The CYPome of the model xenobiotic-biotransforming fungus Cunninghamella elegans

The fungus Cunninghamella elegans is recognised as a microbial model of mammalian drug metabolism owing to its ability to catabolise xenobiotic compounds in an analogous fashion to animals. Its ability to produce phase I (oxidative) metabolites of drugs is associated with cytochrome P450 (CYP) activity; however, almost nothing is known about these enzymes in the fungus. In this paper we report the in silico analysis of the genome sequence of C. elegans B9769, which contains 32 genes putatively coding for CYPs. Based on their predicted amino acid sequences these were classified as belonging to CYP509, 5203, 5208, 5313, 5210, 61 and 51 families. Reverse transcription-quantitative PCR revealed that the gene coding for CYP5313D1 was significantly upregulated when C. elegans DSM1908 was cultivated in sabouraud dextrose in contrast to its expression in cells grown in Roswell Park Memorial Institute medium. This corresponded to the fungus’ xenobiotic biotransformation ability when grown in the two media. Heterologous expression of cyp5313D1 in Pichia pastoris resulted in a recombinant strain that biotransformed flurbiprofen to 4′-hydroxyflurbiprofen, the same metabolite generated by C. elegans cultures. This is the first report of a xenobiotic-biotransforming CYP from this biotechnologically important fungus.

www.nature.com/scientificreports www.nature.com/scientificreports/ partial proteins, inferred by their shorter length (<300 amino acids) or were identified as other enzymes when queried using BLASTP on GenBank ® . Fungal CYPs are mainly class II 21 , which are membrane bound proteins and require a redox partner (see following section) to deliver electrons to the haem. Using the web-based algorithm SOSUI (http://bp.nuap.nagoya-u. ac.jp/sosui), the solubility of the C. elegans CYPs was investigated and curiously only five (5208A3, 5206U1, 51, 5203A22, 509Q1) were predicted to have a transmembrane region. It seems highly unlikely that so many of the CYPs would be soluble, since to date only one soluble fungal CYP, P450nor, has been identified 21 . Further biochemical analysis of the CYPs will help shed light on this anomaly.
Of particular interest for this study were the potential xenobiotic-biotransforming CYPs. Apart from CYP51 and CYP61, which are highly conserved in fungi and are involved in sterol biosynthesis 22 , the functions of the other CYPs are either unknown or not experimentally demonstrated. As C. elegans is known for mimicking mammalian phase I (oxidative) biotransformations, relatedness to mammalian CYPs was used as a key criterion to identify potential xenobiotic-biotransforming enzymes. CYPs 5313D1 and 5313E1 were of particular interest as their amino acid sequences showed comparatively high identity (approx. 31%) to the main mammalian CYP involved in drug detoxification, CYP3A4. In contrast, the previously cloned CYP509A1 12 had a 24% identity to this enzyme. Furthermore, members of the CYP53 family in fungi are associated with 4-hydroxylation of benzoic acid and its derivatives 22 . In C. elegans, para-hydroxylation of aryl rings of xenobiotics such as diclofenac and flurbiprofen, is the predominant phase I pathway 23 . BLASTP searches of the predicted C. elegans proteome using the amino acid sequence of CYP53A1 from A. niger and CYP53A19 from F. oxysporum revealed that the highest similarity was to CYP5313E1 (26 and 31%, respectively).

Identification of redox partner proteins.
To identify cytochrome P450 reductases (CPRs), the known C. elegans CPR sequence 14,15 was used as a sequence for BLASTP analysis. The hypothetical CPRs were checked for a flavin adenine dinucleotide (FAD)-binding domain and a NAD(P)H-dependent flavin mononucleotide (FMN) domain. Additionally, cytochrome b5 (Cyt-b5) and cytochrome b5 reductase (Cyt-b5r) sequences from Phanerochaete chrysosporium 24 were used to identify possible alternative CYP redox partners. Eight sequences were identified as possible NADPH reductase partners (Fig. S1). Three were identified as CPRs (Fig. S2), one of which (g4301) has been previously reported 14 . One sequence was identified as a NADPH-dependent diflavin oxidoreductase; however these enzymes are typically involved in the biogenesis of FeS proteins 25 . Four more sequences were identified as alternative redox partners and one sequence was recognised as a cytochrome b5. The remainder were identified as nicotinamide adenine dinucleotide (NADH) cytochrome b5 reductases.
