Medical Mycology, 2017, 0, 1–14 doi: 10.1093/mmy/myx081 Advance Access Publication Date: 0 2017 Original Article Original Article Transcriptional profile of the human skin pathogenic fungus Mucor irregularis in response to low oxygen Wenqi Xu1 , Jingwen Peng1 , Dongmei Li2 , Clement K. M. Tsui3 , Zhimin Long4 , Qiong Wang1 , Huan Mei1 and Weida Liu1,∗ 1 Department of Mycology, Institute of Dermatology, Chinese Academy of Medical Sciences and Peking Union Medical College, Nanjing 210042, Jiangsu, People’s Republic of China, 2 Department of Microbiology and Immunology, Georgetown University Medical Center, Washington, DC 20057, USA, 3 Division of Infectious Diseases, University of British Columbia, Vancouver, BC V6H 3Z6, Canada and 4 Demo Lab, Shanghai AB Sciex Analytical Instrument Trading Co., Ltd, IBP, Shanghai, 200335, People’s Republic of China ∗ To whom correspondence should be addressed. Dr. Weida Liu, No. 12 Jiangwangmiao Street, Nangjing 210042, Jiangsu, People’s Republic of China. Tel: +86 25 85470580; Fax: +86 25 85414477; E-mail: email@example.com Received 18 January 2017; Revised 28 April 2017; Accepted 25 August 2017; Editorial Decision 10 May 2017 Abstract Mucormycosis is one of the most invasive mycosis and has caused global concern in public health. Cutaneous mucormycosis caused by Mucor irregularis (formerly Rhizomucor variabilis) is an emerging disease in China. To survive in the human body, M. irregularis must overcome the hypoxic (low oxygen) host microenvironment. However, the exact molecular mechanism of its pathogenicity and adaptation to low oxygen stress environment is relatively unexplored. In this study, we used Illumina HiSeq technology (RNA-Seq) to determine and compare the transcriptome profile of M. irregularis CBS103.93 under normal growth condition and hypoxic stress. Our analyses demonstrated a series of genes involved in TCA, glyoxylate cycle, pentose phosphate pathway, and GABA shunt were down-regulated under hypoxic condition, while certain genes in the lipid/fatty acid metabolism and endocytosis were up-regulated, indicating that lipid metabolism was more active under hypoxia. Comparing the data with other important human pathogenic fungi such as Aspergillus spp., we found that the gene expression pattern and metabolism in responses to hypoxia in M. irregularis were unique and different. We proposed that these metabolic changes can represent a species-specific hypoxic adaptation in M. irregularis, and we hypothesized that M. irregularis could use the intralipid pool and lipid secreted in the infection region, as an extracellular nutrient source to support its hypoxic growth. Characterizing the significant differential gene expression in this species could be beneficial to uncover their role in hypoxia adaptation and fungal C The Author 2017. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org 1 2 Medical Mycology, 2017, Vol. 00, No. 00 pathogenesis and further facilitate the development of novel targets in disease diagnosis and treatment against mucormycosis. Key words: Mucor irregularis, cutaneous mucormycosis, hypoxic response, lipid/fatty acid metabolism, endocytosis. Introduction Mucormycosis, a fungal disease typically occurs in sinuses, lungs and epidermal tissues, is spreading rapidly and exhibiting high rates of morbidity and mortality. The incident rate of mucormycosis has increased by 7.4% per year (from 0.7 to 1.2 cases/million persons) in the last decade.1 Mucormycosis is most commonly caused by members of Mucorales. This fungal infection can be transmitted by spores in the air, ingestion, or direct contact with injured skin.2,3 Based on clinical presentations, mucormycosis can be divided into two types—invasive and cutaneous mucormycosis. The former is a severe life-threating fungal infection, commonly prevalent in individuals with impaired immunity, while the latter has mostly a milder condition which can manifest at even immunocompetent patients. Mucor irregularis (renamed from Rhizomucor variabilis4 ) was first isolated from a skin lesion, which had been presented as a primary cutaneous infection in the hand of a Chinese patient in 1991.5 Since then, approximately 30 cases of primary cutaneous infection caused by M. irregularis have been documented,6–8 of which 23 cases were from China. In contrast to the angioinvasive mucormycosis (commonly caused by Rhizopus oryzae), infection caused by M. irregularis presents as a chronic disease, tending to be limited to dermal and subcutaneous tissues without vascular invasion. Most patients with M. irregularis infections were immunocompetent or at least had no apparent immunodeficiency, but some patients were badly disfigured due to misdiagnoses in early stage.9–13 M. irregularis is a unique pathogen in its temperature tolerance among the Mucorales. In general, Mucorales spp. are highly thermotolerant or even thermophilic, with maximum growth temperatures up to 37–45◦ C. However, M. irregularis cannot grow above 37◦ C, which may be one of the major reasons for its inability to cause deep-tissue infections.14 Therefore, this pathogen may have special genetic traits for environmental adaptation that differ from other Mucorales species for high temperature tolerance. To date, most studies on mucormycosis have focused on species that cause invasive infections,12,13,15 but the knowledge for disease management remains limited, especially the pathogenesis/pathophysiology of M. irregularis is poorly understood. To survive in a human host, the pathogenic fungi need to tolerate and overcome in vivo micro-environmental stress conditions. One of the stresses is hypoxia (low oxygen) condition. It is well established that the oxygen levels in most human tissues are considerably below atmospheric level (20.9%); the oxygen concentration varies across different tissues in the body, and ranging from 2.5% in the kidney to 9% in the lung. Furthermore, oxygen partial pressure of the skin is only 41 mmHg (∼6% oxygen concentration).16,17 Since oxygen is a critical component to many essential biochemical processes, the ability to survive under hypoxic conditions has been hypothesized to be a necessary virulence attribute of human pathogenic fungi.18–21 To understand how human pathogenic fungi adapt and survive in low oxygen conditions, several investigations have examined the global fungal transcriptome responses to hypoxia in Aspergillus nidulans, A. fumigatus, Candida albicans, and Cryptococcus neoformans.22–25 The results demonstrated that, pathogenic fungi possess different mechanisms to maintain energy in order to survive and grow in oxygen-limited environments. However, some transcriptome changes and gene regulation patterns are common and can be found across multiple species, such as the up-regulation of genes involved in glycolysis, steroid, and secondary metabolite metabolism, as