ABSTRACT
Recently, the species concept of opportunistic Mucor circinelloides and its relatives has been revised, resulting in the recognition of its classical formae as independent species and the description of new species. In this study, we used isolates of all clinically relevant Mucor species and performed susceptibility testing using the EUCAST reference method to identify potential species-specific susceptibility patterns. In vitro susceptibility profiles of 101 mucoralean strains belonging to the genus Mucor (72), the closely related species Cokeromyces recurvatus (3), Rhizopus (12), Lichtheimia (10), and Rhizomucor (4) to six antifungals (amphotericin B, natamycin, terbinafine, isavuconazole, itraconazole, and posaconazole) were determined. The most active drug for all Mucorales was amphotericin B. Antifungal susceptibility profiles of pathogenic Mucor species were specific for isavuconazole, itraconazole, and posaconazole. The species formerly united in M. circinelloides showed clear differences in their antifungal susceptibilities. Cokeromyces recurvatus, Mucor ardhlaengiktus, Mucor lusitanicus (M. circinelloides f. lusitanicus), and Mucor ramosissimus exhibited high MICs to all azoles tested. Mucor indicus presented high MICs for isavuconazole and posaconazole, and Mucor amphibiorum and Mucor irregularis showed high MICs for isavuconazole. MIC values of Mucor spp. for posaconazole, isavuconazole, and itraconazole were high compared to those for Rhizopus and the Lichtheimiaceae (Lichtheimia and Rhizomucor). Molecular identification combined with in vitro susceptibility testing is recommended for Mucor species, especially if azoles are applied in treatment.
INTRODUCTION
Mucormycosis is a rare but severe fungal infection caused by members of the order Mucorales. Rhizopus, Lichtheimia, Mucor, and Apophysomyces are the main genera potentially causing mucormycoses (1). In patients without underlying diseases, Mucor species usually cause cutaneous or subcutaneous infections (2–4) or severe localized infections related to extensive burns (5, 6) or deep trauma (6, 7). Systemic infections caused by species of the Mucor circinelloides complex and Mucor indicus have been reported in patients with impaired immunity, often as a result of hematologic malignancies (8–13) or uncontrolled diabetes (14). Intestinal infections are mainly caused by M. indicus (3, 11, 15, 16), but M. circinelloides (17) and Mucor velutinosus (18) have also been isolated from blood cultures of patients with severe intestinal dysfunctions. Food contaminated with M. circinelloides has been reported to cause gastrointestinal disorders (19). In contrast to other Mucor species, Mucor irregularis, formerly known as Rhizomucor variabilis var. variabilis, causes chronic cutaneous and subcutaneous infections mainly in Asia. Usually, exposed parts of the body are affected mostly due to small trauma such as bites or lesions caused by plant material (4, 20). The patients are apparently immunocompetent but may have hidden immune disorders such as CARD9 deficiency (4, 21).
Treatment of mucormycosis is difficult because of the intrinsic resistance of Mucorales against echinocandins (22, 23) and voriconazole (24–26). In contrast to classical views in the literature, the response against the remaining antifungal drugs is not homogeneous within the Mucorales. Specific and generic differences in susceptibility to amphotericin B, azoles, and terbinafine have been reported (27–30). To date, no breakpoints have been defined for any member of the Mucorales, but epidemiological cutoff values exist for Mucor circinelloides and amphotericin B and posaconazole (31).
For mucoralean genera with recently updated taxonomy, such as Rhizopus and Lichtheimia (32–34), antifungal susceptibility profiles of the opportunistic members have been published (28, 29, 35). In Mucor, currently twelve species are known to cause infections in humans and animals (6), but comprehensive antifungal susceptibility (AFS) data only exist for M. circinelloides, M. indicus, and M. ramosissimus (23, 28, 31, 36). Mucor circinelloides and its relatives were recently taxonomically revised resulting in the recognition of its classical formae as independent species (M. circinelloides sensu stricto , M. griseocynanus, M. janssenii, and M. lusitanicus), the acceptance of M. velutinosus as a discrete species (37, 38), and the description of several new Mucor species (6).