CYp expression in C. elegans DSM1908. C. elegans DSM1908 is one of a number of strains that has been used by researchers to study drug metabolism 26,27 . To confirm that the genes identified in the in silico analysis of the C. elegans B9769 are also present in the DSM1908 strain, a set of internal primers were designed to amplify partial sequences of seven genes (Table S1); the genes chosen were selected based on their similarity to mammalian CYPs. These primers were used in PCR experiments with both genomic DNA and complementary DNA (cDNA) isolated from C. elegans DSM1980. PCR was performed initially using gDNA template and the targets were visualised by gel-electrophoresis, for which the expected amplicon sizes were detected (Fig. 1A). The same primers were then employed to investigate gene expression by amplifying targets from cDNA. As the internal primers were designed to span exon-exon boundaries, the products amplified from cDNA were smaller due to www.nature.com/scientificreports www.nature.com/scientificreports/ an absence of intron regions (Fig. 1B). Furthermore, only a single band was visualised by gel-electrophoresis representing an effective extraction of messenger RNA and subsequent reverse-transcription. The correct amplicon size also confirmed the accuracy of the gene prediction by supporting the position and length of the introns within the ORF. Finally, the amplicons were sequenced and confirmed that the genes were the same in both strains with >92% sequence identity. The sequence data were inputted into the genome browser Artemis and validated the correct position of the sequence.
Based on the visualisation of gene transcripts in Fig. 1B, CYPs 509Q1, 5206U1, 5206U3, 5205A8, 5208A3, 5201C1 and 5313E1 are expressed by C. elegans when cultivated for 72 h in the growth medium sabouraud dextrose broth (SDB), which is the medium often used to ensure good xenobiotic biotransformation in this fungus. Expression of CYP5313D1 under the same conditions was confirmed in a separate experiment (Fig. S3). Additionally, the cDNA amplicons reinforced the accuracy of the gene prediction through confirmation of correct product sizes.

CYP5313D1 is up-regulated in C. elegans grown in sabouraud dextrose medium. A previous
investigation demonstrated that when C. elegans biofilm was grown in Roswell Park Memorial Park (RPMI) medium, biotransformation of flurbiprofen was dramatically reduced compared with cells grown in SDB 28 . In the present study suspended cultures of C. elegans grown in these media showed the same pattern when incubated with the non-steroidal anti-inflammatory drug flurbiprofen (5 mg). Analysed by HPLC, 3.30 mg (66%) of 4′-hydroxyflurbiprofen was recovered from cultures grown in SDB after 72 h, whereas cells cultured in RPMI only yielded 0.1 mg (2%) of the metabolite.
This metabolite is the result of CYP monooxygenase activity, but to date the specific enzymes involved in the biotransformation are not known. The in silico analysis of the CYPome has revealed two CYPs (5313D1 and 5313E1) that may be involved in xenobiotic biotransformations. To investigate this further RT-qPCR was employed to look for changes in gene expression of cyp5313D1 and cyp5313E1 under the conditions optimal for biotransformation. In addition the change in expression of the previously identified cyp509A1 and cpr was also measured. cDNA from four experimental conditions was prepared for investigation, which were C. elegans cultured in sabouraud dextrose broth and RPMI for 72 h and 96 h. Comparison of Cq values of technical replicates resulted in 86.9% falling within 0.5 Cq and 98.8% within 0.8 Cq. As expected, biological replication produced a higher level of variation with 50% of the replicates falling within 1.21 Cq and 95.8% within 2.65 Cq. Amplification efficiencies measured from serial dilutions of cDNA of all but the reference genes cpr and sarA were 1.87 or higher (Table S4). To normalise the sample data, a geNorm analysis was initiated on 11 samples and 3 reference targets (M = 0.440, V = 0.181). The genes ubcB (ubiquitin carrier protein) and sarA (secretion associated GTP-binding protein) were assessed as reference targets based on a previous study that identified a number of stable reference genes in filamentous fungi 29 . In addition, given its high expression stability, cpr was repurposed as a reference gene. The optimal number of reference targets in this experimental situation was 2 (geNorm V = 0.143): ubcB and cpr.
