Surveillance of Amphotericin B and Azole Resistance in Aspergillus Isolated from Patients in a Tertiary Teaching Hospital

The genus Aspergillus harbors human infection-causing pathogens and is involved in the complex one-health challenge of antifungal resistance. Here, a 6-year retrospective study was conducted with Aspergillus spp. isolated from patients with invasive, chronic, and clinically suspected aspergillosis in a tertiary teaching hospital. A total of 64 Aspergillus spp. clinical isolates were investigated regarding molecular identification, biofilm, virulence in Galleria mellonella, antifungal susceptibility, and resistance to amphotericin B and azoles. Aspergillus section Fumigati (A. fumigatus sensu stricto, 62.5%) and section Flavi (A. flavus, 20.3%; A. parasiticus, 14%; and A. tamarii, 3.1%) have been identified. Aspergillus section Flavi clinical isolates were more virulent than section Fumigati clinical isolates. Furthermore, scant evidence supports a link between biofilm formation and virulence. The susceptibility of the Aspergillus spp. clinical isolates to itraconazole, posaconazole, voriconazole, and amphotericin B was evaluated. Most Aspergillus spp. clinical isolates (67.2%) had an AMB MIC value equal to or above 2 µg/mL, warning of a higher probability of therapeutic failure in the region under study. In general, the triazoles presented MIC values above the epidemiological cutoff value. The high triazole MIC values of A. fumigatus s.s. clinical isolates were investigated by sequencing the promoter region and cyp51A locus. The Cyp51A amino acid substitutions F46Y, M172V, N248T, N248K, D255E, and E427K were globally detected in 47.5% of A. fumigatus s.s. clinical isolates, and most of them are associated with high triazole MICs. Even so, the findings support voriconazole or itraconazole as the first therapeutic choice for treating Aspergillus infections. This study emphasizes the significance of continued surveillance of Aspergillus spp. infections to help overcome the gap in knowledge of the global fungal burden of infections and antifungal resistance, supporting public health initiatives.