The aim of the present study was to provide AFS profiles for opportunistic Mucor species on the basis of the revised taxonomy and compare AFS profiles of Mucor species with other mucoralean genera of medial importance. Therefore, we studied the MICs of 101 strains of 18 species against six antifungals according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) protocol (39). We included Mucor species that have been reported to cause superficial as well as invasive infections on the basis of a molecular identification (4, 6, 37, 38, 40). We also studied Mucor amphibiorum that is only known to infect frogs, toads, and platypus (41) and Cokeromyces recurvatus because this species is related to Mucor (40) and has been isolated from clinical sources as well (42–44). Mucor hiemalis was not included, although it has been reported from clinical sources, because these cases most likely refer to Mucor irregularis (syn. Rhizomucor variabilis) (20), as M. hiemalis does not grow at temperatures higher than 30°C (45). All case reports of M. hiemalis are morphology based only, published before Mucor irregularis was described (37), and mostly describe chronic cutaneous infections of healthy individuals, which is in agreement with the typical clinical picture of M. irregularis infections (20). To compare the antifungal profiles between Mucor species and the remaining main mucoralean opportunists, strains of Rhizopus, Lichtheimia, and Rhizomucor were also studied.
RESULTS
Molecular species identification based on internal transcribed spacer (ITS) sequences (see Table S1 in the supplemental material) was unambiguous for all strains studied. ITS sequences of strains that were not identified in former taxonomic studies (6, 40) showed ITS sequence similarities of ≥99% with the respective ex-type strains.
The results of the AFS testing are shown in Table 1. The most active drug for Mucor spp. and Cokeromyces recurvatus was amphotericin B, with a maximum MIC value of 0.5 mg/liter. In the genus Rhizopus, amphotericin B MIC values of 1 mg/liter (R. microsporus) and 2 mg/liter (R. arrhizus) were found. Similar results were obtained for natamycin, the second polyene in our study with relatively high MIC values. Natamycin MIC values were also elevated in Rhizopus.
Ranges of antifungal susceptibilities, geometric means, and MIC50s of the species studied
Large differences were found for the susceptibility against terbinafine. While nearly all Mucor species and Cokeromyces showed high MICs for terbinafine, members of the Lichtheimiaceae showed low MICs for this drug. For Rhizopus, the results were species dependent, R. microsporus having low and R. arrhizus high MIC values. For azoles, species-specific AFS profiles were determined, including four species, Cokeromyces recurvatus, Mucor ardhlaengiktus, M. lusitanicus, and M. ramosissimus, with high MIC values for all three azoles. The MIC values for isavuconazole were distinctly higher for the Mucoraceae (Mucor and Cokeromyces) (geometric mean [GM], 6.59 mg/liter) than for Rhizopus (GM, 1.37 mg/liter) and the Lichtheimiaceae (Lichtheimia and Rhizomucor) (GM, 2 mg/liter). Several Mucor species, namely, M. amphibiorum, M. ardhlaengiktus, M. indicus, M. irregularis, and M. lusitanicus, had a MIC50 value (MIC at which 50% of isolates are inhibited) of >8 mg/liter for isavuconazole. Large differences between species of the test set were noted with itraconazole. MIC values of the Lichtheimiaceae (GM, 0.57 mg/liter) were lower than those of the Mucoraceae (GM, 3.57 mg/liter) and Rhizopus (GM, 2 mg/liter). Only in species of the Mucoraceae, itraconazole MIC values of 4 mg/liter and higher were found. Compared to those of other antifungals, posaconazole MIC values were very variable among strains of the same species. In the Mucoraceae, eight of thirteen species included strains with posaconazole MIC values of ≥2 mg/liter, while posaconazole MICs of this level were absent from Lichtheimiaceae and only once found in Rhizopus. The great majority of strains of Cokeromyces recurvatus, M. ardhlaengiktus, M. indicus, M. lusitanicus, and M. ramosissimus showed high MICs for posaconazole.