The results shown in Fig. 2 were normalised to ubcB and cpr and calibrated against the RPMI 72 group, with positive and negative values showing up-and down-regulation of the gene of interest, respectively. The largest effect in the dataset was observed as a dramatic up-regulation of cyp5313D1 in C. elegans when grown in SDB (~20 fold change, p < 0.0001). The same gene was 3.4 times down-regulated in the RPMI 96 group, possibly because the cells grown in this medium were under severe nutritional limitation at this time. The other genes remained consistently expressed in the various conditions (cyp5313E1) or were significantly down regulated in SDB (cyp509A1).

CYP5313D1 is a xenobiotic-biotransforming enzyme. The in silico analyses and expression studies
suggested that CYP5313D1 is involved in phase I biotransformation of xenobiotics, thus the gene was overexpressed in a heterologous host. The yeast P. pastoris was selected as the host, since it has been used successfully   30 and has endogenous reductases. The gene coding for CYP5313D1 was cloned into pPICZ A without the N-terminal signal peptide so that the protein would not be exported from the cell, resulting in the expression plasmid pPICZ-cyp8. Positive transformants were confirmed by PCR. The recombinant strain was cultivated in BMY medium and protein expression was induced by the addition of methanol. Expression was confirmed by Western blot analysis of the lysate (Fig. 3) and showed that the expressed protein was localised in the insoluble fraction, which is consistent with fungal CYPs. Whilst proteins with both His-and myc-epitopes were detected in the recombinant P. pastoris cell extracts, only a faint band at the correct molecular weight (approx. 59 kDa) was detected using anti-his antibodies, indicating that extensive degradation of the protein had occurred.
Activity of recombinant CYP5315D1 was determined by adding the drug flurbiprofen to cultures after induction. The entire culture was extracted with ethyl acetate after 48 h incubation, and the extract dried and the residue silylated for GC-MS analysis. Figure 4 shows that after incubation with the drug, the metabolite 4′-hydroxyflurbiprofen was detected in extracts from the recombinant culture, whereas no 4′-hydroxylated metabolite was observed in control (wild type) cultures incubated with drug (Fig. 4B). The same metabolite was observed as the most prominent metabolite when C. elegans was incubated with flurbiprofen 26 , thus CYP5313D1 is the first xenobiotic-biotransforming CYP to be identified from the fungus.

Discussion
Cunninghamella spp. have been studied for decades for their ability to produce phase I metabolites of drugs and other xenobiotics that are analogous to those produced in mammals 1 . Despite this, their CYPs have until now proven inaccessible. In this paper we identified the CYPome of the fungus C. elegans B9769 through in silico analysis of the genome, and found that there are 32 putative CYPs based on sequence homology and the presence of conserved motifs. The CYPs were classified according to the established methods. Two CYPs (5313D1 and 5313E1) were identified as putative xenobiotic-biotransforming enzymes, based on their sequence similarity to mammalian CYP3A4, which is involved in phase I metabolism, and members of the CYP53 family, which are known to catalyse hydroxylation of benzoic acid in basidiomycetes 31 . www.nature.com/scientificreports www.nature.com/scientificreports/ Since the C. elegans genome sequence that was available is from a strain not previously investigated for xenobiotic biotransformation (B9769), PCR experiments were conducted using gDNA and cDNA isolated from C. elegans DSM 1908, which is an established xenobiotic-metabolising strain. The sequences of amplicons obtained matched those of the B9769 strain, confirming the presence of the same genes in 1908. End point RT-PCR experiments demonstrated the expression of eight putative CYPs when the fungus was cultivated in the rich medium sabouraud dextrose broth, which is routinely used in biotransformation experiments. RT-qPCR confirmed that CYP5313D1, identified as coding for a likely xenobiotic-transforming enzyme, was dramatically up-regulated when the fungus was grown in sabouraud dextrose broth compared to RPMI. When C. elegans was grown in the latter medium the biotransformation of the drug flurbiprofen was very poor in comparison to cells grown in SDB, which was further evidence that CYP5313D1 is an enzyme involved in xenobiotic metabolism in the fungus. Other researchers [15][16][17] previously investigated CYP expression in Cunninghamella spp. by RT-PCR, using primers based on the conserved haem-binding region of cytochromes P450. Whilst these studies have demonstrated up-regulation of cytochrome P450, they did not take account of the different CYP genes.