Introduction
The genus Aspergillus harbors numerous cosmopolitan fungal species frequently found in diverse natural habitats, especially soil and decaying organic matter.They are responsible for food spoilage, mycotoxin contamination, and various human and animal mycoses [1].The clinical manifestations of human infections caused by species of this genus can vary from superficial mycoses to allergic conditions, asthma, and invasive aspergillosis, the latter especially in immunocompromised patients [2].The fungal infection varies according to the pathogen versus host relationship.Further, diverse species and strains within the Aspergillus genus present assorted virulence patterns and resistance to commercial antifungals.Thus, the precise identification of infection-causing agents is highly relevant to selecting optimal antifungal therapy, patient management, and prevention efforts [3].
In recent decades, the use of the polyphasic approach to identify Aspergillus spp.has become more common.The technique combines several data and information, such as phenotypic, molecular (including genomics, proteomics, and secondary metabolites), genetic, and phylogenetic data to reach a consensus taxonomy [4].The genus Aspergillus is currently divided into subgenera and sections, and the commonly isolated species from clinical samples are mostly found within the Aspergillus sections Fumigati (A. fumigatus sensu stricto), Flavi (A. flavus), Terrei (A. terreus), and Nigri (A. niger).Additionally, cryptic species in the section Fumigati (A. udagawae, A. lentulus, and A. pseudofischeri) are also isolated from clinical samples.They present an intrinsic resistance to antifungals, which impose an additional complication to antifungal treatment [5].
The management of aspergillosis requires early infection recognition and species identification.Depending on the severity of the infection, rapid initiation of appropriate antifungal treatment is highly recommended.Amphotericin B was previously considered first-line treatment of aspergillosis.Currently, professional societies have recommended voriconazole, isavuconazole, or even posaconazole as first-line therapy.Combination therapy with voriconazole and echinocandin is indicated for selected patient groups, although the latter should not be used as monotherapy.Itraconazole is indicated for non-invasive aspergillosis infections.The antifungal prophylaxis for patients at increased risk of invasive mold infections is the long-term administration of posaconazole or voriconazole [6].
Over the past decade, studies have reported an increase in the frequency of antifungal resistance among clinical isolates of Aspergillus spp.[7].In A. fumigatus, triazole resistance is related to a higher MIC of at least one triazole antifungal agent based on the epidemiological cutoff values (ECV) values or clinical breakpoints (CBP) [8,9].The emergence of azole-resistant strains has been attributed to the use of azoles in agriculture or prolonged treatments in the clinic [10,11].The azole resistance is a complex one-health challenge that involves many stakeholders.Among the diversity of involved professionals are medical doctors and fungi researchers [7].In 2022, the World Health Organization (WHO) ranked A. fumigatus among the four 'critical threat' pathogens due to the high perceived public health importance.This fungal prioritizing list guides research, development, and public health action [12].
Standard protocols have been established to test the antifungal susceptibility of clinically relevant microorganisms.In this context, reliable antifungal minimal inhibitory concentration (MIC) values are a priority for antifungal resistance screening and antifungal therapy guidance for patients [13].For MIC interpretation, ECVs identifies wild-type (WT) and non-WT profile for diverse Aspergillus spp.against commercial antifungals [8,14].In contrast, the CBP defines the susceptibility (susceptible, intermediate, or resistant) of A. fumigatus isolates against voriconazole [9].These data are potentially helpful for antifungal treatment guidance.
The azole antifungal target is the 14-α sterol demethylase, encoded by the cyp51A allele, which catalyzes a crucial step in the ergosterol biosynthesis of fungi.Many azoleresistant A. fumigatus clinical isolates do not present mutations within the cyp51A allele [15].Nonetheless, most azole-resistant strains have mutations in the cyp51A allele and modifications in the promoter region.The most prevalent azole-resistance-related modifications in the cyp51A promoter region are tandem repeats (TR) of 34 bp (TR34), 46 bp (TR46), and 53 bp (TR53), which lead to increased expression of cyp51A.The single point mutations in the A. fumigatus cyp51A allele associated with the azole-resistant strains are the Cyp51A amino acid substitutions G54, P216, M220, G138, and G448 [16].Particular combinations of point-mutations in the cyp51A allele and promoter modifications are involved in multiazole resistance, such as TR34/L98H, TR34/R65K/L98H, and TR46/Y121F/T289A.In addition, specific azole-binding amino acids that do not compromise Cyp51A activity can reduce the enzyme's drug affinity and alter the stability and functionality of Cyp51A.This alteration makes substrate recognition more challenging, leading to varied patterns of azole resistance.These substitutions can also obstruct the azole entry channel or disturb the heme group's position within the protein, thereby diminishing the azole's ability to bind to the heme group effectively and permitting substrate replacement [16].The clusters of three and five Cyp51A amino acids substitutions ('F46Y, M172V, and D255E'; 'F46Y, M172V, N248T, D255E, and E427K') have been described with different profiles of azole susceptibility, albeit with notably higher azole MICs than wild-type strains of A. fumigatus.The strains with these amino acid substitutions also present three silent mutations that encode the Cyp51A amino acids G89, L358, and C454, which present phylogenetic importance [16][17][18].The cluster of five Cyp51A amino acid substitutions was first described in a clinical strain isolated in 2001 [10].Since then, 10% of described A. fumigatus sensu stricto isolates harbor this cluster [10,18].
For triazole-resistant A. fumigatus strains, amphotericin B (AMB) has been commonly recommended by experts as a salvage and last-line antifungal treatment of fungal infections in a hospital environment [19,20].Little is known about the susceptibility pattern of Aspergillus sp. to AMB in many parts of the world [21].To date, it is known that the resistance to AMB has been frequently observed in A. terreus, with low frequency in A. flavus [22,23] and rare frequency in A. fumigatus [21,24].Although the resistance to AMB is less common than the resistance to fungistatic agents such as triazoles [19], populations of A. fumigatus from Korea, Canada, and Brazil have been emerging with high AMB minimal inhibitory concentration (≥2 µg/mL), becoming a significant public health concern [21,23,25,26].
Understanding the epidemiology and antifungal resistance of Aspergillus spp. is essential to improving prevention and patient management efforts in the medical field.In this way, we studied 64 Aspergillus spp.clinical isolates recovered over 6 years (2013 to 2019) from patients at the Clinical Hospital of the Medical School of Ribeirão Preto, University of São Paulo, Brazil, regarding molecular identification, antifungal susceptibility, resistance to AMB and azoles, biofilm, and virulence in Galleria mellonella.