well as the down-regulation of genes responsible for ribosomal and purine/pyrimidine biosynthesis.22–25 In contrast to these invasive pathogens, far less is known about hypoxia responses of M. irregularis, which is confined to causing skin lesions. For better understanding of its adaptation in reduced oxygen level condition and its consequential significance in the pathogenicity of mucormycosis during skin infection, we performed transcriptome sequencing (RNA-seq) to examine the transcriptional responses of M. irregularis to hypoxia. The RNA-Seq data of M. irregularis (CBS103.93) under atmospheric environment was compared to the one under 6% O2 concentration. The latter was used to mimic the skin oxygen concentration.16,17 The main goal of the study is to provide an overview of genes that are up- or down- regulated by low level of oxygen stress, particularly on genes involved in pathogenesis and hypoxia pathway that may distinguish it from other clinical important pathogenic fungi. Methods Growth comparison under normoxic and hypoxic atmospheres Mucor irregularis standard strain CBS103.93 was maintained at the National Fungi Strain Reserve Center (Nanjing, China). Since the fungus causes skin infection, it was Xu et al. cultured on solid medium instead of liquid broth, which can be used for blood or blood vessel infection fungus. The fungus was initially incubated on Malt Extract Agar (OXOID, Basingstoke, UK) at 27◦ C for 5 days in normoxic condition (20.9% O2 ), then the mycelia was homogenized and filtered through filter paper with average pore size of 40 µm. The concentration of spores (conidia) was determined with a hemocytometer under a microscope, and 5 µl conidial suspensions adjusted to 1 × 103 conidia/µl were spotted onto the center of the plates. Plates were pre-incubated at 27◦ C under normoxic condition for 12 hours, then either kept normoxic (20.9% O2 ) or shifted to a hypoxia chamber (HuaXi Elctronics Technetronic, Changsha, China) with 6% O2 and incubated until the mycelia covered the entire cultivation dish. The diameter of the colony was measured and averaged from two separate experiments. To compare the growth of M. irregularis on medium with or without triglycerides in normoxic and hypoxic conditions, 5 µl conidial suspensions were spotted onto the center of the plates containing 70% MEA or 70% MEA supplemented with 7.5% (v/v) triglycerides, respectively, followed by incubation at 27◦ C in a hypoxic atmosphere (6% O2 ) for 3 days. RNA extraction and RNA-Seq library sequencing Six plates of M. irregularis were pre-incubated for 3 days at 27◦ C under normoxic condition. Of these, three plates were then incubated in hypoxic condition (6% O2 ), while the remaining three plates were still kept under normoxic condition. After 6 hours of incubation, all the repeat plates were collected, and the mycelium was ground to fine powder for total RNA isolation using Qiagen RNeasy Plant Mini kit (Qiagen, Hilden, Germany). Total RNA concentration was quantified with an Ultrospec 2100 Pro (Amersham Pharmacia, Little Chalfont, England). Messenger RNA (mRNA) was purified by polyA selection method using oligo(dT) beads and was separated in fragmentation buffer to elute 100-bp to 400-bp fragments. Total mRNAs were then reverse-transcribed into complementary DNAs (cDNAs) for library construction. RNA-seq transcriptome library was prepared according to a TruSeqTM RNA sample preparation kit from Illumina Technology (San Diego, California) using 5 µg of RNA. Afterward, double-stranded cDNA was synthesized using a SuperScript double-stranded cDNA synthesis kit (Invitrogen, Carlsbad, California) with random hexamer primers (Illumina, San Diego, California). Then the synthesized cDNA was subjected to end-repair, phosphorylation and ‘A’ base addition according to Illumina’s library construction protocol. Libraries were size selected for tar- 3 geting fragments of 200–300 bp on 2% Low Range Ultra Agarose followed by 15 cycles of polymerase chain reaction (PCR) amplification using Phusion DNA polymerase (NEB, Ipswich, Massachusetts). After quantification by TBS-380 mini-fluorometer (Promega, Madison, Wisconsin), the paired-end RNA-seq sequencing library was sequenced by Illumina HiSeq platform (2 × 150 bp read length) at Biozeron Biotech Company (Shanghai, China). The sequencing of M. irregularis under normoxia and hypoxia were performed, respectively. Transcriptome data processing and assembly After Illumina sequencing, the raw reads were in FASTQ format. The adapter sequences and reads of poor quality were trimmed (ambiguous bases and quality value ≤5). Reads obtained for M. irregularis under normoxia and hypoxia conditions were assembled de novo, respectively by Trinity (http://trinityrnaseq.sourceforge.net/) using the default parameters (–min_contig length 200 –min kmer cov 1 –max reads per graph 20000 –group pairs distance 500 –path reinforcement distance 75), which had been reported in other publications.26,27 The contigs and unigenes of less than 200 bp were discarded due to low annotation rate.27,28 The raw data of M. irregularis under normoxia and hypoxia were deposited in NCBI as BioProject PRJNA347489, in which SRA483033, SRA483036, and SRA483037 were derived from normoxia, and SRA483176, SRA483180, and SRA483179 from hypoxic condition. The assembled sequencing data were deposited in the NCBI-TSR database (TSR: GFBC00000000). Functional annotation Functional annotations were performed by sequence comparison to public databases including the NCBI nonredundant protein database (NR database, http://www.ncbi.nlm.nih.gov), Swiss-Prot database (http://www.expasy.ch/sprot), clusters of orthologous groups for eukaryotic complete genomes database (KOG) (ftp://ftp.ncbi.nih.gov/pub/COG/KOG/kyva), and Kyoto encyclopedia of genes and genomes (KEGG) pathway database (http://www.genome.jp/kegg/) using BLASTX alignment with an E-value of 1.0E−5 , respectively. In addition, Blast2GO program and BlastX29 were also used to perform GO annotation of unigenes, and then WEGO30 software was used to perform GO classification, assigning biological processes, molecular functions, and cellular components. 