Confirmation that CYP5313D1 was capable of hydroxylating a xenobiotic was by heterologous expression of the cyp5313D1 gene in P. pastoris and biotransformation of flurbiprofen to the known phase I metabolite 4′-hydroxyflurbiprofen by the recombinant yeast. At present the recombinant system is unoptimised, relying on the endogenous Cpr of P. pastoris to supply reducing power to the CYP rather than co-expressing a C. elegans reductase. Future studies will focus on improving product yield through optimising the expression system which will enable a more comprehensive analysis of the enzyme's properties to be conducted. This is the first xenobiotic-biotransforming cytochrome P450 to be identified in this biotechnologically important fungus and the in silico analysis of the Cypome provides a platform for further studies on the other genes.

Materials and Methods
Fungal strain and reagents. Cunninghamella elegans DSM1908 was obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ, Germany) and maintained on sabouraud dextrose agar. Roswell Park Memorial Institute 1640 (RPMI) medium and flurbiprofen were acquired from Sigma-Aldrich, Arklow, Ireland. Pichia pastoris X-33 was employed as a fungal host for recombinant protein expression (Fisher Scientific, Dublin, Ireland) and pPICZ A was the vector. YPD medium (Yeast extract Peptone Dextrose) that was used routinely for www.nature.com/scientificreports www.nature.com/scientificreports/ maintenance and propagation of P. pastoris X-33 and BMY(G) medium used for protein expression were prepared according to the EasySelect ™ Pichia Expression Kit manual. For the preparation of transformed P. pastoris cells, sorbitol (182.2 g/l) was added to the YPD agar to stabilise the cells to osmotic pressure. Restriction enzymes (New England Biolabs) were purchased from Brennan and Company, Dublin, Ireland. Gibco ® Zeocin ™ Selection Reagent (100 mg/ ml) and T4 DNA Ligase 1 U/µL (Invitrogen) were acquired from Fischer Scientific. CYp discovery. Identification of candidate sequences were verified through a multiple sequence alignment tool, MUSCLE 32 , to search for conserved regions among the cytochromes P450. These conserved motifs included the haem-binding domain (FXXGXRXCXG), the PERF domain (PXRX) and the K-helix region (EXXR) 22,33,34 . The genome browser Artemis 35 was used to visualise the surrounding sequence features to check for errors generated by the gene prediction program. The deduced amino acid sequences were submitted to David Nelson and named according to the International P450 Nomenclature Committee.
Culture conditions. Cunninghamella elegans DSM1908 was cultivated on sabouraud dextrose agar for 120 h at 28 °C. Inoculum was prepared by homogenising the mycelium in sterile 0.8% NaCl (100 ml). The homogenate (5 ml) was used to inoculate 250 ml Erlenmeyer flasks containing 45 ml of either sabouraud dextrose broth (SDB) or RPMI, and then incubated with rotary agitation (150 rpm) at 28 °C.
