Sterol Composition of Clinically Relevant Mucorales and Changes Resulting from Posaconazole Treatment

Mucorales are fungi with increasing importance in the clinics. Infections take a rapidly progressive course resulting in high mortality rates. The ergosterol biosynthesis pathway and sterol composition are of interest, since they are targeted by currently applied antifungal drugs. Nevertheless, Mucorales often exhibit resistance to these drugs, resulting in therapeutic failure. Here, sterol patterns of six clinically relevant Mucorales (Lichtheimia corymbifera, Lichtheimia ramosa, Mucor circinelloides, Rhizomucor pusillus, Rhizopus arrhizus, and Rhizopus microsporus) were analysed in a targeted metabolomics fashion after derivatization by gas chromatography-mass spectrometry. Additionally, the effect of posaconazole (POS) treatment on the sterol pattern of R. arrhizus was evaluated. Overall, fifteen different sterols were detected with species dependent variations in the total and relative sterol amount. Sterol analysis from R. arrhizus hyphae confronted with sublethal concentrations of posaconazole revealed the accumulation of 14-methylergosta-8,24-diene-3,6-diol, which is a toxic sterol that was previously only detected in yeasts. Sterol content and composition were further compared to the well-characterized pathogenic mold Aspergillus fumigatus. This work contributes to a better understanding of the ergosterol biosynthesis pathway of Mucorales, which is essential to improve antifungal efficacy, the identification of targets for novel drug design, and to investigate the combinatorial effects of drugs targeting this pathway.


Introduction
The order Mucorales represent the most prominent order of zygospore-forming fungi, which was formerly placed in the phylum Zygomycota and was referred to as Zygomycetes. Over the years, this phylum has undergone constant taxonomic rearrangements. Now, the Mucorales, or mucormycetes, are placed in the phylum Glomeromycota and subphylum Mucormycotina [1][2][3]. Mucorales are mainly saprophytes commonly found in soil and decomposing material. Members of this group play various important roles for human life-e.g., Mucor spp., are of biotechnological importance due to their high growth rates and their great potential in the production of secondary metabolites [4]. Other Mucorales, such as Rhizopus spp., have been used for food fermentation in Asia for centuries [5]. Contrary to these positive effects of mucoralean fungi on human life, they also cause a wide range Mucorales that were isolated in high abundance from patient material belong to the following genera: Rhizopus, Lichtheimia (formerly Absidia), Mucor, Rhizomucor, and Cunninghamella. Worldwide, Rhizopus arrhizus is the most common species that was isolated from clinical specimen. The clinical presentation of mucormycosis varies, with mostly rhinocerebral, pulmonary and gastrointestinal manifestations. Primarily, immunocompromised patients are at risk to develop mucormycosis, but some forms, such as cutaneous mucormycosis, have also been seen in otherwise healthy persons [1,7]. The aggressive course of the disease, delayed diagnosis, and poor treatment options result in unacceptably high mortality rates (40-70%), even with antifungal therapy. Often, mucormycosis is seen as so called breakthrough infection when voriconazole is applied as prophylactic treatment regime [6]. Antifungal drugs currently licensed for treatment of mucormycosis are the polyene antifungal amphotericin B (AMB), preferably the liposomal formulation as first line treatment, and the triazole posaconazole as salvage treatment. In 2015, another azole, isavuconazole (ISA), was approved by the Food and Drug Administration (FDA) as a stand-alone treatment for mucormycosis [1,[8][9][10][11][12][13]. AMB forms 1:1 adducts with fungal ergosterol and induces an accumulation of reactive oxygen species in the cytoplasm [14,15]. Azoles conduct their antifungal activity by inhibiting the enzyme sterol C14-demethylase (named ERG11 in yeasts, and CYP51 in Aspergillus spp., Figure 1, enzyme A) [16,17]. POS is also very active against Aspergillus isolates and used in the therapy of invasive aspergillosis. In contrast, another second-generation triazole, voriconazole (VRC), was shown to have no in vitro nor in vivo activity against Mucormycetes [13,15,[18][19][20]. The limited arsenal of highly active anti-mucoralean drugs, plus the occurrence of AMB and azole resistant strains highlight the importance of illuminating the mode of action of antifungal drugs on, and resistance mechanisms in this group of fungi [13,15]. As the target of the currently used antifungal drugs against Mucorales is ergosterol or its biosynthesis, respectively, it is of great importance to understand the biosynthetic pathway and decipher the differences to other human pathogenic fungi. For two major human pathogenic fungi, the yeast Candida albicans, and the mold Aspergillus fumigatus, differences in the ergosterol biosynthesis pathways have already been elucidated. In Candida spp., and this is similar to baker´s yeast S. cerevisiae (Figure 1, upper panel), lanosterol is the preferred substrate for the sterol C14-demethylase ( Figure 1, enzyme A), which converts lanosterol into 4,4-dimethylcholesta-8,14,24-trien-3β-ol [21][22][23][24][25]. In molds, e.g., Aspergillus fumigatus, the preferred ergosterol biosynthesis route from lanosterol starts with a methylation at C-24 by sterol C24-methyltransferase (enzyme B) in order to give eburicol, and is then followed by a demethylation at C-14 by sterol C14-demethylase to give 4,4-dimethylergosta-8,14,24(28)-trien-3β-ol [24][25][26] (Figure  1, lower panel).
In the presence of azoles, the accumulation of physiological substrates for sterol C14-demethylase and a depletion of ergosterol are observed in both yeasts and molds [16,25,[27][28][29][30]. The accumulation of C14-methylated sterols (e.g., lanosterol (10) and eburicol (13)) and lower levels of ergosterol as the predominant native sterol in fungi or changes in the relative sterol composition results in alterations of the plasma membrane that impact its function and the activity of membrane-bound enzymes [31][32][33].
So far, very little is known about the ergosterol biosynthesis pathway, the sterol content, and the composition of clinically relevant Mucorales, and most studies were carried out with one species or one strain only, which makes the comparison in-between the group of Mucorales difficult [26,[34][35][36][37][38][39]. Therefore, a full understanding of the ergosterol biosynthesis pathway and the sterol pattern of this group of pathogenic fungi is essential to determine the efficacy of antifungal therapy, to identify targets for novel drug design, and to investigate the combinatorial effects of drugs targeting the ergosterol biosynthesis pathway. Here, the sterol pattern and sterol content of six clinically relevant Mucorales (Rhizopus (R.) arrhizus, Rhizopus microsporus Lichtheimia (L.) corymbifera, Lichtheimia ramosa, Mucor (M.) circinelloides, and Rhizomucor (Rh.) pusillus) were analysed in a targeted metabolomics fashion by gas chromatography-mass spectrometry (GC-MS). In addition, sterols in hyphae of R. arrhizus that was confronted with sub-lethal concentrations of posaconazole were analysed and compared to untreated hyphae. R. arrhizus was chosen for this test because it represents the most abundant Mucorales species from patient material worldwide. Furthermore, the obtained data were compared to the well-studied ergosterol biosynthetic pathway of Aspergillus fumigatus [16,24,25,27,30,[40][41][42][43][44][45][46].

