New antifungal 4-chloro-3-nitrophenyldifluoroiodomethyl sulfone reduces the Candida albicans pathogenicity in the Galleria mellonella model organism

Candida albicans represents an interesting microorganism to study complex host-pathogen interactions and for the development of effective antifungals. Our goal was to assess the efficacy of 4-chloro-3-nitrophenyldifluoroiodomethyl sulfone (named Sulfone) against the C. albicans infections in the Galleria mellonella host model. We assessed invasiveness of CAI4 parental strain and mutants: kex2Δ/KEX2 and kex2Δ/kex2Δ in G. mellonella treated with Sulfone. We determined that KEX2 expression was altered following Sulfone treatment in G. mellonella-C. albicans infection model. Infection with kex2Δ/kex2Δ induced decreased inflammation and minimal fault in fitness of larvae vs CAI4. Fifty percent of larvae died within 4–5 days (P value < 0.0001) when infected with CAI4 and kex2Δ/KEX2 at 109 CFU/mL; survival reached 100% in those injected with kex2Δ/kex2Δ. Larvae treated with Sulfone at 0.01 mg/kg 30 min before infection with all C. albicans tested survived infection at 90–100% vs C. albicans infected-PBS-treated larvae. Hypersensitive to Sulfone, kex2Δ/kex2Δ reduced virulence in survival. KEX2 was down-regulated when larvae were treated with Sulfone: 30 min before and 2 h post-SC5314-wild-type infection respectively. kex2Δ/kex2Δ was able to infect larvae, but failed to kill host when treated with Sulfone. Sulfone can be used to prevent or treat candidiasis. G. mellonella facilitates studding of host-pathogen interactions, i.e., testing host vs panel of C. albicans mutants when antifungal is dosed.


Introduction
Since incidences and frequency of the Candida albicans infections dominate over non-albicans species, there is still need to study the virulence of this pathogen, which represents an interesting window into the evolution of complex hostpathogen interactions and for the development of effective antifungal treatments [1][2][3]. The serine protease encoded by KEX2 is among the C. albicans' virulence factors that mediate its success as a pathogen. It is a representative of subtilisin family of proteins, processes enzymes, and its critical role in virulence was demonstrated previously [4][5][6][7][8]. As this approach still needs further analyses, here we continue studies assessing the Galleria mellonella-C. albicans infection model and the efficiency of novel antimicrobial agent. Recent studies [9][10][11][12][13][14][15][16][17][18][19][20] increased interest in the G. mellonella larvae as an alternative in vivo model due to the immunological and developmental similarities between insects and mammals. Therefore, the results obtained using these insects can serve as a starting point to study the C. albicans pathogenesis and generate hypotheses to be further tested in vertebrate models.
Since candidiasis is difficult to eradicate with the existing antimycotics [21], a new antifungal compound overcomes these deficiencies. We found [8] that the 4-chloro-3-Responsible Editor: Celia Maria de Almeida Soares nitrophenyldifluoroiodomethyl sulfone (named below Sulfone) treatment reduced the C. albicans pathogenicity in G. mellonella. It was shown [8] that the Sulfone displayed the minimal fungicidal concentration (MFC) against C. albicans at 0.25 μg/mL in in vitro studies and it was non-toxic against G. mellonella (lethal dose > 16 μg/mL). However, the impact of the Sulfone on metabolic pathways' inhibition in C. albicans is not well understood yet. Our goals were to assess the efficacy of the Sulfone against the C. albicans infections in the G. mellonella host model and to compare the KEX2 mutants' susceptibility with the Sulfone in vivo. Additionally, we investigated whether the KEX2 expression plays a role in the virulence of C. albicans in G. mellonella. The role of KEX2 in virulence was tested by screening for attenuation in the C. albicans mutants. We determined in histopathological examinations whether the KEX2 mutations affected the C. albicans' tissue invasion capabilities. We examined if the KEX2 expression is altered following the Sulfone treatment in the G. mellonella-C. albicans infection model.

