human cells. In vivo conditions are increasingly being mimicked using three-dimensional (3D) cell cultures. Such full skin models consist of a collagen matrix populated with primary human fibroblasts as dermis and a fully differentiated epidermis made of primary human keratinocytes (Reddersen et al., 2019;Wiegand et al., 2016). For the investigations in this study, the model was artificially infected with spores to enable the analysis of the sporicidal influence of the substances to be examined under relevant in vivo-like conditions. Compared to a simple 2D model, the 3D full skin model has the advantage that the cells are in their natural 3D environment and show corresponding behavior. The antifungal and sporicidal effects of various compounds can therefore be examined in situ in the presence of human cells. In addition, the influence on cell viability and the release of lactate dehydrogenase (LDH) over the incubation period was examined. Damage to the cell membrane can be measured by the release of the cytosolic enzyme LDH and can be regarded as a sign of cell necrosis, late apoptosis, and other forms of cellular damage (Wiegand & Hipler, 2009;Wiegand et al., 2010). Because the biocompatibility of the materials is further determined by other cellular and molecular factors, an analysis of cellular interleukin release can help in elucidating proinflammatory effects, which could not be demonstrated by cytotoxicity tests alone (Wiegand & Hipler, 2009;Wiegand et al., 2010). Such mediators are, for example, interleukins-1α, -6, and -8, which coordinate cell proliferation, cell migration, and possible inflammatory reactions.
This study aimed to investigate the sporicidal effects of a ser-

| Antifungal stock solutions and preparation of test solutions
Sertaconazole nitrate (Grupo Ferrer Internacional, S.A.), terbinafine, or ciclopirox olamine (Sigma-Aldrich; PHB standard quality) were dissolved in dimethyl sulfoxide (Sigma-Aldrich) at a final concentration of 10 mg/ml to prepare stock solutions. For the determination of antifungal activity, the corresponding test concentrations were prepared by diluting the stock solutions in the Sabouraud glucose medium (Merck). To identify effective concentrations, sertaconazole nitrate was used in the range of 0.001-50 μg/ml, terbinafine concentrations were varied between 0.00005 and 50 μg/ml, and ciclopirox olamine was tested in concentrations ranging from 20 to 100 μg/ml. To investigate the influence of the samples in the infected full skin models, the corresponding test concentrations (Table 1)   . In brief, the spores of dermatophytes and Scopulariopsis spp. were gently scraped from the agar plate surface and dispersed in 5 ml of sterile isotonic NaCl solution (9 g/L; Fresenius Kabi), which was then filtered through a cell strainer with a mesh size of 40 μm (Greiner Bio-One). Spore solutions (molds) were counted manually using disposable counting chambers (type Neubauer improved; Carl Roth). Candida cell cultures were obtained by inoculation of 20 ml of 2% Sabouraud glucose medium (Merck) with 1-2 colonies and by incubating the culture overnight at 30°C with shaking. Dilutions were made in a 2% Sabouraud glucose medium (Merck) to a final concentration of 2 × 10 3 spores/ml. The viability of the spores/cells was determined by plating on Sabouraud dextrose agar (Merck). Colonies were counted after 48 h at 37°C (yeast and mold) or up to 7 days of cultivation at room temperature (dermatophytes), and viability was determined as the percentage of the initial concentration.