DISCUSSION
The main results of our study are the species-specific AFS profiles of opportunistic Mucor species based on the revised taxonomy, which further support the recent taxonomic revision (6) and the differences in AFS profiles of Mucor spp. compared to those of Lichtheimia and Rhizopus. Our study reveals that some of the clinically relevant Mucor species show high in vitro MICs to isavuconazole, itraconazole, and/or posaconazole, while the clinically relevant Rhizopus and Lichtheimia species present low MICs for all these antifungals. Mucor circinelloides sensu stricto, M. griseocyanus, M. janssenii, M. lusitanicus, and M. velutinosus, which were treated as formae of M. circinelloides sensu lato in the past (6), exhibited different AFS profiles. Mucor lusitanicus showed high MIC values for all azoles tested. It is a limitation of this study that some species were only represented by a low number of strains. However, for species such as M. amphibiorum, M. ardhlaengiktus, and M. variicolumellatus , only a low number of strains are available worldwide, and for other species (especially M. ramosissimus, M. janssenii, and M. griseocyanus), the number of incorrectly identified strains is high; consequently, it was impossible to test high numbers of strains for each species. The MIC ranges of these species should be interpreted with care because they may not cover the entire variability of these species.
AFS data in some previous studies that included Mucor species other than M. circinelloides already suggested that AFS profiles differed between Mucor species, but often only a single strain per species was studied (28, 31, 46). Espinel-Ingroff et al. (31) studied susceptibility, applying several strains of M. circinelloides sensu lato, M. indicus, and M. ramosissimus against posaconazole using the CLSI protocol with reading after 24 h. Although our tests were performed according to the EUCAST protocol, our data for posaconazole against M. indicus (MIC range and MIC50 of 0.25 to >8 mg/liter and 1 mg/liter, respectively) were similar to those obtained by Espinel-Ingroff et al. (31) (0.25 to 8 mg/liter and 0.5 mg/liter, respectively). Larger deviations were encountered in M. circinelloides: Espinel-Ingroff et al. (31) found a MIC range of 0.06 to 16 mg/liter and a MIC50 of 1 mg/liter, while we determined 0.125 to 4 mg/liter and 0.5 mg/liter, respectively. One possible reason for this difference could be that the concept of M. circinelloides of Espinel-Ingroff et al. (31) differed from the revised species concept of Wagner et al. (6) and may have included, e.g., M. lusitanicus, which shows high MICs in vitro to posaconazole. In this case, the epidemiologic cutoff values provided by Espinel-Ingroff et al. (31) for M. circinelloides need to be recalculated. Another reason could be the different protocols used (CLSI versus EUCAST). Two studies (36, 46) addressed the effect of EUCAST versus CLSI protocols on susceptibility in Mucorales and found essential agreement, although EUCAST MICs were generally somewhat higher for triazoles (87% for posaconazole, 98.3% for isavuconazole [46]). Arendrup et al. (36) found only 50% agreement for posaconazole and 25% to 67% for isavuconazole in M. circinelloides sensu stricto. Based on the reference sequences given by Arendrup et al. (36), the strains used in the study belong to M. circinelloides sensu stricto. Therefore, CLSI MICs of the studied species are expected to be lower than the EUCAST MICs.
Antifungals that lack in vitro activity may nevertheless lead to successful treatment (47). The lower posaconazole susceptibility of Rhizopus spp. than of Lichtheimia spp. appears to be method dependent, because these differences were not detected when tested by CLSI (36).
In agreement with our results, Arendrup et al. (36) found isavuconazole MICs to be in general 1 to 3 steps higher than those for posaconazole, which is compensated by the higher exposure at standard dosing in clinical settings. The authors also reported high isavuconazole MICs in the range between 1 and 16 mg/liter for M. circinelloides sensu stricto. With M. amphibiorum, M. ardhlaengiktus, M. indicus, M. irregularis, and M. lusitanicus, other species with high isavuconazole MICs were detected in this study.
