ABSTRACT
Azole antifungal resistance in Aspergillus fumigatus is a worldwide concern. As in most public hospitals in Brazil, antifungal susceptibility tests are not routinely performed for filamentous fungi at our institution. A 4-year retrospective azole antifungal resistance screening revealed two azole-resistant A. fumigatus clinical isolates carrying the CYP51A TR34 (34-bp tandem repeat)/L98H (change of L to H at position 98)/S297T/F495I resistance mechanism mutations, obtained from two unrelated patients. Broth microdilution antifungal susceptibility testing showed high MICs for itraconazole, posaconazole, and miconazole. Short tandem repeat (STR) typing analysis presented high levels of similarity between these two isolates and clinical isolates with the same mutations reported from the Netherlands, Denmark, and China, as well as environmental isolates from Taiwan. Our findings might indicate that active searching for resistant A. fumigatus is necessary. They also represent a concern considering that our hospital provides tertiary care assistance to immunocompromised patients who may be exposed to resistant environmental isolates. We also serve patients who receive prophylactic antifungal therapy or treatment for invasive fungal infections for years. In these two situations, isolates resistant to the antifungal in use may be selected within the patients themselves. We do not know the potential of this azole-resistant A. fumigatus strain to spread throughout our country. In this scenario, the impact on the epidemiology and use of antifungal drugs will significantly alter patient care, as in other parts of the world. In summary, this finding is an important contribution to alert hospital laboratories conducting routine microbiological testing to perform azole resistance surveillance and antifungal susceptibility tests of A. fumigatus isolates causing infection or colonization in patients at high risk for systemic aspergillosis.
INTRODUCTION
Aspergillus is a genus of fungi comprising over 300 different species distributed worldwide, Aspergillus fumigatus being the most commonly isolated species (1). Aspergillus infections demonstrate a wide variety of clinical manifestations, including allergic bronchopulmonary aspergillosis (ABPA), skin infections that result in cutaneous aspergillosis, aspergilloma, chronic pulmonary aspergillosis, and invasive aspergillosis (IA). Currently, IA is the main cause of morbidity and mortality in immunocompromised patients (2, 3). Although the first-choice antifungals for aspergillosis treatment are azoles, a wide variety of mutations in the CYP51A gene have been reported in azole-resistant isolates; the CYP51A gene encodes the cytochrome P450 14-α-lanosterol demethylase, the main target of azole antifungals (1, 4).
Many studies have reported the existence of resistance attributed to mutations in the CYP51A gene. Strains harboring these substitutions were reported in Europe, Asia, and the United States (5–8). In South America, such findings have already been reported from Colombia (2017), describing TR34 (34-bp tandem repeat)/L98H (change of L to H at position 98), TR46/Y121F/T289A, and TR53 mutations found in environmental isolates, from Argentina (2018), relative to a clinical isolate collected in 2009 carrying TR46/Y121F/T289A mutations, and from Peru (2017), in a clinical isolate carrying TR34/L98H mutations (9–11).
In Brazil, many researchers are studying the azole susceptibility profiles of A. fumigatus strains (12–14), and at present, there is one abstract from ISHAM 20th describing TR34/L98H mutations found in a clinical A. fumigatus isolate and a CYP51A gene point mutation resulting in an M220R mutation obtained from an isolate collected from a crop in southern Brazil (15).
The University of Campinas Clinical Hospital (UNICAMP) (https://www.hc.unicamp.br/node/76) is a reference tertiary hospital for the city of Campinas and a macroregion of 86 municipalities of São Paulo State, with about 6.5 million inhabitants. Also, patients from several other states, such as Minas Gerais, Paraná, Bahia, Rio de Janeiro, and Mato Grosso, are treated on an outpatient basis in almost all specialties and clinical and surgical subspecialties, including rarer and more complex ones. In addition to over 400 beds, more than 10,000 people circulate daily at our institution.
In this hospital, as in most public hospitals in Brazil, antifungal susceptibility tests are not routinely performed for filamentous fungi. A retrospective azole antifungal resistance screening revealed our first two azole-resistant A. fumigatus clinical isolates carrying the CYP51A TR34/L98H/S297T/F495I resistance mechanism mutations.