Fungal Isolates, Conidia Production, and Ethics
Sixty-four Aspergillus spp.clinical isolates were included in the study.The clinical isolates were recovered over 6 years (2013 to 2019) from respiratory and other samples from patients at the Clinical Hospital of the Medical School of Ribeirão Preto, University of São Paulo, Brazil.The clinical isolates were deposited in both culture collections of Mycology Laboratory Hospital and Laboratório de Micologia Clínica (LMC)-FCFRP/USP, Brazil.The control strains were A. flavus ATCC204304 and A. fumigatus ATCC46645.The clinical isolates were also characterized and speciated by molecular methods at Laboratório de Micologia Clínica (LMC)-FCFRP/USP.The clinical isolates were freshly sub-cultured on yeast extract agar dextrose (YAG) at 37 • C for 2 days.The conidia were suspended with autoclaved double-distilled water, filtered through autoclaved Miracloth (22 to 25 µm, Merck EMD Millipore Corporation, Billerica, MA, USA), and counted under a hemocytometer.The conidia concentration was adjusted according to the needs of each experiment.

Molecular Identification of Aspergillus spp. Clinical Isolates
Genomic DNA (gDNA) was extracted according to modifications of the method described by Junghans and Metzlaff [27].The concentration of 2 × 10 7 Aspergillus spp.conidia were inoculated in 30 mL of dextrose yeast extract (YG) liquid culture medium and incubated at 37 • C at 180 rpm (Shaker Ecotron-Infors HT, Basel, Switzerland) overnight to produce mycelium, which was collected by vacuum filtration in funnel porcelain, frozen in liquid nitrogen, and crushed with the aid of a sterilized mortar and pestle.For every 40 mg of ground mycelium, 500 µL of extraction buffer (200 mM Tris-HCl pH 8.5, 250 mM sodium chloride, 25 mM EDTA, 0.5% SDS) was added.Then, an equal volume of phenol (Sigma Aldrich, St. Louis, MI, USA) and chloroform (JT Baker, Radnor, PA, USA) 1:1 (v/v) was added, and the mixture was homogenized for about 10 min in a tube shaker (Vortex Mixer-Labnet International, Edison, NJ, USA) to precipitate proteins and break down cell walls and membranes.The samples were centrifuged for 15 min at 12,000× g (Eppendorf, Hamburg, Germany) to sediment the precipitated proteins and cell debris.The aqueous phase was transferred to a new 1.5 mL microtube (Axygen Scientific, Glendale, AZ, USA), and the same volume of chloroform was added to remove phenol and/or chloroform residue.The samples were centrifuged again at 12,000× g for 5 min, and the upper aqueous phase was transferred back to a new 1.5 mL microtube where 0.54 volumes of isopropanol (JT Barker, Radnor, PA, USA) were added and gently shaken to DNA precipitation.After centrifugation at 12,000× g for 1 min, the pellet was washed with 70% ethanol and centrifuged again at 12,000× g for 1 min.The supernatant was discarded, and the ethanol residue evaporated at room temperature for 30 min.The sediment was suspended in sterilized water (Nuclease Free Water-Promega, Tokyo, Japan) and stored at 4 • C. RNA was eliminated from each sample by treatment with 300 ng/mL of RNAse (Pure Link A-Invitrogen Thermo Fisher, Waltham, MA, USA) at 37 • C for 1 h.The concentration and purity of genomic DNA were evaluated by a spectrophotometer (Implen-München, Germany) at 260 and 280 nm wavelengths.The DNA quality was observed in 1% agarose gel electrophoresis.