4-Chloro-3-nitrophenyldifluoroiodomethyl Sulfone's synthesis
4-Chloro-3-nitrophenyldifluoroiodomethyl sulfone was synthesized according to the scheme in Fig. 1, starting with the commercial 4-chlorophenyldifluoromethyl sulfone, which was iodinated through the reaction with iodine bromide, carried out in carbon tetrachloride with potassium hydroxide as a base [8]. The next step was nitration by fuming nitric acid and concentrated sulfuric acid.

Strains culture, reagents, and growth conditions
Candida albicans strains used in the study are listed in Table 1. C. albicans cultures were grown in YEPD medium [24] at 30°C overnight. Cultures were washed in sterile PBS and adjusted to the required cell density. YEPD medium supplemented with uridine at the final concentration of 50 μg/mL [25] was used when required. Transformants were selected on the MMD (0.7% wt/vol yeast nitrogen base without amino acids, 2% wt/vol glucose, 770 μg/mL Complete Supplement CSM-URA) liquid and agar (2%) media. The 5-FOA medium (4% wt/vol glucose, 0.2% wt/vol 5-fluoroorotic acid, 1.34% wt/vol yeast nitrogen base with amino acids, 4% wt/vol agar) was used when required.
Generation of the C. albicans KEX2 mutant strains using the mini-ura-blaster technique [23] The C. albicans strains attenuated in KEX2 (Table 1) were constructed as follows: the two KEX2 alleles in C. albicans CAI4 [23] were disrupted using the URA3-dpl200 disruption cassette, that was PCR amplified from plasmid pDDB57 [25] using the specific primer pairs, containing complementary sequences to the 5′-and 3′-regions of the target ORF 19.4755 ( Table 2). The transformants were generated in two rounds using the lithium acetate/single-stranded carrier DNA/PEG method as described in [25,26] and the URA3 marker was recycled by growing transformants on the 5-FOA medium. Briefly, in the first round of the transformation, we used primers ( Table 2) amplifying plasmid DNA (URA3-dp1200 disruption cassette) in PCR with 70 bp of KEX2 flanking homology on either side and then transformed the PCR product into CAI4. The URA + transformants grown on the MMD medium were screened with PCR that had undergone homologous integration at KEX2. Following this, the transformants were placed on 5-FOA to recycle the selection marker and use this strain in the second round of gene disruption. Due to high transformation frequencies (37 ± 5.7 CAI4 transformants/μg DNA [27]), the Candida integrity plasmid CIP10 was used as the positive control in the lithium acetate procedure. Consistent with this, CIP10 transformed C. albicans CAI4 [23] about 20 times more efficiently than the control plasmid (YPB-ADHpt).