4 Medical Mycology, 2017, Vol. 00, No. 00 Identification of differentially expressed genes To better interpret the expression levels of each unigene, FPKM (Fragments Per kb per Million reads)31 was used to eliminate the influence of differences of gene lengths and sequencing depth. Then, the adjusted expression level can be used for direct comparison of differences between samples. The false discovery rate (FDR) method was also introduced to determine the threshold P-value in multiple tests using Cufflink software package. An FDR ≤ 0.05 was used as the threshold to determine the significance of gene expression differences between the two tested temperatures.32 GO functional enrichment and KEGG pathway analysis were carried out by Goatools (https://github.com/tanghaibao/Goatools) and KOBAS (http://kobas.cbi.pku.edu.cn/home.do).33 Confirmation of gene transcripts using quantitative RT-PCR The total RNA used for RNA-Seq was also employed to verify the expression levels of selected target genes. Firststrand cDNA was synthesized from 2 µg DNase-treated R Q RT SuperMix for qPCR total RNA using the HiScript (+gDNA wiper) (Vazyme, Nanjing, China). Quantitative real time polymerase chain reaction (qPCR) amplifications in 20 µl volumes, containing 10 µl SYBR Green Master Mix (Vazyme, Nanjing, China), 20 ng template, and 4 pmol of each primer, were performed using the Stratagene Mx3000p Real-Time PCR instrument (Agilent Technologies, Hansen, Connecticut). The temperature profile was 95◦ C for 5 min, 40 cycles of 95◦ C for 15 s, 55◦ C for 25 s, and 72◦ C for 25 s. The reaction was conducted in triplicate. The threshold values for each target gene were normalized using the glyceraldehyde-3-phosphate dehydrogenase gene (GAPDH). The relative expression was estimated by employing the 2−CT method.34 Seven predicted genes involved in glycolysis/gluconeogenesis, GABA shun, glycerolipid metabolism, fatty acid metabolism pathways were selected according to the RNA-seq data. The RTqPCR primers were designed with MacVector 11 (Accelrys, San Diego, California) (Table S1). Results Comparison of growth status under hypxia and normoxia The effect of hypoxia on the growth of pathogenic fungus M. irregularis was tested under 6% O2 conditions in an oxygen-controlled chamber, which was mimicking the level of oxygen in human skin. After incubation for 3 days, the Figure 1. Growth of Mucor irregularis CBS103.93 in normoxic and hypoxic culture. A total of 5 × 103 conidia was spot inoculated onto MEA plates initially incubated in air at 27◦ C. After 12 h, plates were either kept normoxic (20.9% O2 ) or shifted to a hypoxic condition (6% O2 ) and incubated for a further 3 days. B. The diameters of the colonies were measured over 78 h and are expressed in cm. Values represent the mean of three biological replicates. Bar = 2 cm. sizes of M. irregularis colonies on MEA plates were always smaller than those kept in the normoxic incubator (Fig. 1A). The average of colony diameters for M. irregularis mycelia under hypoxia condition was 5.3 ± 0.2 cm after 72 hours, while the colonies were 7.6 ± 0.3 cm in diameter under normoxic growth condition (Fig. 1B). Transcriptomes of M. irregularis under normoxia and hypoxia To better understand the transcriptomic profile of M. irregularis in low oxygen condition, total RNAs were extracted from two conditions in triplicates. A total of 126 and 205 million reads were generated from the hypoxic and normoxic samples, respectively (Table S2), from which variable sequences with Q30 bases > 92 % and G+C% about 44 % were assigned to 30,761 consensus unigenes (Table S2). The mean length of these unigenes was 1247 bp, with N50 of 1761bp (Table S3). The variance among the three biological replicates from normoxic and hypoxic conditions was evaluated using the scatter plots of gene expression (Fig. S1). The correlation among the three Xu et al. 5 replicates in each tested condition was significantly higher than the results between the two conditions for any particular given replicate, comfirming the data were reliable for downstream analysis. Functional annotation of the unigenes Figure 2. Sequence similarity and species distribution of the top BLASTx hits against the NR database for each unigene. A total of 26,693 (86.76%) genes had positive hits at least once in the NR, the Swiss-Prot protein, KEGG or KOG database, in which 85.4% of the sequences were homologous to the gene sequences listed in the NR database when the e-value frequency distribution has fixed significant hits as 1.0E−60 (Fig. 2A). However, the blast hits against SwissProt, KEGG and KOG databases only resulted in 58.30% (17,933 transcripts), 38.73% (11,915 transcripts) and 8.44% (2597 transcripts) of positive hits from genomic sequences, respectively (Table S3). The closest related species to M. irregularis in the NR database was Rhizopus delemar with a 36.98% identity between the two species, and only 6802 (22.11%) unigenes could be annotated, by searching against the NR database (Fig. 2B), while the remaining unigenes were either hypothetical genes or functionally uncharacterized. Based on the NR annotation and gene ontology classification, 21085 unigenes were assigned with GO terms. The GO-annotated unigenes were distributed in 45 Figure 3. GO annotations of nonredundant consensus sequences. Best hits were aligned to the GO database, and 21085 unigenes were assigned to at least one GO term. Most consensus sequences were grouped into three major functional categories, namely biological process, cellular component, and molecular function. 6 Medical Mycology, 2017, Vol. 00, No. 00 Figure 4. Histogram of clusters of orthologous groups (KOG) classification. All unigenes were aligned to KOG database for prediction and classification based on possible functions. categories in terms of biological processes, cellular components, and molecular functions clusters (Fig. 3, Table S4). For the biological processes category, the genes related to metabolic processes (7585, accounting for 35.97%) were dominant, followed by those related to cellular processes (6617, 31.38%), then single-organism processes (3744, 17.76%) and biological regulation (2218, 10.52%) (Fig. 3). Among the cellular components category, cell part and cell (both 4594, 21.79%) were the dominant groups, followed by organelles (2555, 12.12%) and membrane (2335, 11.07%) (Fig. 3). In terms of molecular functions, 48.35% (10195) of the unigenes was assigned to catalytic activity, followed by in descending order 42.86% for binding (9038), 5.94% for transporter activity (1253), and 3.11% for molecular function regulator (655) (Fig. 3). Based on the KOG database, the putative proteins were functionally classified into 25 molecular families such as cellular structure, biochemistry metabolism, molecular processing, and signal transduction (Fig. 4). Within these broad categories, we found a few notable groups, such as translation, ribosomal structure, and biogenesis (accounting for 334, 10.14%), followed by posttranslational modification, protein turnover and chaperones at 10.00% (329), signal transduction mechanisms at 7.69% (253), and transcription bringing up the rear at 5.65% (186). Apart from the largest group of functionally uncharacterized genes (337, 10.24%), the remaining genes were involved either in nuclear structure, cell motility, extracellular structure, or defense mechanisms (5, 3, 3, and 1 unigenes, respectively) (Fig. 4). To further explore the molecular interaction between genes, the KEGG database was also used to