Genomic DNA extraction. C. elegans cultures were grown for 72 h and the biomass (200 mg) was collected aseptically using sterile tweezers after separation from the supernatant by centrifugation (Eppendorf 5810 centrifuge). The biomass was then flash frozen with liquid nitrogen and ground into a powder using a pestle and mortar as described previously 17,36 . The crushed biomass was resuspended immediately in 300 μl of MicroBead Solution from the DNeasy UltraClean Microbial Kit (Qiagen) and processed using the steps outlined in the instruction manual. An increased yield of genomic DNA (gDNA) was achieved by heating the preps at 65 °C before bead beating. total RNA extraction. RNA was extracted from cultures that had been grown in either SDB or RPMI media for 72 h and 96 h. The biomass was collected and handled as described for gDNA isolation. The crushed biomass was resuspended in 1 ml of TRIzol reagent (Invitrogen), vortexed and incubated at room temperature for 5 min. The manufacturer's instructions were followed and the final pellet was air-dried for 5 min before suspending in diethylpyrocarbonate (DEPC)-treated water (90 μl) and incubated at 65 °C for 5 min. To digest DNA, Turbo DNA-free TM kit (Invitrogen) was employed, before finally extracting the RNA from the supernatant using an RNeasy mini extraction kit (Qiagen) following the manufacturer's instructions. Aliquots of the RNA samples were prepared for measuring the final RNA concentration was measured using an Epoch Microplate Spectrophotometer (Biotek) and a Take3 plate (Biotek). RNA samples were stored at −80 °C. preparation of complementary DNA. Template using extracted RNA (100 ng) was prepared on ice and diluted in DEPC treated water to a final volume of 6.1 μl in sterile pre-chilled PCR tubes and random hexamer primers (5 μl) were added. Tubes were incubated at 70 °C for 5 min and then held 4 °C. Reverse transcription was then carried out using an ImProm-II ™ reverse transcription system (Promega). Amplification was conducted with a thermocycler (Techne TC-3000 thermocycler) with annealing at 25 °C for 5 min, extension at 42 °C for 1 h followed by heat inactivation at 70 °C for 15 min and then held at 4 °C. Complementary DNA (cDNA) samples were stored at −20 °C for future polymerase chain reactions. pCR. Primer design was based on the genome sequence of C. elegans B9769 for the gDNA and cDNA produced from C. elegans DSM1908. Oligonucleotide primers were synthesised by Eurofins Genomics (Ebersberg, Germany) and are listed in Table S1. Amplification of target sequences was enabled using either Phusion (New England Biolabs) or Taq (Bioline) DNA polymerases. PCR products were routinely purified using a QIAquick PCR Purification Kit (Qiagen) and sent for Sanger DNA sequencing (SUPREMERUN, GATC sequencing). DNA concentrations were measured using the Epoch Microplate Spectrophotometer and Take3 plate. The nucleotide sequence was then aligned with the predicted gene sequence from C. elegans B9769 using the pairwise sequence alignment tool, EMBOSS Needle. To build full length gene sequences full length and internal overlapping primers were used to 'primer walk' to build a curated gene sequence. This was done to search for any incorrect coding sequence data created by the gene prediction (e.g. exon-exon junctions) or any single nucleotide polymorphisms that existed between the bioinformatics strain (C. elegans B9769) and the experimental strain (C. elegans DSM1908).

Quantitative pCR (qpCR).