Relative and Absolute Amounts of Sterol Intermediates in Mucorales
In total, 14 different sterols were detected (Table 1)    As expected, ergosta-5,7,22-trien-3β-ol (ergosterol, 2) was identified as the dominating sterol in all the species studied. This is in agreement with a study that was carried out with R. arrhizus [39], in which ergosterol was also identified as the sterol exhibiting highest abundance. This finding is supported by a study that was investigating the phylogenetic distribution of fungal sterols [39]. One contrary result was shown by Weete et al. [37], explaining that ergosta-7,22-dien-3β-ol (3) is the major sterol (56.0 to 59.9%) and ergosterol (2) is only the second most abundant sterol (21.1-28.4%) in R. arrhizus. Importantly, the relative content of ergosterol was highly variable between the species, ranging from 38.0% in Rh. pusillus to 76.3% in R. arrhizus. In A. fumigatus, the relative amount of ergosterol reached 95.0% of all the sterols in our experimental setup. This amount is slightly higher  Table 1. X-axis presents retention time. Selected ions for chromatogram (A) m/z 363 + 366 + 365 + 407 + 408 and selected ions for chromatogram (B) m/z 343 + 377 + 379 + 393 + 472 + 486. Table 1. Sterol content and sterol composition of six different mucormycete species and A. fumigatus. Cultures were grown in RPMI 1640 medium over night at 37 • C before sterol extraction. The results are presented as the average of two independent experiments, comprising 6 technical replicates in total. Sterol composition is given as the relative amount of the respective sterol (%, in bold letters) of all the sterols detected. Sterol content is expressed as µg sterol intermediate/mg biomass (dry weight). Standard deviation is given in brackets. * n.d. = not detected, ** i.t. = in traces (< 0.04 µg/mg); major sterols are indicated in red.