Conidia Measurements and Biofilm Quantification
The freshly cultured conidia were observed under light microscopy (1000× magnification, N120, Coleman, Chicago, IL, USA), and the images were captured with HDCE-X5 camera/ScopImage 9.0 software.The conidia diameter measurements were performed with ImageJ version 1.53e software.The average and standard deviation of 20 measurements were performed per clinical isolate.The results were expressed in a scatter plot, and an unpaired t-test was performed to analyze the difference in diameter between groups.The fungal biofilm was produced in a 96-well flat-bottom plate with 0.5 × 10 6 conidia/mL in RPMI 1640 culture medium buffered with 0.165 M 3-(N-morpholino)propanesulfonic acid (MOPS, Sigma Aldrich Chemical Corporation, Karnataka, India), pH 7.0 at 37 • C for 48 h.Subsequently, the non-adherent conidia were washed three times with sterile PBS1×.The adhered conidia were fixed with 100% methanol for 15 min.The biofilm and matrix biomass quantification were performed according to Li and collaborators [30] and Seidler and collaborators [31], respectively.Briefly, 100 µL of 0.5% Crystal Violet (Sigma Aldrich Chemical Corporation, Karnataka, India) for biomass quantification or 100 µL of 1% Safranin (Sigma, St. Louis, MI, USA) for matrix quantification were added to each well with pre-cultured biofilm and incubated for 20 min at room temperature.The dyes were removed, and the excess was washed five times with PBS 1×.The adhered cells were discolored with 200 µL of 33% acetic acid for 5 min at room temperature.One hundred microliters of the supernatant were transferred from each well to a new 96-well plate, and the absorbance was read in a plate reader (680XR-BioRad, Hercules, CA, USA).The wavelengths were 570 nm and 492 nm for biomass and matrix, respectively.The results were expressed by a correlation graph (biomass × matrix) using the Pearson method in the GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA).Data were analyzed considering a 95% confidence interval.Data were arbitrarily split into low, medium, and high biofilm producers within the absorbance range of 0.0 to 0.5, 0.501 to 1.050, and 1.051 to 2.5, respectively.

Galleria Mellonella Virulence Experiments
The virulence of Aspergillus spp.clinical isolates were characterized in the invertebrate model G. mellonella, according to Renwick and collaborators [32], with modifications.Groups of 10 G. mellonella larvae at the sixth instar of development (225 ± 25 g) were selected for each tested clinical isolate, PBS 1× control, and naïve.Artificial infection in G. mellonella larvae was performed, and the conidia concentrations were 1 × 10 6 conidia/mL for section Fumigati and 1 × 10 5 conidia/mL for section Flavi.The experiment was carried out for 10 days with three biological replicates, and the most significant result was expressed in a graph.The graphs and statistical analyzes by the Log-rank method (Mantel-Cox) were performed with GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA).Data were arbitrarily split into high, medium, and low virulence by considering the death of all larvae in each group at the respective intervals of zero to 4 days, 5 to 7 days, and 8 to 10 days after the fungal inoculum in each larva.

Antifungal Susceptibility Testing and Data Analyses
Minimal inhibitory concentration (MIC) testing was performed according to the broth microdilution method protocol M38-A2 of Clinical and Laboratory Standards Institute [33] for amphotericin B (AMB), voriconazole (VOR), posaconazole (POS), and itraconazole (ITR).Fungal isolates were grown on yeast extract agar dextrose (YAG) for conidia production at 37 • C for 48 h.A suspension of fungal conidia on distilled sterile water was filtered with a Miracloth filter and adjusted to a 5 × 10 5 conidia/mL concentration using a hemocytometer.Final concentrations were 2.5 × 10 5 conidia/mL, and serial antifungal dilution (0.0625 to 32 µL/mL) were diluted in RPMI 1640 culture medium buffered with 0.165 M 3-(Nmorpholino)propanesulfonic acid (MOPS, Sigma Aldrich Chemical Corporation), pH 7.0 and incubated at 37 • C for 48 h.The control strain Aspergillus flavus (ATCC 204304) was used to validate the experiment.Data were recorded by visual observation, and the minimal inhibitory concentration (MIC) was defined as the lowest concentration of antifungal that produces inhibition.Results were expressed by reporting the MIC range, geometric mean (GM), and MIC that inhibits the growth of 50% and 90% of the clinical isolates (MIC50/MIC90).The MIC values were evaluated according to the epidemiological cutoff value (ECV) with capture ≥ 97.5% of the statistically modeled population [14].Clinical isolates with MIC ≤ ECV were considered wild-type, while isolates with MIC > ECV were non-wild-type to antifungal, i.e., possible antifungal-resistant strain [14].