predicate the potential pathways in which these genes might be involved. Among the 30761 annotated genes, 11,915 were clustered to 34 processes/pathways, including signal transduction, translation, carbohydrate metabolism, endocrine functions, and other biosynthetic pathways (data not shown). Differentially expressed unigenes The differentiation of gene expression between normoxia and hypoxia was evaluated by FPKM with FDR corrections (P < .05); genes showing at least twofold changes in expression were considered to be differentially expressed. A total of 1112 transcripts (Table S5) had altered expression significantly in response to different oxygen conditions, of which 531 transcripts were significantly up-regulated and 581 transcripts were down-regulated in response to hypoxia (twofold changes cut off, Table S5). The up-regulated genes were associated with the regulation of gene expression (GO:0010468), regulation of metabolic process (GO:0080090, GO:0019222, and GO:0031323), antioxidant activity (GO:0051920, GO:0016209, and GO:0016684), transcription factor activity (GO:0003700, GO:0006355, and GO:0001071), energy metabolism related process (GO:0060590, GO:0004090, and GO:0003959) and mitochondrion (GO:0044429) (Table 1). The down-regulated genes were involved in hydrolases activity (GO:0016798, GO:0004553, and GO:0016811), defense response (GO:0006952 and GO:0009605), carbohydrate metabolism (GO:0030246, GO:0005975, GO:0016810, GO:0015926, GO:0004410, and GO:0004339) and calcium ion binding processes (GO:0005509) (Table 1). The biological functions of the 1112 differentially expressed genes (DEGs) were also evaluated in the KEGG Xu et al. 7 Table 1. Significant GO terms in response to hypoxia. GO term counts GO IDs Terms <Up-regulated in hypoxia> GO:0003700 transcription factor activity, sequence-specific DNA binding GO:0004090 carbonyl reductase (NADPH) activity GO:0006355 regulation of transcription, DNA-templated GO:0010468 regulation of gene expression GO:0016209 antioxidant activity GO:0016684 oxidoreductase activity, acting on peroxide as acceptor GO:0019222 regulation of metabolic process GO:0031323 regulation of cellular metabolic process GO:0044429 mitochondrial part GO:0051920 peroxiredoxin activity GO:0060590 ATPase regulator activity GO:0080090 regulation of primary metabolic process <Down-regulated in hypoxia> GO:0004339 glucan 1,4-alpha-glucosidase activity GO:0004410 homocitrate synthase activity GO:0004553 hydrolase activity, hydrolyzing O-glycosyl compounds GO:0005509 calcium ion binding GO:0005975 carbohydrate metabolic process GO:0006952 defense response GO:0009605 response to external stimulus GO:0015926 glucosidase activity GO:0016798 hydrolase activity, acting on glycosyl bonds GO:0016810 hydrolase activity, acting on carbon-nitrogen (but not peptide) bonds GO:0016811 hydrolase activity, acting on carbon-nitrogen (but not peptide) bonds, in linear amides GO:0030246 carbohydrate binding database by pathway enrichment analysis. The pathways that were significantly enriched included starch and sucrose metabolism (ko00500), glycerolipid metabolism (ko00561), glycerophospholipid metabolism (ko00564), pyruvate metabolism (ko00620), amino sugar and nucleotide sugar metabolism (ko00520), oxidative phosphorylation (ko00190), and carbon metabolism (ko04141). Also some functional categories such as carbon metabolism, fatty acid/lipid metabolism and endocytosis were observed during comparison (Fig. 5). Many of the findings reported in M. irregularis (Fig. 6, Table S6) were different from that reported in other human pathogenic fungi (Table 2). For example, the genes involved in glycolysis, fatty acid metabolism, oxidative phosphorylation, steroid biosynthesis, and pentose phosphate pathway were often found to be up-regulated in other fungi such as Cryptococcus neoformans, Candida albicans, and Aspergillus nidulans under hypoxic conditions.35–39 Observed Annotated Expected P-value 23 615 10.62 .000743 2 45 45 5 4 54 50 7 3 2 52 7 1241 1207 73 48 1825 1575 176 17 6 1591 0.12 21.42 20.84 1.26 0.83 31.50 27.19 3.04 0.29 0.10 27.46 .0059 3.29E−06 1.43E−06 .00863 .00938 1.25E−04 3.76E−05 .0337 .0029 .00426 1.08E−05 6 2 23 12 22 6 4 8 24 11 81 5 427 240 596 37 21 120 438 232 1.53 0.09 8.06 4.53 11.26 0.70 0.40 2.27 8.27 4.38 .00434 .00343 8.21E−06 2.18E−03 3.26E−03 6.25E−05 5.84E−04 2.04E−03 3.93E−06 .00488 5 61 1.15 .00589 6 62 1.17 1.11E−03 Reduced ability to degrade the carbohydrate under hypoxia In M. irregularis many transcripts involved in glycolysis were reduced in response to hypoxia. For example, acetyl-coenzyme A synthetase (Acs, GFBC01001446), one of the most highly reduced transcripts, two fructosebisphosphatase aldolases (FbaA, GFBC01011669 and GFBC01021420), which are involved into the early steps of converting fructose 1,6-phosphate to glyceraldehyde 3-phosphate, pyruvate kinase (Pyk, GFBC01029099) along with aldehyde dehydrogenases (Aldh, GFBC01005625), which converts acetaldehyde to acetate for central carbohydrate and lipid metabolism tended to be transcriptionally reduced (Fig. 5, 6). Of the seven genes involved in starch and sucrose metabolism, six decreased in the transcriptional levels after a shift to hypoxic growth condition. Somewhat 8 Medical Mycology, 2017, Vol. 00, No. 00 Figure 5. Hypoxia decreased levels of expression for genes involved in carbon metabolism, oxidative phosphorylation, and increased the expression of genes involved in fatty acid/lipid metabolism and endocytosis processes. Heat map showing the cluster analysis of genes differentially expressed prior to hypoxia (0 hour) and in hypoxia (6 hours) of Mucor irregulars CBS103.93. surprisingly, only one exception, the trehalose 6-phosphate phosphatase (TPS, GFBC01001384), which catalyzes the dephosphorylation of trehalose 6-phosphate to form trehalose, was up-regulated. Altered expression in steroid biosynthesis and GABA shunt pathway genes In response to hypoxia, the steroid biosynthesis pathway of M. irregularis was not significantly affected, only two sterol reductase genes egr4 (GFBC01007963, GFBC01007636) and a cytochrome P450 (GFBC01017808) reduced transcription, while the transcription level of sterol regulatory element-binding protein SREBP was not significantly altered. These results suggested that the steroid biosynthesis of M. irregularis appeared to not be affected or partially decreased in response to hypoxia, which was incongruent to the patterns of many other fungi (Table 2). In addition, seven differentially expressed genes were identified to be involved in the GABA shunt of M. irregularis, including two glutamate decarboxylase (Gad, GFBC01008155 and GFBC01027882), one 4aminobutyrate transaminase (GatA, GFBC01030508), and four vesicular inhibitory amino acid transporter (VGAT). We found that the expression of Gad and GatA was reduced, but VGATs