RESULTS
Azole-resistant A. fumigatus isolates were detected after retrospective azole resistance screening of a selection of 199 isolates maintained at the Laboratory of Fungal Investigation (LIF) collection of the Clinical Pathology Department, School of Medical Sciences, UNICAMP. The isolates were obtained from patients with different underlying diseases for routine microbiological diagnosis from 2014 to 2017. Only 2 isolates (LIF 2444-6 and LIF 2552-4.9) from 2 different patients (which represent 1% of the isolates and 4.5% of the patients) were able to grow at the highest concentration in the screening test for itraconazole and showed high MICs for almost all azoles after performing the broth microdilution antifungal susceptibility test (Table 1). A tandem duplication of a 34-bp sequence (5′-X-3′) was found between positions 163 and 230 of the start codon (Fig. 1A). In addition, sequence analysis of the CYP51A open reading frame (ORF) showed three substitutions, T293A, T889A, and T1483A, that resulted in amino acid changes L98H, S297T, and F495I (Fig. 1B). Short tandem repeat (STR) typing analysis for nine microsatellite markers, 2A, 2B, 2C, 3A, 3B, 3C, 4A, 4B, and 4C (16), showed high levels of similarity between the two isolates and clinical isolates with the same mutations reported from the Netherlands, Denmark, and China, as well as environmental isolates from Taiwan (Fig. 2) (16, 17).
In vitro antifungal susceptibilities of the 2 A. fumigatus isolates and CYP51A sequencing results
(A) Nucleotide sequence alignment showing TR34 repeat on isolates LIF 2444-6 and LIF 2552-4.9. (B) Protein sequence alignment.
Genotypic relationship between LIF 2444-6 and LIF 2552-4.9 and Aspergillus fumigatus isolates with TR34/L98H/S297T/F495I mutations from China, the Netherlands, Denmark, and Taiwan (16, 17). The dendrogram is based on a categorical analysis of nine microsatellite markers in combination with unweighted pair group method with arithmetic mean (UPGMA) clustering using Phyloviz 2.0a.
DISCUSSION
Isolate LIF 2444-6 was obtained from sputum in routine culture controls for cystic fibrosis patients on 28 July 2015, from a 24-year-old male patient with chronic lung disease, followed at the Pulmonology Outpatient Clinic. There is no hospitalization record for him and no data in his folder from 1999 to 2016.
A. fumigatus LIF 2552-4.9 was isolated from sputum in routine microbiological diagnosis culture on 2 December 2015. Clinical data available in the Clinical Laboratory were that the 54-year-old male patient was diabetic with bronchiectasis in the right lung. Samples collected on the same date had negative results in examination for tuberculosis, a positive result in direct examination for Paracoccidioides spp., and after standard incubation time, a positive culture for A. fumigatus.
According to routine laboratory procedures in 2015, identification was based only on macro- and micromorphology, no antifungal susceptibility test was performed, and the isolates were added to the Laboratory of Fungal Investigation (LIF) collection of the Clinical Pathology Department, School of Medical Sciences, University of Campinas.
In this paper, we report two azole-resistant A. fumigatus strains that harbor the CYP51A TR34/L98H/S297T/F495I mutations, detected in our institution after an azole resistance screening test. These mutations have already been described in Denmark, the Netherlands, China (16), and Taiwan (17) and were related to high MICs for itraconazole and imidazoles.
According to Chen et al. (16), it was possible to show that the F495I substitution is related to high MICs to imidazoles. It was possible to evaluate an imidazole (miconazole), which is already included in the commercial plate used (Eiken Chemical Co., Tokyo, Japan), with a MIC of >16 μg/ml, confirming the relationship of high MICs for this antifungal class and the F495I substitution.
According to STR analysis, our isolate, although it seems to be closer to the isolates of Denmark and the Netherlands, has its own pattern. Isolates with this type of mutation were mostly considered environmental. These isolates of A. fumigatus were considered by the attending clinicians as colonizing agents, and the patients received no treatment.
What stands out is the fact that the two microorganisms were isolated in the same year (2015), although 5 months apart. These patients live in different cities, and as neither patient was admitted during this period, the only common space attended by both was the outpatient clinic of the hospital.
As this is a retrospective study, the resistant microorganisms having been detected among the isolates from 2015, a limitation that cannot be solved is to perform related environmental sampling. But surveillance air collection was conducted by our study group in each of the four seasons during the whole of 2015 in two different environments in the hospital: the bone marrow transplant ward (with filtered air) and the hematology ward (with open windows, thus reflecting the air composition around the hospital) (18). Among the 123 Aspergillus isolates found were 62 A. fumigatus isolates, none of which showed resistance to azoles in the azole screening test. After CYP51A gene sequencing, one isolate showed the amino acid exchange N248K derived from a point mutation and 12 isolates showed F46Y, M172V, N248T, D255E, and E427K mutations, found not only in environmental isolates but also in isolates from patients preexposed to azole drugs (19).
Our findings might indicate that resistant A. fumigatus strains from our region may carry these or other mutations. Due to the protocol for collecting a single colony from each distinct morphology in routine diagnostic cultures, we may have missed the opportunity to detect other equally resistant isolates, whether carrying these mutations or not. This represents a concern, especially considering that our hospital provides tertiary care assistance to immunocompromised patients who may be exposed to resistant environmental isolates. Other patients receive prophylactic antifungal agents, and patients with invasive fungal infections receive antifungal therapy for years. In these two situations, isolates resistant to the antifungal in use may be selected within the patients themselves (1).