Galleria mellonella collection and treatment
The last instar larvae weighing between 180 and 250 mg were selected for the study. For killing assays, the C. albicans inoculum of 1 × 10 9 cells per 10 μL of PBS was tested per larva. Larvae were incubated at 28°C in 9cm Petri dishes without food for up to 96 h post-infection and inspected every 24 h for survival. Since larvae deprived of nutrition demonstrated increased susceptibility to infection with the fungal pathogen C. albicans [28], we used the same conditions to show unequivocally the Sulfone activity/toxicity using the C. albicans-G. mellonella model. It is worthy to note that despite slight differences in the level of hemocytes and antimicrobial peptides between non-and starved larvae, their effectivity in killing C. albicans is not affected [28]. Since it was found that prophylaxis test is more likely to be successful because a compound is administered before infection, here we tested the Sulfone for either its prophylaxis or post-exposure activity [29]. Thus, we chose four time points to test the Sulfone dosing, e.g., 30 min and 1 h before infection (b. i.) with C. albicans as well as 30 min and 1 h post-infection (p. i.). In one experiment, 10 larvae per group were tested, where each one was injected with 10 μL of the Sulfone (at 0.01 mg/kg of larva) and with the C. albicans inoculum of 1 × 10 9 cells per 10 μL of PBS (treated). The Sulfone and C. albicans inoculum were administered by injection into a different pro-leg using a Hamilton syringe. The Sulfone dose (0.01 mg/kg) was selected based on the previous studies [8] which proved that this dose is fungicidal against the C. albicans wild type and mutants (KEX2/kex2Δ and kex2Δ/kex2Δ) in in vitro studies. Syringes were changed between treatments with different strains. Ten larvae without the C. albicans inoculum and 10 larvae inoculated with RPMI or PBS were included for control purposes. The larvae were incubated for 96 h in the above-described conditions. Host health index [30] of the treated larvae was assessed to rank the virulence of the C. albicans strains against G. mellonella [30]. Thus, larvae were monitored daily for the following attributes: activity and survival. Larvae were considered dead if they did not move after stimulation. Randomly selected live, non-melanized larvae, and treated larvae (each in triplicate) were suspended in 0.5 mL of PBS and mechanically disrupted (homogenized), then centrifuged at 100,000g for 1 h. Briefly, 100 μL of aliquots from the serial dilution (1/10,000,000) of supernatants were placed onto YEPD agar medium with penicillin (100 μg/mL) and incubated at 37°C for 48 h. Thus, we confirmed the Sulfone's fungicidal activity (MFC = 0.25 μg/mL) in vivo by comparison of the number of CFU of the C. albicans recovered from the larvae treated with the Sulfone and with the starting inoculum. The Sulfone's antifungal activity was calculated using the formula: log reduction R = log CFU/ mL control Candida, log CFU/mL Candida treated with the Sulfone, where R means the relative number of live fungal cells eliminated by the antifungal agent. To interpret the Sulfone activity in vivo, we adopted the criteria described by Majoros et al. [31], where fungicidal agent is Reverse primer for ACT1 gctggtagagacttgaccaacca that causing a ≥ 3 log reduction; i.e., after the treatment with the agent, the number of fungal cells is 1000 times smaller than the initial (control) number of fungal cells.

Fungal cell staining with hematoxylin and eosin and periodic acid Schiff
Larvae were infected as described above. After 96 h of the maintenance of the larvae at 28°C (see above), the surviving larvae from each group such as controls (PBS, RPMI, untreated) and those exposed to Candida and the Sulfone were anesthetized by chilling at − 20°C for 18 h and then fixed, dehydrated, and stained, as described previously [8]. The stained tissues were observed using an Olympus FLUOROVIEW FV1000 confocal laser scanning microscope CLSM (Olympus, USA). Images were assembled using the Photoshop software (Adobe Photoshop CS3 Extended, France). Three larvae from each experimental group were used and the experiment was repeated on 3 independent occasions.

RNA isolation and reverse transcription-quantitative polymerase chain reaction
Larvae were injected as described above. Three surviving larvae originated from each experimental group after 96 h, i.e., the larvae injected with Candida and Sulfone in time intervals as well as the untreated control group; they were frozen in liquid nitrogen and ground to powder with a mortar and pestle. The samples were homogenized and total RNA of C. albicans was isolated as described previously [32]. The total RNA from each sample was used in separate reverse transcription reactions with oligo dT. Total RNA from each sample (5 μL [33] was then used to process these data to calculate the relative gene expression for the KEX2 experiment. The experiment was performed in triplicate.

Statistical analysis
Each experiment was performed at least in three replicates and the data were presented as mean values ± standard deviations (SD