were up-regulated in response to hypoxia. In contrast to the fungi such as Aspergillus nidulans and C. neoformans in which the genes of the GABA shunt were up-regulated for energy metabolism, the inconsistency of these genes alternation elicited that the GABA shunt was inactive in M. irregularis under low oxygen (Table 2). Bolstered catabolic potential in lipid metabolism and activation in endocytosis The hypoxic exposure also led to marked changes in gene expression involved in lipid metabolism. We found Xu et al. 9 Figure 6. Overview of metabolic responses of Mucor irregularis in hypoxia. The figure showed the major changes in M. irregularis metabolism during hypoxia for 6 h. Up-regulated genes are highlighted in black, while down-regulated genes are colored in gray. Genes are abbreviated in capital letters: PYK, pyruvate kinase; FbaA, Fructose-bisphosphatase aldolase; ACS, Acetyl-coenzyme A synthetase; ALDH, aldehyde dehydrogenase; ADH, alcohol dehydrogenase (NADP+); PckA, phosphoenolpyruvate carboxykinase; GAD, glutamate decarboxylase; ACOX, acyl-CoA oxidase]; ACSL, long chain fatty acyl-CoA synthetas; TAG lipase, Triacylglycerol lipase. Table 2. Comparison of metabolic pathways (predicted by KEGG) involved in hypoxic adaptation in Mucor irregularis and six other fungi. Glycolysis Gluconeogenesis (PEP carboxykinase) Pentose phosphate pathway Oxidative phosphorylation β-oxidation Lipid/Fatty acid metabolism GABA shunt/ergosterol biosynthesis Endocytosis M. irregularis A. nidulans A. niger A. oryzae A. fumigatus C. albicans C. neoformans ↓ ↑ ↓ ↓ ↑ ↑ ↓ ↑ ↑ ↓ ↑ NR ↑ ↑ ↑ NR ↑ NR ↑ NR ↑ ↑ ↑ NR ↑ ↑ NR NR NR NR ↑ NR ↑ NR ↑ ↓ NR ↓ ↑ NR ↑ NR ↑ ↓ ↑ ↑ ↑ NR − NR NR NR ↑ ↑ ↑ NR ↑, up-regulated; ↓, down- regulated; −, a general lack of changes; NR, Not reported; References: Aspergillus nidulans,25 ,37,65 Aspergillus niger,54 Aspergillus oryzae,25 Aspergillus fumigatus,18 ,19,48 Candida albicans,24 ,39 Cryptococcus neoformans.23 ,36 that in M. irregulairs, the triacylglycerol (TAG) lipase (GFBC01029180) that degrades triacylglycerol into fatty acid, and a member of the cytosolic phospholipase A2 group IV family (GFBC01029364), which catalyzes phosphatidylcholine to acyl-glycero-3-phosphocholine for cellular energy, were up-regulated under hypoxia (Fig. 5, 6). In addition, both the long chain fatty acyl-CoA synthase (ACSL, GFBC01014149) and acyl-CoA oxidase (ACOX, GFBC01011826) of M. irregulairs, which were both involved in β-oxidation, were also up-regulated. These results led us to speculate that the hypoxic M. irregularis might use the intracellular lipid pool as energy source. Along with this speculation, M. irregularis would need an approach to maintain the lipid homeostasis under hypoxic growth conditions. Previous studies showed that endocytosis was important for bringing nutrients into the cell to maintain lipids and protein homeostasis in the cell;40,41 hence, analyses on endocytosis were conducted in M. irregularis. Eight genes associated with endocytosis were differentially expressed in hypoxia, of which five genes (GFBC01019239, GFBC01021838, GFBC01028777 [HSP71-like], GFBC01016923 [HSP70], and GFBC01029894 [HSP71-like]) were up-regulated under hypoxic condition. It was noteworthy that such 10 Figure 7. Gene expression for validation in (A) qRT-PCR and (B) RNAseq assays in normoxia and hypoxia exposed strains. The data represent the mean ± standard deviation from three biological replicates. ACS (GFBC01001446), ALDH (GFBC01005625), GAD (GFBC01008155), SREBP (GFBC01022405), TAG lipase (GFBC01029180), ACOX (GFBC01001611), ACSL (GFBC01014149). induction of endocytosis process including three HSP genes under hypoxia has not been reported in other pathogenic fungi such as A. nidulans (Table 2). Validation of RNA-seq findings with real-time PCR The RNA-seq results were validated using qRT-PCR by using the same biological RNA samples. Seven target genes from different functional categories that were verified as up- or down-regulated through qRT-PCR as an independent measure of differential gene expression (Fig. 7A). All the genes used in validation showed the same pattern of expression as that of the RNA-seq results (Fig. 7B) demonstrating the reliability of the RNA seq data. Discussion The comparison of mycelial diameter between normoxia and hypoxia clearly demonstrated that the reduced oxygen supply was a growth limiting factor in M. irregularis. Similarly, M. plumbeus, another member of Mucorales, also had slower growth in hypoxia than in atmospheric oxy- Medical Mycology, 2017, Vol. 00, No. 00 gen.42 Almost all clinical cases caused by M. irregularis were chronic cutaneous infection, except the case that reported a facial lesion with an extraordinary pulmonary infection.13 Xia et al. observed that the patient’s symptoms subsided in summer and were aggravated in winter, which was different from most superficial fungal infection in human, suggesting that different optimal temperatures for fungal growth may be responsible for this behavior. Apart from this explanation, since the oxygen content is the highest in the lung of human body, which could be a favorable condition for its lung infection and invasion rather than other inner organs. Thus, we hypothesized that the growth of M. irregularis was retarded during the invasion of human skin in hypoxic environment, having the ability to overcome the low oxygen condition is essential establishment of the skin infection. To understand the molecular mechanisms of hypoxia adaptation in this pathogenic fungi, we investigated the transcriptome profiles of M. irregulairs using Illumina RNA-seq. Many transcripts related to glycolysis were reduced in response to hypoxia such as Acs (GFBC01001446), FbaA (GFBC01011669 and GFBC01021420), Pyk (GFBC01029099), and Aldh (GFBC01005625). However, these glycolic genes were strongly induced in response to hypoxia in A. nidulans and C. albicans,37,39 which was contrary to the response of M. irregularis (Table 2). Furthermore, seven differentially expressed unigenes were identified in starch and sucrose metabolism, of which the expression levels of six genes were decreased after a shift to the hypoxic conditions, except TPS (GFBC01001384) which was up-regulated. Previous studies had shown that trehalose can be used as an alternative carbon source43 and can be metabolized during adaptive response to various stress conditions including dehydration, oxidative stress, heat, cold, and freezing stress in yeast and filamentous fungi.44–46 The elevated level of TPS might played an important role in regulating carbon utilization in respond to hypoxia stress for M. irregularis. It will be important to further verify the role of TPS in the infection mechanism. Decreased expression of genes involved in the TCA cycle and aerobic respiration has been demonstrated in other fungi such as C. albicans and