Statistical Analysis
Statistical analysis was based on the one-way analysis of variance method (one-way ANOVA), followed by Tukey's post-test for comparative analysis of the experiments.For the virulence test, the Long-rank statistical method (Mantel-Cox) was adopted to assess the significance of the survival curves.p values less than 0.05 were considered significant, and the result was considered statistically different.All analyses were performed using the GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA).S1 and S2).

Aspergillus
The virulence factors (conidia size and biofilm) were investigated for all clinical isolates.The mean diameter of the conidia was 2.796 µm and 4.511 µm for Aspergillus Section Fumigati (n = 40) and Flavi (n = 24) clinical isolates, respectively.An unpaired t-test comparing the mean values of the two groups was performed using GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA), and the smallest conidial size was observed for Aspergillus Section Fumigati (p < 0.0001) in comparison with Aspergillus Section Flavi clinical isolates (Figure 1).
South Brisbane, Australia) and analyzed by the multiple alignment methods using MEGA version 11.0.8 [37] (KUMAR et al., 2018).The sequences were compared with the sequence of a wild-type strain of A. fumigatus (GenBank accession number AF338659.1)[34].

Statistical Analysis
Statistical analysis was based on the one-way analysis of variance method (one-way ANOVA), followed by Tukey's post-test for comparative analysis of the experiments.For the virulence test, the Long-rank statistical method (Mantel-Cox) was adopted to assess the significance of the survival curves.p values less than 0.05 were considered significant, and the result was considered statistically different.All analyses were performed using the GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA).S1 and S2).

Aspergillus
The virulence factors (conidia size and biofilm) were investigated for all clinical isolates.The mean diameter of the conidia was 2.796 µm and 4.511 µm for Aspergillus Section Fumigati (n = 40) and Flavi (n = 24) clinical isolates, respectively.An unpaired t-test comparing the mean values of the two groups was performed using GraphPad Prism software (v 5.0; GraphPad Software, La Jolla, CA, USA), and the smallest conidial size was observed for Aspergillus Section Fumigati (p < 0.0001) in comparison with Aspergillus Section Flavi clinical isolates (Figure 1).Pearson's correlation coefficient was used to understand the existing correlation between the data of the biofilm biomass and matrix of Aspergillus sections Fumigati and Flavi clinical isolates.The analysis demonstrated a significant positive correlation between biomass and matrix (p < 0.0001) (Figure 2A).Thus, the amount of biomass corresponds to a similar amount of matrix for each clinical isolate.Among Aspergillus section Fumigati clinical isolates, 25% (n = 10/40), 65% (n = 26/40), and 10% (n = 4/40) are low, medium, and high biofilm producers, respectively.Similarly, for Aspergillus section Flavi clinical isolates, 20.8% (n = 5/24), 66.6% (n = 16/24), and 12.6% (n = 3/24) are low, medium, and high biofilm producers, respectively (Figure 2, Supplementary Table S3).No statistical differences were found between the amount of biomass (p < 0.803) and matrix (p < 0.054) formation capacity between Aspergillus sections Fumigati and Flavi clinical isolates (Figure 2B,C).
t-test comparing the mean values of the two groups was performed.The p-value was performed with a significance level of 5% (p < 0.05).
Pearson's correlation coefficient was used to understand the existing correlation between the data of the biofilm biomass and matrix of Aspergillus sections Fumigati and Flavi clinical isolates.The analysis demonstrated a significant positive correlation between biomass and matrix (p < 0.0001) (Figure 2A).Thus, the amount of biomass corresponds to a similar amount of matrix for each clinical isolate.Among Aspergillus section Fumigati clinical isolates, 25% (n = 10/40), 65% (n = 26/40), and 10% (n = 4/40) are low, medium, and high biofilm producers, respectively.Similarly, for Aspergillus section Flavi clinical isolates, 20.8% (n = 5/24), 66.6% (n = 16/24), and 12.6% (n = 3/24) are low, medium, and high biofilm producers, respectively (Figure 2, Supplementary Table S3).No statistical differences were found between the amount of biomass (p < 0.803) and matrix (p < 0.054) formation capacity between Aspergillus sections Fumigati and Flavi clinical isolates (Figure 2B,C).The virulence of Aspergillus sections Fumigati and Flavi clinical isolates was determined in the alternative virulence model Galleria mellonella.The advantages of this model are the similarity between the defense mechanism used by larvae and the innate immune system of vertebrates, the high reproduction rate, easy maintenance in the laboratory, and the larvae survival in a wide range of temperature (18 • C to 37 • C), which facilitates the study with human pathogens [38,39].The G. mellonella larvae were artificially infected with conidia of Aspergillus sections Fumigati and Flavi clinical isolates, and the fungal virulence profile was scored according to the larvae survival percentage.Whereas Aspergillus section Fumigati clinical isolates presented 29.6% (n = 8/27), 18.5% (n = 5/27), and 51.85% (n = 14/27) of high, medium, and low virulence, respectively, Aspergillus section Flavi clinical isolates presented 82.6% (n = 19/23) and 17.4% (n = 4/23) of high and low virulence in G. mellonella larvae, respectively (Figure 3 and Supplementary Table S3).We performed a Mantel-Cox test based on the survival experience of the subjects in the different groups being compared.The analysis showed that Aspergillus section Flavi clinical isolates were more virulent than section Fumigati clinical isolates (p < 0.0001).