| Determination of the antifungal effect using MLN
The MLN assay was performed as previously described

| Preparation of the 3D full skin models and infection with the microorganisms
Normal human dermal fibroblasts (Promocell) and normal human epidermal keratinocytes (Promocell) were used to produce the full skin models. Fibroblasts were maintained in Dulbecco's modified Eagle's medium (DMEM; BioConcept) with 2% fetal calf serum (FCS; PAN-Biotech), 5 ng/ml human fibroblast growth factor (CellSytems), and 5 μg/ml insulin (PeloBiotech), and keratinocytes were cultivated in keratinocyte basal medium (KBM; Promocell). Cell cultivation took place over 7 days in 75 cm² cell culture bottles (Greiner Bio-One) at 37°C in a 5% CO 2 atmosphere. The cells were then detached with trypsin-EDTA (Gibco, Thermo Fisher Scientific) and resuspended in the appropriate medium. To produce the dermis equivalent, fibroblasts were placed in 12-well inserts (Greiner Bio-One) with DMEM + 10% FCS + 1% gentamycin (Gibco, Thermo Fisher Scientific) + 150 μg/ml ascorbic acid (Sigma-Aldrich) and cultivated for 3 weeks at 37°C under 5% CO 2 . The medium was changed every 2-3 days. Before the keratinocytes were seeded to produce the epidermis, the medium surrounding the insert was completely removed, and the top of the dermis was coated with fibronectin (50 μg/ ml; Promocell). After 30 min incubation of the dermis equivalents with fibronectin, the keratinocytes were seeded at a density of 2 × 10 5 cells/insert in KBM + 5% FCS + 1% gentamycin + ascorbic acid (150 μg/ml). The full skin models were then incubated for 45 min in an incubator before being flooded with KBM + 5% FCS + 1% gentamycin + ascorbic acid (150 μg/ml). The skin models were cultivated for 7 days, submerged in the medium with decreasing FCS concentration (5%). To do this, the medium was changed every 2-3 days. After submerged cultivation, the full skin models were cultivated at the air-medium border (airlift cultivation) for a further 12 days. For this purpose, the full skin models were transferred to ThinCert 12-well plates (Greiner Bio-One) and supplemented with Green's medium (1:1 DMEM + DMEM HAMS-F12 (Gibco, Thermo Fisher Scientific), as previously reported (Reddersen et al., 2019;Wiegand et al., 2016). The medium was changed every 2-3 days. For artificial infection, collagen pellicles were prepared from a 6 mg/ml collagen suspension (Fraunhofer Institute) and loaded with 25 μl of a suspension of spores of the respective microorganisms (10 4 /ml, C. albicans: 2 × 10 2 /ml). The suspension was allowed to dry for 1 h. Then skin models were placed on the spore-loaded collagen pellicles.

| Statistical analysis
Experiments were carried out in two independent tests. Measurements were performed four times (MLN) or twice (full skin model).
The mean values ± standard deviation is given.

| DISCUSSION
Several antifungal drugs have been recommended for the treatment of dermatomycoses. However, it is still challenging to eliminate not only growing fungi but also fungal spores. Besides featuring a thick cell wall, fungal spores show minimal cell metabolism, which both reduces their susceptibility toward adverse environmental conditions, such as reduced oxygen content or antifungal therapy (Seidl et al., 2015). The main reasons for treatment failure and disease recurrence are a patient's failure to continue treatment (Behnam et al., 2020) and the formation of fungal spores (Seidl et al., 2015).
Reports of drug-resistant fungal strains are steadily increasing (Afshari et al., 2016;Ebert et al., 2020;  F I G U R E 2 (a) Influence of sertaconazole nitrate, ciclopirox olamine, and terbinafine on the viability of the full skin models as well as the release of (b) LDH and (c) IL-1α as a sign of damage to the full skin models. IL-1α, interleukin 1α; LDH, lactate dehydrogenase F I G U R E 3 (a) Behavior of 3D skin models infected with T. rubrum, E. floccosum, S. brevicaulis, and C. albicans either untreated or after treatment with sertaconazole nitrate, ciclopirox olamine, or terbinafine, with regard to the viability of the cells in the full skin models. Release of (b) LDH and (c) IL-1α as signs of damage to the full skin models, as well as monitoring of the secretion of the proinflammatory mediators (d) IL-6 and (e) IL-8. 3D, three-dimensional; IL-1α, interleukin 1α; LDH, lactate dehydrogenase Salehi et al., 2018;Yamada et al., 2017). Consequently, it is important to test the sporicidal effects, as well as the susceptibility of growing cells to old and new antifungal substances, using a variety of strains implicated in dermatomycoses.