Disruption of KEX2 alleles reduces fungal injury during the host-fungus interaction
Using the mini-ura-blaster technique, we disrupted KEX2 to study its relevance in the C. albicans virulence in vivo. The electrophoresis results of the CAI4 parental strain and KEX2 disruptants are presented in Fig. 2. The CAI4 strain (Ura − ) with two deleted alleles of URA3 was used as the parental strain (Fig. 2, line 1). The KEX2 alleles were replaced by the C. albicans URA3 gene flanked by direct repeats of the hisG sequence from Salmonella typhimurium (Fig. 2, lines 2-6). After CAI4 had been transformed with the URA blaster cassette, the Ura + transformants were selected on uracil-deficient medium. The cells that had lost URA3 by homologous recombination of the hisG sequences were selected on the medium containing 5-FOA because of its toxicity against the Ura + cells (Fig. 2, lines 7-8). The larvae were infected with 1 × 10 9 blastoconidia per larva of each strain: CAI4, KEX2/kex2Δ, and kex2Δ/kex2Δ, and the survival was monitored (Fig. 3).
Killing of the larvae depended on the strain injected. The larvae were killed significantly faster (P value< 0.0001) when infected with CAI4 and kex2Δ/KEX2 at the inoculum density tested. Fifty percent of the larvae died within 4-5 days, while the survival still reached 100% in those injected with kex2Δ/ kex2Δ, and after 6 days, all the latter larvae survived. The tissue sections were performed on the sixth day after the initiation of the fungal infection (Fig. 4). In the larvae treated with kex2Δ/kex2Δ, there were only very few infected areas compared with the larvae injected with CAI4 or KEX2/ kex2Δ. The larvae inoculated with kex2Δ/kex2Δ showed smaller nodules; these limited to the peripheral larval tissues. There was a relationship between the larvae death rate (Fig. 3) and progression to pathogenesis (mature nodules entrapping CAI4 or KEX2/kex2Δ appeared, Fig. 3a-d). Therefore, we found evidence of the resistance of the larvae against the infection with kex2Δ/kex2Δ (Figs. 3 and 4e, f).
The efficiency of the sulfone during the infection of larvae with C. albicans  (Table 3). We observed a reduction in the number of CFUs recovered from the Sulfonetreated larvae p. i. with C. albicans compared with the untreated larvae. In the Sulfone-treated larvae, the CFUs did not increase over the 6-day period as compared with the C. albicans control CFUs in the Sulfone-untreated larvae (Table 3). Collectively, the Sulfone increased the larval survival and caused 8.8-log reduction of CFUs when injected 1 h b. i. with CAI4 vs control. In contrast, the Sulfone inhibited the proliferation of C. albicans kex2Δ/KEX2 and kex2Δ/kex2Δ regardless of its administration (Table 3). Each infected larva was treated with the Sulfone at time course (b. i. and p. i.). The two major attributes such as activity and survival were assessed. The surviving larvae exhibited mobility after 96-h incubation. The wax worms injected with the Sulfone 1 h b. i. and p. i. with C. albicans showed a higher health index score (higher activity and survival) compared with those PBS-treated (Fig. 6, insignificant variation, P > 0.063).
The sulfone treatment modulates the KEX2 expression in the G. mellonella-C. albicans infection model KEX2 was quantified as the relative gene expression data from each sample, and was normalized to the endogenous reference ACT1 gene of C. albicans [32]. To evaluate the KEX2 expression in the C. albicans treated with the antifungal Sulfone, we chose fungal inoculum at conc. of 10 9 CFU/mL of PBS. Using the 2 −ΔΔCq method, it was shown that KEX2 was slightly   (Table 4).