Discussion
Aspergillus spp.are the major filamentous fungus that causes invasive infection, mainly in immunocompromised patients [40].The identification of Aspergillus species was previously usually based on phenotypic features.However, molecular characterization has strongly influenced it in recent decades [4].The precise identification of these species is necessary since the infection varies according to the pathogen versus host relationship.Regarding the pathogen, it is mainly due to the great diversity of species and strains within the genus Aspergillus that present an assortment of infection patterns, resistance to commercial antifungals, mainly to drugs from the polyene class (amphotericin B), azoles, and echinocandin group.Nonetheless, the clinical incidence of fungal infections, especially resistance to antifungals, is not widely known [3].
Our study identified fungal clinical isolates from Brazilian patients in the two main Aspergillus sections, Fumigati and Flavi.The pathogenic species here identified (A.fumigatus s.s., A. flavus, A. parasiticus, and A. tamarii) are globally associated with superficial, subcutaneous, chronic, and systemic mycoses [26,41,42].Fungal infection occurs through inhalation of airborne fungal propagule, although traumatic inoculation of this propagule also occurs less frequently.The conidia and spores play a crucial role in Aspergillus spp.infections, and their characteristics, such as pigment and size, directly impact the fungus's pathogenicity.For instance, Aspergillus spp.have two pigments associated with stress resistance mechanisms, dihydroxynaphthalene melanin (DHN)-melanin and pyomelanin [43].Regarding the size of the fungal conidia, A. fumigatus produces small conidia size (2-3 µm) reaching the intimate parts of the lung tissue and causing the most invasive infections.In contrast, A. flavus presents a large conidia size (4-6 µm), which is easily retained in the sinuses of the face, causing sinocranial infections [43].Our study observed the larger conidia of Aspergillus section Flavi (4.5 µm) compared to the clinical isolates of Aspergillus section Fumigati (2.7 µm).Additionally, a higher virulence in G. mellonella larvae was observed for Aspergillus section Flavi clinical isolates.The higher virulence of A. flavus was previously described in G. mellonella and other experimental models as mice [44,45].The experimental procedures possibly justify the high virulence of Aspergillus section Flavi clinical isolates.The artificial inoculation of G. mellonella larvae disrupted the initial physical barriers, easing these larger conidia's entry.Once inside the host, the conidia of the clinical isolates of both Aspergillus sections Fumigati and Flavi had an equal environment to express their pathogenicity.However, despite the previous knowledge of the higher virulence [44], little is known about the specific virulence factors of Aspergillus section Flavi that cause the increased virulence compared with Aspergillus section Fumigati.The virulence factors among Aspergillus spp.are multifactorial.Thermotolerance, adhesins, biofilm production, toxic metabolites, and hydrolytic enzyme secretion allow for the fungal cells to overcome the host defenses and contribute to the fungal pathogenesis [46].Thus, we investigated the biofilm formation capacity of Aspergillus sections Fumigati and Flavi clinical isolates.The clinical isolates were assigned to the three groups of biofilm producers (low, medium, and high).The comparative analysis with the virulence profile of each clinical isolate in G. mellonella larvae indicated no relationship with biofilm production.Therefore, the fungal invasion and evasion of the host immune response have a multifactorial nature, which modulates the Aspergillus spp.virulence and pathogenicity in high eukaryote organisms.
A crucial concern with the positive diagnosis of Aspergillus spp.infection (e.g., chronic invasive aspergillosis) is chronicity, difficulty in fungal eradication, and high patient mortality rate.It is partly related to emerging resistance to antifungals frequently used in clinical medicine.The resistance of several species to the available commercial antifungals shall be considered when treating the infection [2,47].Thus, the susceptibility of Aspergillus sections Fumigati and Flavi clinical isolates to antifungals was studied.Amphotericin B is widely used in hospitals as a salvage and last-line drug to treat severe and urgent cases of triazole-resistant Aspergillus spp.infections [19,20].In our sampling, we observed a majority of Aspergillus spp.clinical isolates with high MIC of amphotericin B. Although little is known about the worldwide susceptibility of Aspergillus spp. to AMB, up to now, the identical incidence of high AMB MIC has been described in Korea, Canada, and Brazil.Nonetheless, the