The allylamine terbinafine is the best-studied antifungal drug, with most studies using T. rubrum or E. floccosum for testing, but several studies have also investigated T. indotineae, Candida species, or S. brevicaulis. Terbinafine inhibits the enzyme squalene epoxidase, a crucial enzyme involved in ergosterol biosynthesis (Osborne et al., 2005). Inhibition of this enzyme leads to the accumulation of squalene and depletion of ergosterol inside the fungal cells, ultimately leading to cell death (Afshari et al., 2016). Terbinafine is highly effective against the dermatophytes T. rubrum, T. soudanense, T. interdigitale, and E. floccosum, findings that confirmed those of earlier  (Jessup et al., 2000;Mock et al., 1998;Singh et al., 2007) and E. floccosum (Esteban et al., 2005;Jessup et al., 2000;Salehi et al., 2018). Nevertheless, T. indotineae and rubrum strains with Erg1 point mutations encoding for squalene epoxidase show an increase in terbinafine MIC 90 values of several magnitudes Rudramurthy et al., 2018). T. rubrum is an anthropophilic species, and together with T. interdigitale, it is the most common cause of skin and nail infections worldwide (Nenoff et al., 2014;Seebacher et al., 2008). Before T. rubrum and T. interdigitale, E. floccosum was the most common pathogen associated with tinea pedis (Zhan & Liu, 2017). Currently, a high incidence of Epidermophyton spp. infections is only reported from countries such as Iran or Nigeria (Zhan & Liu, 2017). However, E. floccosum remains associated with chronic courses of dermatophytosis (Vineetha et al., 2018). Only one report of the MIC values of terbinafine against T. soudanense was found (Mock et al., 1998), and this study produced results similar to ours. T. soudanense is an anthropophilic dermatophyte that occurs mainly in the tropics, especially in sub-Saharan Africa. Hence, infections with this dermatophyte are expected among travelers and immigrants from African countries and their contact persons, who require dermatological consultation (Nenoff et al., 2018). Silva et al. (2014) reported that slightly lower concentrations of terbinafine are needed to inhibit T. interdigitale growth than were identified in our study. However, higher concentrations were required in other studies (Behnam et al., 2020;Rudramurthy et al., 2018;Salehi et al., 2018).

The differences between the MICs determined in different studies
probably depend on the test methods, the growth conditions (e.g., incubation time) used, and the usage of fungal mycelia or spores.
Moreover, the MIC depends strongly on the isolates selected for testing and their susceptibility. This was most distinctly demonstrated for T. mentagrophytes, for which the terbinafine-resistant variant T. indotineae was tested here. Consistently higher concentrations were needed to kill the resistant isolate than previously reported (Badali et al., 2015;Behnam et al., 2020;Carrillo-Muñoz et al., 2008;Esteban et al., 2005;Jessup et al., 2000;Mock et al., 1998;Singh et al., 2007).
There is a veritable epidemic of chronic recalcitrant dermatomycoses due to T. indotineae in India . However, these dermatophytes are now also increasingly identified in Germany and other European countries Nenoff et al., 2020;Süß et al., 2019). Some studies also reported higher MIC values for terbinafine against T. rubrum, which might be attributed to the heightened occurrence of resistant isolates (Badali et al., 2015;Carrillo-Muñoz et al., 2008;Esteban et al., 2005;Rudramurthy et al., 2018;Salehi et al., 2018;Silva et al., 2014). Higher concentrations of terbinafine were also needed to kill the mold S. brevicaulis in this study than were found in previous reports (Carrillo-Muñoz et al., 2008;Cuenca-Estrella et al., 2003;Skóra et al., 2014). Scopulariopsis spp. are common soil saprophytes and have been isolated from a wide variety of substrates (Cuenca-Estrella et al., 2003). They can cause superficial, subcutaneous, and invasive infections in humans that are very difficult to treat. S. brevicaulis is resistant to amphotericin B, flucytosine, and azole compounds in vitro without a prior history of antifungal treatment (Cuenca-Estrella et al., 2003).
No inhibition of C. albicans could be achieved with terbinafine in soluble amounts, a finding that is consistent with reports that terbinafine has little to no effect on Candida spp., although the filamentous form is thought to be susceptible. Hence, a wide effective concentration range is usually found with Candida spp. (Craik et al., 2017;Jessup et al., 2000;Ryder et al., 1998;Silva et al., 2014).
Ciclopirox olamine has been examined in a number of studies as a promising sporicidal agent. It acts through the chelation of polyvalent metal cations, leading to the inhibition of many cellular activities and modification of the fungal plasma membrane (Monti et al., 2019). It was believed to be the only antifungal drug with a sporicidal action as spores are resting cells and therefore do not synthesize ergosterol and, consequently, antifungal drugs that inhibit enzymes necessary for ergosterol production are ineffective (Seidl et al., 2015). However, only Schaller et al. reported a similar MIC as observed here against T. rubrum (Schaller et al., 2009), and mostly, lower concentrations were reported to kill T. rubrum (Rudramurthy et al., 2018;Singh et al., 2007), T. interdigitale (Rudramurthy et al., 2018), T. mentagrophytes (Singh et al., 2007), E. floccosum (Singh et al., 2007), S. brevicaulis (Skóra et al., 2014), and the yeasts C. albicans and C. parapsilosis (Craik et al., 2017;Figueiredo et al., 2007).