We do not know the potential of this azole-resistant A. fumigatus strain carrying the CYP51A mutations to spread throughout our country. In this scenario, the impacts on epidemiology and the use of antifungal drugs will significantly alter patient care, as in other parts of the world.
In summary, this description of the CYP51A mutations in our health care area is an important contribution to alert hospital laboratories conducting routine microbiological testing to perform azole resistance surveillance and antifungal susceptibility tests of A. fumigatus strains causing infection or colonization in patients at high risk for systemic aspergillosis.
MATERIALS AND METHODS
Aspergillus fumigatus isolates.One hundred ninety-nine sequential A. fumigatus isolates obtained from different clinical specimens from 44 patients with 2 or more isolates from January 2014 to March 2017 were selected. The age of the patients ranged from 10 to 71 years, considering the date of first isolation. The number of isolates ranged from 2 to 12 per patient, whose main underlying diseases were cystic fibrosis (78%), immunosuppression (9%), tuberculosis (2%), paracoccidioidomycosis (2%), diabetes mellitus (2%), and not available (2%). Only for 9 patients (20%) is there a record of the administration of itraconazole prior to fungal isolation. At the time of preparation of the manuscript, death was recorded for 20% of patients, which does not include the patients with resistant isolates. All isolates were stored in distilled water at room temperature using Castellani’s method (20). The macromorphology and micromorphology of each isolate were observed after pure growth on Sabouraud dextrose agar (SDA) and potato dextrose agar (PDA; Difco, Sparks, MD, USA).
Molecular identification.Genomic DNA was extracted after 48 to 72 h of growth at 25°C on SDA plates using a DNeasy tissue kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. Comparative DNA analyses of beta-tubulin (β-tubulin 2A/-B) sequences were performed for species confirmation (21). The resultant nucleotide sequences were analyzed using Geneious 8.1 (2015; Biomatters Ltd., Newark, NJ, USA) and compared with external databases.
Azole resistance screening.Mueller-Hinton agar plates (MH) with 2% dextrose were prepared by adding the antifungal dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) to reach a final concentration of 2 μg/ml for itraconazole and voriconazole and 0.5 μg/ml of posaconazole. (Sigma-Aldrich, St. Louis, MO, USA). After up to 3 days fungal growth at 35°C in PDA slants, inoculums were prepared to final concentrations of 5.5 × 104 to 2 × 105 CFU/ml. Amounts of 20 μl of inoculum suspensions were deposited on the agar plates containing the antifungal agents. For growth controls, the same inoculums were also deposited on plates without the antifungal agents. The tests were read after 48 h of incubation at 35°C.
Broth microdilution.The isolates capable of growth in the screening test were submitted to antifungal susceptibility tests according to the CLSI M38-A2 (22) protocol, using preprepared plates for itraconazole, voriconazole, and miconazole (Eiken Chemical Co., Tokyo, Japan) and plates prepared in-house for posaconazole (Sigma-Aldrich, Brazil). Aspergillus flavus strain ATCC 204304, Candida parapsilosis strain ATCC 22019, and Candida krusei strain ATCC 6258 were included as quality controls in each test.
Detection of CYP51A mutations.DNA was extracted from 48-h fungal cultures as described above. Oligonucleotides used for amplification and sequence analysis of CYP51A were AF1P1 (F/R), AF2P1 (F/R), AF3P1 (F/R), and AF4P1 (F/R). Sequences were aligned with an azole-susceptible strain sequence (GenBank accession no. AF338659) using Geneious 8.1 (2015; Biomatters Ltd., Newark, NJ, USA) (12).
Microsatellite genotyping.For genotyping, PCR was performed using 9 pairs of primers and sequencing ≈400 bp from the isolates as described previously. Repeat numbers of each region (2A, 2B, 2C, 3A, 3B, 3C, 4A, 4B, and 4C) of the sequence were counted (16).
Ethical considerations.The present study was approved by the local Ethics Committee (CAAE51794615.0.0000.5404).
ACKNOWLEDGMENTS
L. Pontes and C. A. G. Beraquet received scholarships financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES), finance code 001. This project was supported by grants from the collaborative research project Science and Technology Research Partnership for Sustainable Development (SATREPS), Japan, and the University of Campinas, Brazil, no. 02-P-9427/2018.
L.P and A.Z.S. conceived and designed experiments. C.A.G.B., T.A., G.L.P., and L.L. contributed to the acquisition, analysis, and/or interpretation of data. L.P., C.A.G.B., and A.Z.S. drafted the manuscript, and A.W. and M.L.M. revised it critically for important intellectual content.
FOOTNOTES
- Received 20 October 2019.
- Returned for modification 15 December 2019.
- Accepted 20 December 2019.
- Accepted manuscript posted online 23 December 2019.
- Copyright © 2020 American Society for Microbiology.