fission yeast when exposed to hypoxia.39,47 We found that three NADH-quinone oxidoreductase (Ndh, GFBC01024053, GFBC01024863, GFBC01007772) were also down-regulated, thus, mitochondrial respiration chain complexes I in M. irregularis was diminished by limited oxygen supply. The underlying mechanism to maintain energy flow in M. irregularis during oxygen depletion could be different from other filamentous fungi, in which mitochondrial respiration for ATP production was active under hypoxia.48 Another finding was that Xu et al. gluconeogenesis in M. irregularis was up-regulated, demonstrated by increased levels of phosphoenolpyruvate carboxykinase (PckA, GFBC01011023) and transcription activator of gluconeogenesis ERT1 (GFBC01017450). PckA is a key enzyme in the reductive branch of the TCA cycle, which in other pathogens. For example, this enzyme is important for the re-oxidation of intracellular NADH during the hypoxic growth of Mycobacterium tuberculosis49 and is activated in A. oryzae upon hypoxia (Table 2). The induced transcripts of PckA and ERT1 suggested that M. irregularis might use the reductive branch of the TCA cycle to adapt to the hypoxia conditions. Taken together, these results suggested that energy generating through carbohydrate metabolism activity was likely decreased in response to hypoxia in M. irregularis, indicating that M. irregularis was evolved to have different metabolism properties from other fungi such as A. nidulans upon hypoxia. The steroid biosynthesis pathway has been identified as responsive to hypoxia in A. fumigatus, C. neoformans and C. albicans,18,50,51 and sterols were considered as an oxygen sensing system due to its high oxygen requirement for sterol biosynthesis. Transcripts of steroid biosynthesis enzymes such as the C-14 sterol reductases ERG24, and the C-4 methyl sterol oxidases ERG25, which has been shown to require oxygen, were highly induced under hypoxia in C. neoformans.51 However, the steroid biosynthesis pathway in M. irregularis was not significantly affected. Only two sterol reductase genes egr4 (GFBC01007963, GFBC01007636) and a cytochrome P450 (GFBC01017808) were seen to be reduced, while the transcripts of SREBP, which has been assigned as a key transcriptional regulator for ergosterol biosynthesis in many eukaryotes,52 were not significantly altered in response to hypoxia. This minor alternation of ergosterol biosynthesis when compared to other fungus may be caused by the different oxygen levels in each testing experiment.53 In terms of oxygen consumption, mitochondrial electron transport chain for ATP formation and sterol biosynthesis are two main sinks. We speculated that the decrease in gene expression involved in sterol biosynthesis, and the down regulation of three Ndh in hypoxic M. irregularis were likely to be a response to the balance shift in TAG and fatty acids catabolism, which normally require more O2 and generate higher amounts of NADH would be generated. This speculation could also explain the GABA shunt response mentioning below. Balance of NADH/NAD+ level during hypoxia may play crucial roles for fungal survival. GABA shunt is an important contributor for energy metabolism that can prevent the NADH accumulation in hypoxic-grown fungal cells.37 The activation of GABA shunt has been reported in sev- 11 eral fungi that were cultivated under limited oxygen conditions.18,37,54 For instance, the expressions of almost all genes in the GABA-shunt pathway including the glutamate decarboxylase, 4-aminobutyrate transaminase, and succinate semialdehyde dehydrogenase were induced from 1.6to 5.7-fold in A.nidulans.55 In M. irregularis, seven differentially expressed genes were identified involved in the GABA shunt, of which that the Gad and GatA expression were decreased but the VGATs were up-regulated in response to hypoxia. The KEGG predicted that glutamate pathway also showed that Gad was associated with glutamate biosynthesis, and had a wide range of functions depending on the organism. For example, in S. cerevisiae, GAD1 is critical for cell tolerance to oxidative stress.56 However, M. irregularis may benefit by the decreased levels of transcripts associated with the GABA shunt and glutamate biosynthesis to regulate the intracellular redox status of cell in response to hypoxia. Intracellular lipid homeostasis is vital for normal membrane structure and function, as well as for cell survival in response to lipid metabolism perturbations resulting from environmental stresses.57 Triacylglycerol (TAG) metabolism is a central core for intracellular lipid homeostasis, since the TAG is not only a source of cellular energy, but also a key player in lipid synthesis, particularly in membrane biogenesis.58 The enzyme TAG lipase which always degrades triacylglycerol into fatty acid can also participate in TAG mobilization and phospholipid metabolism through its lysophospholipid acyltransferase activity.58 We found that TAG lipase (GFBC01029180) of M. irregularis was observed to be up-regulated under hypoxia (Fig. 5, 6). However, this gene’s expression had not been described as differentially expressed to hypoxia in the other six pathogenic fungi mentioned above when exposing to hypoxia. Also, the expression level of GFBC01029364, a member of the cytosolic phospholipase A2 group IV family catalyzing phosphatidylcholine to acyl-glycero-3-phosphocholine for cellular energy, was increased as well. In agreement with an activation of lipid/fatty acid degrading process, we found that the ACSL (GFBC01014149) and ACOX (GFBC01011826) of M. irregulairs, encoding the proteins for β-oxidation were also up-regulated, which was different to the decreased ACOX level in A. fumigatus in response to hypoxia.18 These results suggested that the hypoxic M. irregularis rather use the intracellular lipid pool instead of the carbohydrates as energy source. Three HSP proteins associated with endocytosis were upregulated under the hypoxic condition. The HSP71 gene has a high degree of homology to other Hsp70.59 The HSP70 protein can bind to the plasma membrane of macrophage, specifically on its lipid raft-microdomain and functions as 12 Figure 8. Growth comparison of Mucor irregularis CBS103.93 on MEA medium treated with or without triglycerides. A total of 5 × 103 conidia were spot inoculated onto the plates, and incubated in normoxia and hypoxia (6% O2 ) conditions at 27◦ C for 3 days. Bar = 2 cm. Medical Mycology, 2017, Vol. 00, No. 00 In conclusion, this is the first transcriptome study to our knowledge to provide new insights into the molecular mechanisms of pathogenesis in M. irregularis based on its adaptation to hypoxia. Major responses to hypoxia observed in this study include: decreased gene transcription in the glycolysis, oxidative phosphorylation and carbon metabolism in contrast to previous observations in other fungi such