Relative [%] and Absolute [µg/mg] Amount of Sterols in Each Species
No. IUPAC Name   Ergosta-5,8,22-trien-3β-ol (lichesterol, 1) was detected in all of the six species tested. Among the Mucormycetes, highest relative amounts of this intermediate were detected in Rhizopus spp., although these amounts were still lower than in A. fumigatus, which was included for comparison.
As expected, ergosta-5,7,22-trien-3β-ol (ergosterol, 2) was identified as the dominating sterol in all the species studied. This is in agreement with a study that was carried out with R. arrhizus [39], in which ergosterol was also identified as the sterol exhibiting highest abundance. This finding is supported by a study that was investigating the phylogenetic distribution of fungal sterols [39]. One contrary result was shown by Weete et al. [37], explaining that ergosta-7,22-dien-3β-ol (3) is the major sterol (56.0 to 59.9%) and ergosterol (2) is only the second most abundant sterol (21.1-28.4%) in R. arrhizus. Importantly, the relative content of ergosterol was highly variable between the species, ranging from 38.0% in Rh. pusillus to 76.3% in R. arrhizus. In A. fumigatus, the relative amount of ergosterol reached 95.0% of all the sterols in our experimental setup. This amount is slightly higher than what was shown by a comprehensive study of Alcazar-Fuoli et al. [24], where A. fumigatus isolates reached relative ergosterol amounts of 75.8 to 88.4%. Differences in experimental setup, like growth media and the age of the cultures used for sterol extraction, might explain these differences.
Most interestingly, a high amount of eburicol (4,4,14-trimethylergosta-8,24(28)-dien-3β-ol, 13) was detected to a degree that is usually only observed under azole treatment in fungi [25,29,41,43]. Levels reached up to 15.4% in Rh. pusillus, all other species also showed considerably high eburicol levels, ranging from 2.0% (L. corymbifera) to 7.2% (M. circinelloides). In our experimental setup, we did not detect eburicol in A. fumigatus, which is in contrast to the sterol profile of A. fumigatus in Alcazar-Fuoli et al. [24], where an eburicol content of approx. 1.9% was reported. Different growth conditions and different growth media might explain this discrepancy.
Additional to the relative amount of sterol intermediates, we also determined the actual amount of each sterol, which was expressed as µg/mg biomass, (dry weight, Table 1). The total sterol content was proportionately 1-2% of the total mycelial biomass, which is in agreement with what has been detected for other Mucor species before [36]. Lowest amount of total sterols was found in R. arrhizus (9.26 µg/mg) and highest in Rh. pusillus (21.39 µg/mg). The values of each intermediate determined correlate with the relative amounts that were previously described. In A. fumigatus, a total sterol content of 9.78 µg/mg was determined, of which 9.25 µg/mg was ergosterol, an amount that is comparable to previously reported data [24].