Quantitative PCR was carried out with the LightCycler 480 SYBR Green I Master using a LightCycler 480 instrument according to the manufacturer's instructions (Roche LifeScience). Briefly, 15 μl of a PCR mix containing SYBR Green and relevant primers (final concentration 0.5 μM) were aliquoted in a LightCycler 480 multiwell plate-96 and added with 5 ul of cDNA template (previously diluted 1:5). Primers listed in Table S2 were designed to generate amplicons between 100-200 bp. To calculate PCR efficiency, standard curves were prepared using serial dilutions of cDNA (1, 1/4, 1/16 and 1/64). Non-template controls were routinely included in each run. Relevant cycling of parameters is outlined in Table S3. Quantification cycle (Cq) values were obtained using the default settings of the instrument. A final melting curve analysis was performed to demonstrate the formation of a single product. Four experimental conditions were included by growing C. elegans DSM1908 in SDB or RPMI medium for either 72 h or 96 h. Each experimental condition was represented by at least two biological replicates and each sample was analysed in duplicate. The geNorm method was employed to determine the optimal number of reference targets for normalization 37 . Analysis of data to assess the quantitative variation in gene expression against the control group (RPMI 72), including statistical analysis, was performed in qBaseplus 38 . Log 10 fold changes (calibrated normalised relative quantities as referred in qBaseplus) were plotted using SigmaPlot 13. www.nature.com/scientificreports www.nature.com/scientificreports/ Heterologous expression of CYP5313D1. The gene encoding CYP5313D1 was amplified from cDNA using primers pαA-8F (5′-ATTCACGTGCGAAACGATGAATGACTTTAATATTTACAATAAATTGGAACAT-3′) and pαA-8R (5′-GAAGTTAAAGCCCAGAAAATTTCCGCGGAATA-3′). Plasmids were digested with PmlI and SacII prior to ligation to the amplicon. E. coli DH5α was transformed and positive transformants were selected based on Zeocin ™ resistance. Plasmid DNA was purified and the constructs verified by sequencing. Plasmids were purified using a QIAprep Spin Miniprep Kit (Qiagen) and linearised by digesting with SacI. The digested DNA was extracted using a phenol:chloroform:isoamyl alcohol (25:24:1) solution and then concentrated by ethanol precipitation (EasySelect ™ Pichia Expression Kit manual). Transformation of electro-competent P. pastoris cells was performed by electroporation (Eppendorf 2510) and transformants were selected by inoculating YPD agar plates containing increasing concentrations of Zeocin ™ . For protein expression in P. pastoris, 20-40 ml of BMYG in a baffled 250 ml Erlenmeyer flask was inoculated with a single colony and grown for 48 h at 30 °C, 290 rpm. The culture was then harvested by centrifugation at 4000 rpm for 10 min and the cells washed with 20 ml of BMY medium. The suspension was then centrifuged at 4000 rpm for 10 min and the pellet was resuspended in 40 ml of BMY medium and decanted into a sterile, baffled 250 ml Erlenmeyer flask and incubated for 4 h at 30 °C, 290 rpm. Filter sterilised methanol (0.5% (v/v)) was then added to induce gene expression, with further additions every 24 h. Confirmation of gene expression was by Western blotting. P. pastoris cells were harvested, resuspended in lysis buffer (50 mM Tris pH 8, 1 mM EDTA, 100 mM NaCl and 1 mM phenylmethylsulfonyl fluoride) and homogenised using BeadBug ™ . The homogenate was sonicated for 1 min and the lysate finally separated by centrifugation. The soluble and insoluble proteins were separated by SDS-PAGE and the recombinant protein immunodetected using mouse IgG anti-His and anti-myc (BioLegend) and HRP-linked anti-mouse IgG (Cell Signalling Technologies) antibodies; detection was by enhanced chemiluminescence.

Biotransformation of flurbiprofen.
Cunninghamella elegans DSM1908 was cultivated in either sabouraud dextrose broth or RPMI for 72 h with rotary agitation (150 rpm) at 28 °C. Flurbiprofen was dissolved in DMF and added to the flasks, which were incubated for a further 72 h. Organic extracts from shake flasks were analysed by reversed-phase high-performance liquid chromatography (HPLC) with a Varian Prostar HPLC system equipped with Zorbax SB-C18 5 μm 4.6 × 150 mm column (Agilent technologies) and a UV-Vis detector monitoring at 250 nm. Samples were injected (25 µl) onto the column and analytes were eluted with a gradient of acetonitrile / water (20-80% acetonitrile) over 35 min at a flow rate of 1 ml/min. Quantification of 4′-hydroxyflurbiprofen was performed by comparing peak areas with a standard curve prepared using previously isolated metabolite 23 .
Whole cell biotransformation experiments with recombinant P. pastoris cultures were conducted. The cells were cultured as in BMY as described previously and flurbiprofen (1 mg) was added 24 h after induction with methanol. The cultures were incubated for another 48 h then extracted with ethyl acetate; the solvent was removed under reduced pressure and the residue derivatised with N-methyl-N-(trimethylsilyl)trifluoroacetamide (100 °C for 1 h). The silylated extracts were analysed by gas chromatography-mass spectrometry (GC-MS) as described previously 26 .

Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.