Arendrup et al. (36) found a higher susceptibility of Rhizopus spp. for isavuconazole than of Lichtheimia, but only when tested with CLSI methodology, and consequently, they considered these differences as method dependent. Our results, obtained applying the EUCAST protocol, also revealed Rhizopus to present slightly lower MICs for isavuconazole than Lichtheimia, suggesting that this difference exists but is more pronounced if the CLSI methodology is applied. MIC values against terbinafine for the Lichtheimiaceae and Rhizopus microsporus found by Vitale et al. (28) were lower than those in our study, which might be due to the different protocols used for AFS testing (CLSI/EUCAST) and differences in incubation temperatures for Mucor spp. (28°C for all species/35°C for most species). Judging from our results, appropriate molecular species identification is essential when treatment by azoles rather than amphotericin B is installed. In addition, AFS testing is recommended for species with high intraspecific variability, such as M. circinelloides sensu stricto.
In agreement with earlier studies (23, 25, 27, 28, 36, 46), we found amphotericin B to be the most active drug against opportunistic species of Mucor. Natamycin, another polyene, is only topically applied because it is not absorbed from the gastrointestinal tract (48). The MICs obtained for natamycin are relatively high, but high levels of this drug can be reached in cornea tissue. Lalitha et al. (49) considered Aspergillus and Fusarium isolates with MICs of ≤16 mg/liter to be susceptible to natamycin, because adequate levels are reached in the eye during standard therapy of keratitis.
The four tested strains of Mucor racemosus were unable to grow at 32°C in liquid RPMI medium. In addition, the only reports of this species based on a molecular identification refer to infections of the nails and the skin (40). Considering its morphological similarity to M. circinelloides, there is reasonable doubt about the potential of this species to cause invasive infections. The different incubation temperatures for 2 species and different incubation times for 3 species are a limitation of our study, because the obtained MIC values are not fully comparable.
In conclusion, opportunistic Mucor species differed in their azole susceptibility profiles. Mucor circinelloides sensu stricto, M. griseocyanus, M. janssenii, M. lusitanicus, and M. velutinosus that are closely akin to M. circinelloides sensu lato (6) also varied in AFS profiles. Cokeromyces recurvatus, M. ardhlaengiktus, M. lusitanicus, and M. ramosissimus exhibited high MIC values for all azoles tested. Mucor indicus showed high MICs for isavuconazole and posaconazole, and M. amphibiorum and M. irregularis presented high MICs for isavuconazole. In general, MIC values of Mucor spp. for posaconazole, isavuconazole, and itraconazole tend to be higher than those of Rhizopus and the Lichtheimiaceae. The most active drug for all mucoralean species was amphotericin B.
MATERIALS AND METHODS
Strains.Studied strains originated from the reference collection of the Westerdijk Fungal Biodiversity Centre (CBS; Utrecht, The Netherlands), the Instituto de Salud Carlos III National Centre of Microbiology (CNM-CM; Madrid, Spain), the National Reference Center for Invasive Fungal Infections (NRZMyk; Jena, Germany), the Jena Microbial Resource Collection (JMRC; Jena, Germany), or the Belgian Co-ordinated Collections of Microorganisms (IHEM; Brussels, Belgium). A total number of 101 strains of 18 species (see Table S1 in the supplemental material) were studied belonging to three different families: (1) Mucoraceae: Cokeromyces recurvatus (3), Mucor amphibiorum (3), M. ardhlaengiktus (syn. M. ellipsoideus) (2), M. circinelloides (14), M. griseocyanus (4), M. indicus (10), M. janssenii (5), M. lusitanicus (13), M. racemosus f. racemosus (4), M. ramosissimus (1), M. variicolumellatus (2), and M. velutinosus (7); (ii) Rhizopodaceae: Rhizopus arrhizus (6) and R. microsporus (6); and (iii) Lichtheimiaceae: Lichtheimia corymbifera (8), L. ramosa (2), and Rhizomucor pusillus (5). Strains originated from clinical or environmental sources (Table S1).