as Aspergillus spp. In contrast, the levels of transcription in genes involved in the lipid/fatty acid metabolism and endocytosis were up-regulated in response to hypoxia. We hypothesize that M. irregularis cells may use the intra-lipid pool and the lipid absorbed from the extracellular environment through endocytosis as energy source during its infection. This transcriptome (RNA-seq) investigation has provide significant baseline data for future clinical, molecular and genetic studies in M. irregularis towards understanding of infection mechanism and biomarkers development in rapid disease diagnosis and treatment. Supplementary material an enhancer when macrophage-mediated antigen uptake has taken place, which in turn stimulates the phagocytosis process.60,61 Given that the three HSP genes associated with endocytosis in M. irregularis were induced under hypoxia, we speculated that M. irregularis might use this mechanism to absorb the extracellular lipid through endocytosis. Furthermore, M. irregularis infections often present as a destructive skin lesion at exposed surface of the skin, typically in the central face area where the sebaceous glands are abundantly distributed.9,13,62–64 In general, the sebum of human sebaceous glands is primarily composed of triglycerides (∼41%), wax esters (∼26%), squalene (∼12%), and free fatty acids (∼16%). So we hypothesized that hypoxia may increase the endocytosis of M. irregularis for extracellular nutrient sources such as triglycerides and fatty acids. To verify this hypothesis, we investigated the hypoxic growth of M. irregularis on MEA and MEA supplemented with triglycerides. Our results showed that the sizes of M. irregularis colonies on the medium supplemented with triglycerides were larger after 3 days’ incubation under hypoxia (Fig. 8). The phenotypic growth supported our hypothesis that triglycerides or fatty acids could promote the growth of M. irregularis under hypoxia. In summary, the hypoxic microenvironments may suppress carbohydrates metabolism in M. irregularis during infection but accelerate fatty acid metabolism to meet energy demands. Also, the lipid uptake from host serum may provide the extracellular nutrient source as energy/resources for the hypoxic growth during the infection, which then defines the specific pathogenicity of M. irregularis. Supplementary data are available at MMYCOL online. Acknowledgments This work was supported by the National Natural Science Foundation of China [grant No. 81471905], the Postdoctoral Science Foundation of Chinese Academy of Medical Sciences and Peking Union Medical College , and the National Basic Research Program of China (973 Program) [grant No. 2013CB531600]. Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and the writing of the paper. References 1. 2. 3. 4. 5. 6. 7. 8. Bitar D, Van Cauteren D, Lanternier F et al. Increasing incidence of zygomycosis (mucormycosis), France, 1997–2006. Emerg Infect Dis. 2009; 15: 1395–1401. Schofield C, Stern A, Jevtic A. Disseminated zygomycosis due to Mycocladus corymbifera with cutaneous and cerebral involvement. Australas J Dermatol. 2013; 54: e8–e11. Skiada A, Petrikkos G. Cutaneous mucormycosis. Skinmed. 2013; 11: 155–159; quiz 159–60. Alvarez E, Cano J, Stchigel AM et al. Two new species of Mucor from clinical samples. Med Mycol. 2011; 49: 62–72. Zheng RY, Chen CQ. A non-thermophilic Rhizomucor causing human primary cutaneous mucormycosis. Mycosystema. 1991; 4: 45–57. Kang D, Jiang X, Wan H, Ran Y, Hao D, Zhang C. Mucor irregularis infection around the inner canthus cured by amphotericin B: a case report and review of published literatures. Mycopathologia. 2014; 178: 129–133. Xia XJ, Shen H, Liu ZH. Primary cutaneous mucormycosis caused by Mucor irregularis. Clin Exp Dermatol. 2015; 40: 875–878. Xia ZK, Liu C, Cong L et al. Mucormycosis caused by Mucor irregularis: a retrospective review of 20 cases (in Chinese). J Pract Dermatol. 2014; 7: 161–165. Xu et al. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Lu XL, Liu ZH, Shen YN et al. Primary cutaneous zygomycosis caused by Rhizomucor variabilis: a new endemic zygomycosis? A case report and review of 6 cases reported from China. Clin Infect Dis. 2009; 49: e39–43. Spellberg B, Edwards J, Jr, Ibrahim A. Novel perspectives on mucormycosis: pathophysiology, presentation, and management. Clin Microbiol Rev. 2005; 18: 556–569. Tomita H, Muroi E, Takenaka M et al. Rhizomucor variabilis infection in human cutaneous mucormycosis. Clin Exp Dermatol. 2011; 36: 312–314. Wang SB, Li RY, Yu J. Identification and susceptibility of Rhizomucor spp. isolated from patients with cutaneous zygomycosis in China. Med Mycol. 2011; 49: 799–805. Xia Z, Wang W, Yang R. Slowly progressive cutaneous, rhinofacial, and pulmonary mucormycosis caused by Mucor irregularis in an immunocompetent woman. Clin Infect Dis. 2013; 56: 993–995. Lu XL, Najafzadeh MJ, Dolatabadi S et al. Taxonomy and epidemiology of Mucor irregularis, agent of chronic cutaneous mucormycosis. Persoonia. 2013; 30: 48–56. Irga N, Kosiak W, Jaworski R, Komarnicka J, Birkholz D. Hyperthyroidism secondary to disseminated mucormycosis in a child with acute lymphoblastic leukemia: case report and a review of published reports. Mycopathologia. 2013; 175: 123–127. Niehoff K, Barnikol WKR. A new measuring device for non-invasive determination of oxygen partial pressure and oxygen conductance of the skin and other tissues. Oxygen Transport to Tissue XXI. 1999; 81: 705–714. Nizet V, Johnson RS. Interdependence of hypoxic and innate immune responses. Nat Rev Immunol. 2009; 9: 609–617. Barker B, Kroll K, Vodisch M, Mazurie A, Kniemeyer O, Cramer R. Transcriptomic and proteomic analyses of the Aspergillus fumigatus hypoxia response using an oxygen-controlled fermenter. BMC Genom. 2012; 13: 62. Hillmann F, Shekhova E, Kniemeyer O. Insights into the cellular responses to hypoxia in filamentous fungi. Curr Genet. 2015; 61: 441–455. Lima Pde S, Chung D, Bailão AM, Cramer RA, Soares CM. Characterization of the Paracoccidioides hypoxia response reveals new insights into pathogenesis mechanisms of this Important human Pathogenic fungus. PLoS Negl Trop Dis. 2015; 9: e0004282. Wezensky SJ, Cramer RA, Jr. Implications of hypoxic microenvironments during invasive aspergillosis. Med Mycol. 2011; 49: S120–124. Chung D, Barker BM, Carey CC et al. ChIP-seq and in vivo transcriptome analyses of the Aspergillus fumigatus SREBP SrbA reveals a new regulator of the fungal hypoxia response and virulence. PLoS Pathog. 2014; 10: e1004487. Grahl N, Shepardson KM, Chung D, Cramer RA. Hypoxia and fungal pathogenesis: to air or not to air? Eukaryot Cell. 2012; 11: 560–570. Sellam A, van het Hoog M, Tebbji F, Beaurepaire C, Whiteway M, Nantel A. Modeling the transcriptional regulatory network that controls the early hypoxic response in Candida albicans. Eukaryot Cell. 2014; 13: 675–690. Terabayashi Y, Shimizu M, Kitazume T, Masuo S, Fujii T, Takaya N. Conserved and specific responses to hypoxia in Aspergillus oryzae and Aspergillus nidulans determined by comparative transcriptomics. Appl Microbiol Biotechnol. 