Sterol Composition and Sterol Content of Posaconazole Treated R. arrhizus
To investigate the effect of the sterol C14-demethylase inhibitor POS on the sterol composition and content of Mucorales, R. arrhizus hyphae confronted with sublethal concentrations (0.5 µg/mL) of POS were compared to the sterol pattern of untreated hyphae Figures 2 and 3, Table 2. For comparison and also to validate our results that were obtained with Mucorales, we included A. fumigatus, in which the effect of azole treatment on sterol pattern has been extensively studied [24,25,40,41,45].  For detailed information on all sterols extracted, see Table 2. For detailed information on all sterols extracted, see Table 2.   As expected, the relative amount of ergosterol (2) was significantly reduced by 17.8% in R. arrhizus due to POS treatment. On the other hand, eburicol (13), which is one of the substrates of the azoles' target enzyme C14-demethylase, increased by 25.4% (Figure 3), which underlines the enzyme inhibiting properties of POS also in R. arrhizus and it further indicates that 13 is the favored substrate for sterol C14-demethylase in Mucorales ( Figure 4).
As expected, the relative amount of ergosterol (2) was significantly reduced by 17.8% in R. arrhizus due to POS treatment. On the other hand, eburicol (13), which is one of the substrates of the azoles' target enzyme C14-demethylase, increased by 25.4% (Figure 3), which underlines the enzyme inhibiting properties of POS also in R. arrhizus and it further indicates that 13 is the favored substrate for sterol C14-demethylase in Mucorales (Figure 4). Surprisingly, the actual amount of ergosterol (2) relative to the biomass significantly increased in POS treated R. arrhizus hyphae (from 7.07 µg/mg to 7.52 µg/mg). Furthermore, a significant increase of the total sterol content (from 9.26 µg/mg to 12.86 µg/mg) was evident, which is most likely due to the accumulation of new sterols and the increased amount of ergosterol precursors, such as eburicol, which is increased eight-fold when compared to the untreated samples. This finding reflects what was shown by Weete and Wise [26] in propiconazole treated M. rouxii. However, this is contrary to our observations in A. fumigatus, where inhibition of sterol C14-demethylase resulted in ergosterol reduction (from 9.29 µg/mg to 7.41 µg/mg), and in this course resulted in a decrease of the total sterol content from 9.78 µg/mg to 8.81 µg/mg, although this decrease was not significant. Our results that were obtained for A. fumigatus correlate with the results of Alcazar-Fuoli et al. [27], showing that in strains with defective sterol C14-demethylase, caused by mutations in one of the two cyp51 genes, statistically lower amounts of total ergosterol were detected than in the wildtype strain.
The relative amounts of sterols 1, 3-6, 8, 9, 11, 12, 14 were not, or only to a minor extent, affected by POS treatment Table 2. Surprisingly, the actual amount of ergosterol (2) relative to the biomass significantly increased in POS treated R. arrhizus hyphae (from 7.07 µg/mg to 7.52 µg/mg). Furthermore, a significant increase of the total sterol content (from 9.26 µg/mg to 12.86 µg/mg) was evident, which is most likely due to the accumulation of new sterols and the increased amount of ergosterol precursors, such as eburicol, which is increased eight-fold when compared to the untreated samples. This finding reflects what was shown by Weete and Wise [26] in propiconazole treated M. rouxii. However, this is contrary to our observations in A. fumigatus, where inhibition of sterol C14-demethylase resulted in ergosterol reduction (from 9.29 µg/mg to 7.41 µg/mg), and in this course resulted in a decrease of the total sterol content from 9.78 µg/mg to 8.81 µg/mg, although this decrease was not significant. Our results that were obtained for A. fumigatus correlate with the results of Alcazar-Fuoli et al. [27], showing that in strains with defective sterol C14-demethylase, caused by mutations in one of the two cyp51 genes, statistically lower amounts of total ergosterol were detected than in the wildtype strain.
In conclusion, the azole activity in Mucorales results in significant alterations of the sterol composition, but it does not subsequently lead to a reduction of the total sterol content. The high amount of eburicol, even in untreated hyphae of the Mucorales, let us hypothesize that these fungi could be less affected by the accumulation of non-physiological intermediates due to azole treatment, which is reflected in their lower sensitivity to azole drugs when compared to other molds. In the genome of R. arrhizus two paralogous genes encoding for sterol C14-demethylase have been found [20]. It remains to be elucidated if the expression of both the genes is effected to a similar extent in the presence of azoles. So far, our results in R. arrhizus let us hypothesize that sterol C14-demethylase is only partly inhibited, as R. arrhizus is still able to synthesize ergosterol. This has been shown for other fungi as well, and it reflects that molds inherit a wide repertoire of adaption mechanisms to overcome such drawbacks. From the sterol pattern, we learn that in Mucorales the entry into the post-lanosterol pathway of ergosterol biosynthesis is similar to other molds, whereas under azole treatment, differences to molds, e.g., A. fumigatus were detected, resulting in the accumulation of a non-physiological sterol (14-methylergosta-8,24(28)-diene-3β,6α-diol, 15) which has so far only been found in azole treated yeasts ( Figure 4). Because the Mucorales exhibit high variability in their susceptibility to azoles, it is of special interest to decipher the effect of azoles on sterol composition and the content of other clinically relevant mucoralean fungi.

Fungal Strains Used in This Study
The following fungal strains, all being obtained from the strain collection of the Division of

Fungal Growth and Culture Conditions
All of the strains were cultivated on supplemented minimal agar (SUP) at 37 • C until sporulation (five days). Spores were obtained by harvesting them with sterile spore suspension buffer (0.9% NaCl, 0.01% Tween80). To obtain mycelia for sterol extraction, all of the strains were cultivated in RPMI 1640 cell culture medium (Sigmaaldrich, Vienna, Austria) at 37 • C with shaking overnight. Cultures of R. arrhizus and R. microsporus with optimal growth (fine mycelia, no formation of hyphal pellets) were harvested and transferred to a new shake flask containing RPMI 1640 plus 0.5 µg/mL posaconazole (POS). Cultures were further incubated in the presence of posaconazole for 4 h. Untreated controls were transferred into new medium without posaconazole. Then, the cultures were harvested by filtration, washed, and freeze dried to determine the fungal biomass on dry weight basis.

Sterol Extraction
Six mg of dry fungal biomass were used for sterol extraction as described by Müller et al. [25].
The amount of each sterol was expressed as µg per mg dry weight. The results represent the mean (µg/mg ± S.D.) of two independent biological replicates, including six technical parallels. Experiments involving POS treatment were carried out in triplicate. The sterol composition is expressed as relative amount of total sterols (%).