As data regarding sertaconazole are scarce, this study aimed to investigate its effects against some of the most common and increasingly prevalent causes of dermatomycoses. Like other azoles, sertaconazole compounds inhibit the synthesis of the cell membrane component ergosterol by inhibiting sterol 14α-demethylase, which results in the accumulation of sterol precursors and the disruption of mycelial growth and replication (Croxtall & Plosker, 2009;Monod et al., 2019). However, at higher concentrations, sertaconazole has been found to bind directly to nonsterol lipids in the fungal cell membrane, leading to increased permeability of the wall and subsequent lysis of the mycelium (Croxtall & Plosker, 2009). In general, similar MICs against T. rubrum have been reported for sertaconazole (Carrillo-Muñoz et al., 2011;Croxtall & Plosker, 2009;Rudramurthy et al., 2018). In the case of T. interdigitale (Croxtall & Plosker, 2009;Rudramurthy et al., 2018), E. floccosum (Carrillo-Muñoz et al., 2011;Croxtall & Plosker, 2009), C. albicans, and C. parapsilosis (Croxtall & Plosker, 2009) (Croxtall & Plosker, 2009). Our results provide evidence that sertaconazole nitrate has sporicidal effects. In general, spores show an increased resistance due to a thick proteinaceous coat, a relatively impermeable inner spore membrane, a low water content, and proteins that protect the DNA. Despite these properties, spores can be killed, when crucial spore proteins, the spore's inner membrane, and one or more components of the spore germination apparatus are damaged (Setlow, 2014). In addition to inhibiting the ergosterol synthesis, a major constituent of fungal cell wall membranes, sertaconazole nitrate can directly bind to non-sterol lipids in the cell membrane (Agut, Palacin, Sacristan et al., 1992;Agut, Palacin, Salgado et al., 1992;Elewski, 1993), which leads to increased permeability of fungal cell membranes and thereby to a rapid leakage of key intracellular components including adenosine triphosphate. This direct membrane damaging effect was shown in vitro, where sertaconazole produces a dose-dependent reduction of intracellular ATP levels in suspensions of C. albicans (Agut, Palacin, Salgado et al., 1992). The possibility to assert this damage also on spore membranes, differentiating sertaconazole nitrate from other membrane-active antifungals, could be conferred by the lipophilic benzothiophene ether of sertaconazole nitrate, which increases permeation capability through protein-rich layers (Pfaller & Sutton, 2006). In addition, it is also conceivable that sertaconazole nitrate is adsorbed at or into the cell wall and further absorbed into the cell and inhibiting ergosterol biosynthesis from within. Seidl et al. (2015) postulated this additional mode of action, leading to the maintenance of the resting state due to insufficient ergosterol synthesis. The sporicidal activities of terbinafine, ciclopirox olamine, and sertaconazole nitrate were further examined using a 3D skin model artificially infected with T. rubrum, E. floccosum, S. brevicaulis, or C. albicans. Several studies on mostly bacterial, but also fungal, colonization using 3D epidermal models consisting of differentiated keratinocytes on an artificial membrane have been published so far (de Breij et al., 2012;Lerebour et al., 2004;Müller et al., 2014;Shepherd et al., 2009;Son et al., 2014). Some also used 3D full skin models, featuring a dermal layer with fibroblasts and an epidermal layer with differentiated keratinocytes, as they are present in human skin, to reflect a more in vivo-like situation for infection studies (Haisma et al., 2014;Holland et al., 2009Holland et al., , 2008Kitisin et al., 2020;Popov et al., 2014). Recently, such a 3D skin model was successfully employed to investigate the antiseptic treatment of S. aureus infections in a skin-like environment (Reddersen et al., 2019) and to explore the effects of T. benhamiae contagion on human skin cells in vitro (Hesse-Macabata et al., 2019).
To the best of our knowledge, this is the first report on artificial infection of a 3D skin model with different types of fungal spores and subsequent treatment with antifungal agents.
Forty-eight hours after inoculation of the spores into the lower dermal layers, the yeast had spread widely and hyphae growth was even macroscopically recognizable on the surface of the skin model. In summary, these findings demonstrate the necessity of testing antifungal agents under in vivo-like conditions to evaluate their efficacy. Several studies stressed that fungal spores are the primary cause of infection with dermatophytes (Aljabre et al., 1992;Rashid et al., 1995;Seidl et al., 2015). Consequently, only an antifungal with sufficient activity against the resting bodies of fungi and yeasts can deliver sufficient results to achieve acceptable long-term cure rates.