reasons behind the emergence of high AMB resistance rates in these two geographic populations are still unknown [21,[23][24][25][26].A study in a southern Brazilian region showed a low incidence of high AMB MIC for A. fumigatus and A. flavus [48], which indicates a different geographical distribution pattern in Brazil.Thus, the high AMB MIC results for Aspergillus sections Fumigati and Flavi clinical isolates evince the inclusion of the region under study as geographically specific for this susceptibility pattern.Alarmingly, the result indicates a high probability of AMB therapeutic failure.Therefore, the physicians of the region under study should carefully consider the prescription of AMB for Aspergillus infection treatment.
Triazole antifungals are the first-line recommendation therapy for superficial and subcutaneous mycoses, fungal infections in hospitals, and long-term therapy [20,49].Resistance has been reported on all continents, although the frequency of this resistance remains unknown in many regions [7,[50][51][52].Furthermore, long-term azole therapy and unrestricted use of azole compounds in the environment as fungicides in the agricultural and horticultural industry have been identified as risk factors for developing azole resistance [7,53].The MIC of the most recommended triazoles used in the medical field, VOR, POS, and ITR, were evaluated against the Aspergillus sections Fumigati and Flavi clinical isolates of this study.A high presence of non-WT Aspergillus sections Fumigati and Flavi clinical isolates for POS was found, and it differs in European and Asian countries, which have a higher proportion of susceptible isolates [52,[54][55][56].In contrast, non-WT clinical isolates for ITR were uncommon in our region, which differs from the higher prevalence in Europe and southern Brazil [48,57].Voriconazole is highly recommended and the first choice for treatments of aspergillosis [6,47].Our study found a low rate of clinical isolates with a non-WT profile for VOR.Therefore, we demonstrate the utility of the existing strategy for treating Aspergillus infection.
To better understand the high triazole MIC among our clinical isolates, the sequence of the promoter region and cyp51A locus of all A. fumigatus s.s.clinical isolates were investigated.This gene encodes the 14-α-demethylase enzyme critical in the ergosterol biosynthesis pathway [58].Polymorphisms in the azole-binding amino acids that do not compromise Cyp51's activity can reduce the enzyme's affinity for these medications.Such polymorphisms can disrupt the stability and functionality of the enzyme, making it challenging to recognize substrates and ultimately resulting in various patterns of azole resistance.Point mutations in the cyp51 gene are extensively documented in isolates worldwide.Nonetheless, only a few studies have described these mutations within an exclusive Brazilian cohort [16][17][18]26,59].Contrary to the hotspot substitutions G54, P216, M220, G138, and G448 already described as directly related to triazole resistance [16], our study identified the substitution of the F46Y, M172V, N248T, N248K, D255E, and E427K amino acids in the Cyp51A protein of A. fumigatus s.s.clinical isolates.It is important to note that most clinical isolates presented high triazole MICs, as previously described for clinical isolates with the same amino acid substitutions [16][17][18].The same amino acid substitutions, except for N248K, were previously described in Brazilian clinical isolates [26], which indicates that they are common in our region.The N248K amino acid substitution in Cyp51A protein of A. fumigatus s.s. has already been related to azole resistance.However, unlike the single substitution identified in our A. fumigatus s.s.clinical isolates, N248K causes triazole resistance when associated with the V436A Cyp51A amino acid substitution [59].
In summary, the two main Aspergillus sections, Fumigati and Flavi, were isolated from Brazilian patients.A higher virulence of Aspergillus section Flavi clinical isolates was detected in the G. mellonella larvae infection model.Most Aspergillus spp.clinical isolates presented a high amphotericin B MIC, warning of a higher probability of therapeutic failure in the region under study.Furthermore, the amino acid substitutions found in Cyp51A of A. fumigatus s.s.clinical isolates are related to the increase in the MIC of azoles.Even so, most clinical isolates presented a WT profile for itraconazole and voriconazole, supporting them as the first therapeutic choice in controlling Aspergillus infections.Here, we emphasize the need for continuous surveillance of fungal infections in the hospital environment, which aids in overcoming the knowledge gap in the global fungal burden of infections and antifungal resistance, supporting public health interventions.