Discussion
Besides C. albicans, kexin-like proteinases (Kex2) were identified in Saccharomyces cerevisiae, Pichia pastoris, and C. glabrata [34]. These Authors [34] showed that fungal Kex2 proteinases are similar in their substrate activities which have different functions according to the different biological backgrounds of the investigated fungi, including pathogenicity in humans. The KEX2 gene from C. glabrata showed 51% and 62% identity and high structural similarities to its counterparts in C. albicans and S. cerevisiae. Bader et al. [35] revealed that Kex2 is involved in the processing of the proteins that are essential for cell surface integrity of C. glabrata. In C. albicans, Kex2 plays a role in the cell wall formation and interferes with aspartic proteases (Saps) in the cell wall's remodeling mechanisms [4][5][6][7].
Moreover, the C. albicans cells attenuated in KEX2 showed a defect in polarized growth and morphology [4][5][6][7]. Our in vivo results (lack of hyphae in histological analysis Fig. 4) are consistent with the in vitro findings of Newport and Agabian [5] that the double KEX2 mutation affects morphogenesis. Furthermore, we are consistent with [5] showing that the KEX2 disruption in C. albicans has a pleiotropic effect such as elevated sensitivity to the Sulfone. Our studies in C. albicans showed that the KEX2 gene acts in its virulence process in vivo. It is a gene whose product processes enzymes which are critical for the virulence and ability of C. albicans to evade detection and destruction by the host's immune system [4][5][6][7]. Bearing in mind that the understanding of the role of KEX2 needs further investigation, we test (F test) of health index scores of wax worms was performed after 96-h incubation at 37°C. In each graph, the differences between two curves (Sulfone vs PBS) were not significant (P value > 0.063) Table 3 Fungicidal activity of the Sulfone (0.01 mg/mL) in the G. mellonella model in vivo at each time before infection (b. i)