Species identification.All strains were identified to the species level based on ITS sequences. Corresponding GenBank accession numbers are given in Table S1. Species identification of most strains was part of former taxonomic studies (6, 40). The methods for the extraction of DNA, PCR amplification, and sequencing of the remaining strains were described by Wagner et al. (6).
Cultivation and antifungal susceptibility testing.Depending on growth speed and sporulation, the strains were grown for 3 to 5 days at room temperature on malt extract agar (MEA): 40 g/liter malt extract (Becton, Dickinson, Heidelberg, Germany), 4 g/liter yeast extract (Ohly, Hamburg, Germany), 1.5% Europäischer agar (Otto Nordwald, Hamburg, Germany).
In vitro antifungal susceptibilities were determined by broth microdilution technique according to the EUCAST standard methodology with slight modifications for few isolates (39). Pure compounds of known potency of the following antifungals were used: amphotericin B (AMB; European Pharmacopoeia, Strasbourg, France), natamycin (NAT; ChemicalPoint, Deisenhofen, Germany), terbinafine (TRB; Novartis Pharma AG, Cork, Ireland), itraconazole (ITC; Janssen-Cilag GmbH, Neuss, Germany), posaconazole (POS; Merck & Co., Inc., Kenilworth, NJ, USA), and isavuconazole (ISA; Basilea Pharmaceutica International Ltd., Basel, Switzerland). The antifungals were tested in the following concentration ranges: 0.03 to 16 mg/liter for AMB, 0.016 to 8 mg/liter for ISA, ITC, NAT, and POS, and 0.06 to 32 mg/liter for TRB. Microplates containing each antifungal drug in one row were prepared by batch and stored frozen at −80°C for up to 6 months. Spore suspensions were counted with a hemocytometer. The final inoculum ranged between 1 × 105 and 2.5 × 105 spores/ml. With few exceptions, MIC values were determined visually using a mirror after 24 h of incubation at 35°C and defined as a 100% reduction in growth. For slower growing species such as M. ramosissimus and Cokeromyces recurvatus, MIC values were determined after 48 h. Strains of M. amphibiorum needed to be incubated at 32°C, and MIC values were also read after 48 h. The incubation temperature for M. racemosus had to be reduced to 28°C. For the sake of consistency, MIC values of Mucor ardhlaengiktus and Cokeromyces recurvatus were also determined visually, although they grew as yeasts in liquid RPMI medium of the AFS test plate. For Mucor ardhlaengiktus, this is the first report of a yeast state. Two reference strains, Aspergillus fumigatus ATCC 204305 and Candida parapsilosis ATCC 22019, which were recommended by EUCAST (39) for antifungal susceptibility testing using amphotericin B, itraconazole, and posaconazole, were included as quality controls in each set of tests. To allow the calculation of geometric means, high off-scale MICs were raised to the next higher concentration.
ACKNOWLEDGMENTS
We thank Carmen Karkowski and Christiane Weigel for excellent technical assistance during this study. We also thank the companies that kindly provided drugs for the susceptibility testing, namely, Basilea, Merck & Co., Inc. (Kenilworth, NJ, USA), Novartis Pharma AG, and Pfizer, Inc.
This project was funded by the Deutsche Forschungsgemeinschaft (DFG grant number WA 3518/1-1). The work of the NRZMyk is supported by the Robert Koch Institute from funds provided by the German Ministry of Health (grant 1369-240).
FOOTNOTES
- Received 27 March 2019.
- Returned for modification 19 May 2019.
- Accepted 5 June 2019.
- Accepted manuscript posted online 10 June 2019.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00653-19.
- Copyright © 2019 American Society for Microbiology.