Figure 1 .
Figure 1.Conidia size of Aspergillus sections Fumigati and Flavi.Statistical analysis comparing the conidia diameter (µm) of Aspergillus sections Fumigati (n = 40) and Flavi (n = 24) clinical isolates (p < 0.0001).The bars indicate the mean and standard deviation of conidia diameter (µm).An unpaired t-test comparing the mean values of the two groups was performed.The p-value was performed with a significance level of 5% (p < 0.05).

Figure 2 .
Figure 2. Biofilm biomass and matrix of Aspergillus sections Fumigati and Flavi clinical isolates.(A) Pearson correlation between absorbance values of biofilm biomass × matrix of 64 Aspergillus spp.clinical isolates and statistical analysis comparing the (B) biofilm biomass, p < 0.803 and (C) biofilm matrix, p < 0.054 of Aspergillus sections Fumigati (n = 40) and Flavi (n = 24) clinical isolates.The bars indicate the mean and standard deviation.The p-value was performed with a significance level of 5% (p < 0.05) by the Mann-Whitney test and is indicated in the respective graphs.

Figure 2 .
Figure 2. Biofilm biomass and matrix of Aspergillus sections Fumigati and Flavi clinical isolates.(A) Pearson correlation between absorbance values of biofilm biomass × matrix of 64 Aspergillus spp.clinical isolates and statistical analysis comparing the (B) biofilm biomass, p < 0.803 and (C) biofilm matrix, p < 0.054 of Aspergillus sections Fumigati (n = 40) and Flavi (n = 24) clinical isolates.The bars indicate the mean and standard deviation.The p-value was performed with a significance level of 5% (p < 0.05) by the Mann-Whitney test and is indicated in the respective graphs.

Figure 3 .
Figure 3. Survival of Galleria mellonella larvae artificially infected with conidia of Aspergillus sections Fumigati and Flavi clinical isolates.Aspergillus section Fumigati (A) high; (B) medium; and (C) low virulence, Aspergillus section Flavi (D) high; and (E) low virulence.Ten larvae were included in each group, including untouched larvae (naïve), PBS 1× control inoculation, and larvae inoculated with fungal conidia.The survival of the larvae was accompanied daily up to 10 days.PBS, phosphatebuffered saline; p < 0.05.

Figure 3 .
Figure 3. Survival of Galleria mellonella larvae artificially infected with conidia of Aspergillus sections Fumigati and Flavi clinical isolates.Aspergillus section Fumigati (A) high; (B) medium; and (C) low virulence, Aspergillus section Flavi (D) high; and (E) low virulence.Ten larvae were included in each group, including untouched larvae (naïve), PBS 1× control inoculation, and larvae inoculated with fungal conidia.The survival of the larvae was accompanied daily up to 10 days.PBS, phosphatebuffered saline; p < 0.05.

Table 1 .
Minimal Inhibitory Concentration (MIC) of antifungal tested against Aspergillus sections Fumigati and Flavi clinical isolates.GM, geometric mean; MIC, minimal inhibitory concentration; MIC50 and MIC90, the lowest antifungal concentration able to inhibit the growth of 50% and 90% of all strains, respectively, of the tested clinical isolates.

Table 2 .
Minimal Inhibitory Concentration (MIC) of antifungals and Cyp51A amino acid substitutions of A. fumigatus s.s.clinical isolates.