2012; 93: 305–317. Grabherr MG, Haas BJ, Yassour M et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol. 2011; 29: 644–652. Haas BJ, Papanicolaou A, Yassour M et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013; 8: 1494–1512. Shawn TO, Scott JE. Assessing De Novo transcriptome assembly metrics for consistency and utility. BMC Genom. 2013; 14: 465. Conesa A, Gotz S, Garcia-Gomez JM, Terol J, Talon M. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005; 21: 3674–3676. Ye J, Fang L, Zheng H, Zhang Y, Chen J. WEGO: a web tool for plotting GO annotations. Nucleic Acids Res. 2006; 34: W293–297. Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res. 1997; 7: 986–995. 13 32. Benjamini Y, Drai D, Elmer G. Controlling the false discovery rate in behavior genetics research. Behav Brain Res. 2001; 125: 279–284. 33. Xie C, Mao X, Huang J et al. KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Res. 2011; 39: W316–322. 34. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001; 25: 402–408. 35. Askew C, Sellam A, Epp E et al. Transcriptional regulation of carbohydrate metabolism in the human pathogen Candida albicans. PLoS Pathog. 2009; 5: e1000612. 36. Chun CD, Liu OW, Madhani HD. A link between virulence and homeostatic responses to hypoxia during infection by the human fungal pathogen Cryptococcus neoformans. PLoS Path. 2007; 3: e22. 37. Masuo S, Terabayashi Y, Shimizu M, Fujii T, Kitazume T, Takaya N. Global gene expression analysis of Aspergillus nidulans reveals metabolic shift and transcription suppression under hypoxia. Mol Genet Genom. 2010; 284: 415–424. 38. Setiadi ER, Doedt T, Cottier F, Noffz C, Ernst JF. Transcriptional response Candida albicans to hypoxia: Linkage of oxygen sensing and Efg1p-regulatory networks. J Mol Biol. 2006; 361: 399–411. 39. Synnott JM, Guida A, Mulhern-Haughey S, Higgins DG, Butler G. Regulation of the hypoxic response in Candida albicans. Eukaryot Cell. 2010; 9: 1734–1746. 40. Goode BL, Eskin JA, Wendland B. Actin and endocytosis in budding yeast. Genetics. 2015; 199: 315–358. 41. Weinberg J, Drubin DG. Clathrin-mediated endocytosis in budding yeast. Trends Cell Biol. 2012; 22: 1–13. 42. Pitt JI, Hocking AD. Fungi and Food Spoilage, 3rd edn. New York: Springer, 2009; 156. 43. Gancedo C, Flores CL. The importance of a functional trehalose biosynthetic pathway for the life of yeasts and fungi. FEMS Yeast Res. 2004; 4: 351–359. 44. Cao Y, Wang Y, Dai B et al. Trehalose is an important mediator of Cap1p oxidative stress response in Candida albicans. Biol Pharm Bull. 2008; 31: 421–425. 45. Sasano Y, Haitani Y, Hashida K, Ohtsu I, Shima J, Takagi H. Simultaneous accumulation of proline and trehalose in industrial baker’s yeast enhances fermentation ability in frozen dough. J Biosci Bioeng. 2012; 113: 592–595. 46. Zakharova K, Tesei D, Marzban G, Dijksterhuis J, Wyatt T, Sterflinger K. Microcolonial fungi on rocks: a life in constant drought? Mycopathologia. 2013; 175: 537–547. 47. Todd BL, Stewart EV, Burg JS, Hughes AL, Espenshade PJ. Sterol regulatory element binding protein is a principal regulator of anaerobic gene expression in fission yeast. Mol Cell Biol. 2006; 26: 2817–2831. 48. Kroll K, Pähtz V, Hillmann F et al. Identification of hypoxia-inducible target genes of Aspergillus fumigatus by transcriptome analysis reveals cellular respiration as an important contributor to hypoxic survival. Eukaryot Cell. 2014; 13: 1241–1253. 49. Watanabe S, Zimmermann M, Goodwin MB, Sauer U, Barry CE, Boshoff HI. Fumarate reductase activity maintains an energized membrane in anaerobic Mycobacterium tuberculosis. PLoS Pathog. 2011; 7: e1002287. 50. Fu Z, Verderame TD, Leighton JM et al. Exometabolome analysis reveals hypoxia at the up-scaling of a Saccharomyces cerevisiae high-cell density fed-batch biopharmaceutical process. Microb Cell Fact. 2014; 13: 32. 51. Lee H, Bien CM, Hughes AL et al. Cobalt chloride, a hypoxia-mimicking agent, targets sterol synthesis in the pathogenic fungus Cryptococcus neoformans. Mol Microbiol. 2007; 65: 1018–1033. 52. Chung D, Haas H, Cramer RA. Coordination of hypoxia adaptation and iron homeostasis in human pathogenic fungi. Front Microbiol. 2012; 3: 381. 53. Blosser SJ, Cramer RA. SREBP dependent triazole susceptibility in Aspergillus fumigatus is mediated through direct transcriptional regulation of erg11A/cyp51A. Antimicrob Agents Chemother. 2012; 56: 248–257. 54. Diano A, Peeters J, Dynesen J, Nielsen J. Physiology of Aspergillus niger in oxygen-limited continuous cultures: influence of aeration, carbon 14 55. 56. 57. 58. 59. Medical Mycology, 2017, Vol. 00, No. 00 source concentration and dilution rate. Biotechnol Bioeng. 2009; 103: 956–965. Shimizu M, Fujii T, Masuo S, Takaya N. Mechanism of de novo branchedchain amino acid synthesis as an alternative electron sink in hypoxic Aspergillus nidulans cells. Appl Environ Microbiol. 2010; 76: 1507– 1515. Coleman ST, Fang TK, Rovinsky SA, Turano FJ, Moye-Rowley WS. Expression of a glutamate decarboxylase homologue is required for normal oxidative stress tolerance in Saccharomyces cerevisiae. J Biol Chem. 2001; 276: 244–250. Holthuis JCM, Menon AK. Lipid landscapes and pipelines in membrane homeostasis. Nature. 2014; 510: 48–57. Kurat CF, Wolinski H, Petschnigg J et al. Cdk1/Cdc28-dependent activation of the major triacylglycerol lipase Tgl4 in yeast links lipolysis to cell-cycle progression. Mol Cell. 2009; 33: 53–63. Macellaro A, Tujulin E, Hjalmarsson K, Norlander L. Identification of a 71-kilodalton surface-associated Hsp70 homologue in Coxiella burnetii. Infect Immun. 1998; 66: 5882–5888. 60. Wang R, Kovalchin JT, Muhlenkamp P, Chandawarkar RY. Exogenous heat shock protein 70 binds macrophage lipid raft microdomain and stimulates phagocytosis, processing, and MHC-II presentation of antigens. Blood. 2006; 107: 1636–1642. 61. Zhu YZ, Cao MM, Wang WB et al. Association of heat-shock protein 70 with lipid rafts is required for Japanese encephalitis virus infection in Huh7 cells. J Gen Virol. 2012; 93: 61–71. 62. Li DM, Lun LD. Mucor irregularis infection and lethal midline granuloma: a case report and review of published literature. Mycopathologia. 2012; 174: 429–439. 63. Thody AJ, Shuster S. Control and function of sebaceous glands. Physiol Rev. 1989; 69: 383–416. 64. Zhao Y, Zhang Q, Li L, Zhu J, Kang K, Chen L. Primary cutaneous mucormycosis caused by Rhizomucor variabilis in an immunocompetent patient. Mycopathologia. 2009; 168: 243–247. 65. Shimizu M, Fujii T, Masuo S, Fujita K, Takaya N. Proteomic analysis of Aspergillus nidulans cultured under hypoxic conditions. Proteomics. 2009; 9: 7–19.