Strains
Control C. albicans recovered from the untreated larvae CFU × 10 7 ± SD C. albicans recovered from the Sulfonetreated larvae CFU × 10 6 ± SD Logarithm reduction of C. albicans CFU recovered from the Sulfone-treated larvae log R* *Stands for decimal log reduction using the formula: log R = log CFU/mL control Candida, log CFU/mL Candida treated with the Sulfone would like to emphasize below several observations that suggest its role during fungal challenge in the G. mellonella model. In contrast with the parental strain CAI4, among the larvae infected with the kex2Δ/kex2Δ mutant, 100% survived after day 5 (Fig.  3). Notably, the larvae infected with the relevant kex2Δ/KEX2 mutant showed survival closer to the CAI4 parental strain (50% survived after day 4). This could be attributed to the absence of both of the KEX2 alleles which are required for the attenuation the C. albicans virulence trait, while both alleles are not needed to maintain the C. albicans' viability in vivo as shown here, and in vitro as previously reported [5]. Moreover, this phenomenon was described previously for other gene networks [36]. The larvae challenged with the kex2Δ/kex2Δ mutant displayed a decreased inflammatory response, as indicated by low nodule loads seen in the fat body (Fig. 4). Furthermore, this nodule-defective mutant displayed a minimal fault in fitness of the infected larvae. In contrast, the CAI4 parental cells exhibited abundant nodule formation comparable with KEX2/kex2Δ (Fig. 4). Interestingly, this consideration emphasizes further that the role of KEX2 is more attributable to virulence in vivo. These data and the previous reports [18,30] ruled out the ability of C. albicans to proliferate within hemocytes, leading to infection, and overwhelming the larval immune response.
On the basis of our previous study [8], we selected the Sulfone as the most promising lead candidate for further characterization in vivo. The findings presented in [8] and the current studies showed that administration of the Sulfone triggered immune response in larvae. We observed elevated hemocyte aggregation (Fig. 4) and fungicidal activity without the Sulfone's negative influence on survival and pupation (Fig. 6). We included the Sulfone antifungal compound in our screen in order to identify whether it targets KEX2. The kex2Δ/kex2Δ mutant was hypersensitive to the Sulfone, and reduced virulence in the survival assay was noted (Fig. 5). As judged from our results and described for known antifungal drugs elsewhere [37,38], complementation of other fungal genes can be successfully reduced by antifungal drugs when any one of them is deleted. In our study, this could be explained by the inability of other genes to substitute for the absence of KEX2 in that strain when the Sulfone treatment is performed. Inhibition of other genes by the Sulfone is sufficient to prevent the development of infection.
Our study first involved identifying the expression of the responsive KEX2 gene in vivo when treated with the Sulfone at 0.01 mg/kg per larva. The Sulfone affected the C. albicans cells' proliferation in vivo in the time course tested, except for the Sulfone injected 30 min b. i. with SC5314 (Table 4). In this condition, the KEX2 expression was almost unchanged, which indicates that the Sulfone both prevents the fungal cells' proliferation and inhibits the virulence factor, thus it prevents the KEX2 upregulation during infection. Targeting virulence is attractive in terms of antifungal drug development because it expands the extremely limited repertoire of targets in fungi [39]. Since the larvae injected with the CAI4 parental strain displayed lowered health index parameters compared with the relevant Sulfone-treated cells and mutants (Fig. 6), we hypothesized that the Sulfone targets KEX2. Due to the Sulfone's action mode (inhibition of filamentation, adhesion, and biofilm formation, described in [8]), we decided to use a pretreatment modality to mimic its use in a prophylactic antifungal regimen starting 1 h and 30 min b. i. with C. albicans, thus maximizing the chances to detect any protective effect. The larvae which were administered the Sulfone 30 min b. i. with all the C. albicans tested survived the infection at 90-100%, this compared with the mortality observed in the infected and PBS-treated larvae (placebo control group in Fig. 6). We did not observe a perfect correlation between the two experiments, All G. mellonella larvae were washed with 70% alcohol before examination; fungal RNA was isolated 6 days after treatment of larvae; 2 −ΔΔCq , the data are presented as the fold change in the KEX2 expression normalized to the reference ACT1 and relative to the non-targeting control* as measured in survival assays (Fig. 3 vs Fig. 6). For instance, the kex2Δ/kex2Δ mutant was less virulent in the untreated larvae (either with the Sulfone or PBS) than in the PBStreated ones 1 h b. i. with this mutant, yet the larvae displayed a similar fungal load (Fig. 6). One possible explanation is that for the infected larvae, additional stress related to the injection with PBS or the cuticle piercing can be deadly; the larvae started dying in large proportions already on day 1. Thus, the death rate may be misleading, with an error introduced by the PBS injection: the infected and PBS-treated larvae yielded higher mortality, whereas the infected but PBSuntreated ones survived. Thus, a definitive conclusion will be revealed when this experiment has been repeated much more extensively than in the preliminary studies reported here. As shown by Ignasiak and Maxwell [40], sodium chloride is not toxic against G. mellonella, but the cuticle piercing with a needle as such can be traumatic for the larvae. Thus, in each step in the study, three following control groups: untreated, traumatized, and buffer-injected control ought to be included [40]. These Authors [40] found that although the G. mellonella larvae cannot fully replace the mammalian models, they provide the statistical robustness which animal models lack. Moreover, the antibiotic doses recommended for use in humans can be effective in systemic infections in the larvae, and the acute toxicity of compounds in wax moth larvae correlate to the toxicity in mice and rats. Collectively, since the C. albicans virulence involves a complex regulatory network of genes, the presence of overlapping genes could mask any detectable phenotype due to an altered expression of other genes in the kex2Δ/kex2Δ cells. However, our results supported the well-established Sulfone's anti-virulence activity [8].
Our data of the Sulfone revealed striking observations: kex2Δ/KEX2 and kex2Δ/kex2Δ were less virulent when the Sulfone was administered 1 h or 30 min b. i. of the larvae. The C. albicans parental strain's inoculum of 10 9 CFU/mL of PBS induced the larval mortality at 50% compared with that those infected with the C. albicans kex2Δ/kex2Δ mutant. This mutant was able to infect the larvae but failed to kill the host cells when treated with the Sulfone. Moreover, treating the larvae with the Sulfone 30 min or 1 h b. i. with SC5314 prevented further C. albicans' growth and effectively prevented the larvae from death. In conclusion, our novel findings described in this article suggest that the Sulfone can be used in direct therapy to prevent or treat potentially fatal fungal infections. Since no animal or animal-derived model of infection completely replicates human diseases [37], the G. mellonella systemic candidiasis model that we used here facilitates the hostpathogen interactions, i.e., testing a host vs a panel of C. albicans mutants when a new antifungal agent is dosed.