Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Identification of a consensus motif in Erg28p required for C-4 demethylation in yeast ergosterol biosynthesis based on mutation analysis

Identification of a consensus motif in Erg28p required for C-4 demethylation in yeast ergosterol... Abstract The Erg28p protein is localized to the endoplasmic reticulum, where it acts as a scaffold to tether the C-4 demethylase complex involved in the sterol biosynthesis pathway of Saccharomyces cerevisiae. However, due to the challenges involved in characterizing the interactions of membrane proteins, the precise region of Erg28p that is responsible for the assembly of this enzyme complex remains unknown. To address this question, mutants with serial truncations in the C-terminus of Erg28p were constructed based on a topology prediction of its transmembrane domain. Sterol profiles demonstrated that intermediates involved in the stepwise removal of the two C-4 methyl groups from the tetracyclic sterol ring were accumulated in the ERG28Δ135-447 strain. Homologous alignment of Erg28p further identified a highly conserved 10-amino acid sequence (63LS/QARTFGT/LWT72) within the truncated region of ERG28Δ136-273. Complementation of the BY4741/erg28 strain with the ScERG28Δ175-204 plasmid resulted both in a significant growth inhibition and a reduction of ergosterol biosynthesis compared with the plasmid without the Δ175-204 truncation. Furthermore, homology modeling of the Erg28p mutant indicated that the deletion of residues 63–72 significantly disrupted the 3D structure of the four parallel helices in Erg28p. Taken together, the data indicate that the region spanning amino acids 63–72 constitutes a key consensus motif within Erg28p that is required for sterol C-4 demethylation during ergosterol biosynthesis in S. cerevisiae. C-4 demethylation complex, Erg28p; ergosterol synthesis pathway, conserved motif INTRODUCTION Ergosterol is the predominant sterol constituent of the plasma membrane of S. cerevisiae, and its biosynthetic pathway comprising the ERG gene family members is now well understood (Veitia and Hurst 2001). The sterol molecule becomes functional only after removal of the two nuclear methyl groups at the C-4 position, which involves some of the most complicated processes that have been identified in the sterol biosynthetic pathways of eukaryotic organisms (Bloxham, Wilton and Akhtar 1971; Rahier et al.2006; Rahier 2011). In yeast, the C-4 demethylation process involves a series of three redox reactions, catalyzed by the C-4 oxidase (SMO) encoded by ERG25 (Bard et al.1996), C-3 decarboxylase (3βHSD/D) encoded by ERG26 (Gachotte et al.1998) and C-3 ketoreductase (3SR) encoded by ERG27 (Gachotte et al.1999), respectively. The C-4 methyl group is sequentially oxidized, decarboxylated, ortho-ketonized and reduced to a hydroxyl group, so that after two rounds of demethylation, the C-4 dimethyl is finally removed (Wilton and Akhtar 1975; Rahier et al.2006). In addition to the well-defined catalytic mechanisms of the S. cerevisiae ergosterol biosynthesis pathway, the endoplasmic reticulum (ER) transmembrane protein Erg28p, the only known member of the ERG pathway lacking enzymatic activity, is highly conserved across eukaryotes, and homologs of the ERG28 gene have also been observed in Schizosaccharomyces pombe (Sipiczki 2004), Arabidopsis (Mialoundama et al.2013) and C. elegans (Oh et al.2017). This protein was firstly found to be strongly co-regulated with ergosterol biosynthesis, after which its protein–protein interactions with C-4 demethylation enzymes were confirmed (Mo et al.2002; Mialoundama et al.2013). All four components of the sterol C-4 demethylase complex contain potential transmembrane domains, and the role of Erg28p may be either to tether Erg25p, Erg26p and Erg27p to the ER, or to facilitate the interaction between these proteins, acting as a scaffold promoting the co-localization of the ERG enzymes (Gachotte et al.2001). In addition to the critical role in anchoring the C-4 demethylase complex to the ER, Erg28p was also confirmed to be associated with other ERG members such as Erg11p, Erg6p and Erg1p, demonstrating its crucial role in the ergosterol-biosynthetic enzyme complex (Mo, Valachovic and Bard 2004; Mo and Bard 2005a,b). The identification and characterization of protein–protein interactions between transmembrane proteins is highly challenging because of the physicochemically diverse properties of such proteins and the complexity of the enzymatic reactions associated with the ER membrane (Lacapere et al.2007). Typical biochemical methods, such as co-immunoprecipitation, co-purification or cross-linking, have been used to investigate protein complexes (Herrmann, Westermann and Neupert 2001). Alternative genetic methods such as the split-ubiquitin membrane yeast two-hybrid system also offer powerful tools to understand the protein–protein interactions involved in ergosterol synthesis (Mo and Bard 2005b). However, due to the hydrophobic nature of transmembrane proteins as well as the fact that integral and membrane-associated proteins undergo complex post-translational modifications or oligomerize via interactions between their transmembrane domains, there still are significant challenges in determining the interactions of transmembrane proteins (Thaminy et al.2003). Thus, the precise interacting region in the protein complex organized by Erg28p remains to be investigated further. In the present study, we proposed the sequence extent of the hydrophobic transmembrane domain of S. cerevisiae Erg28p (ScErg28p). Correspondingly, a series of ERG28 mutants with sequential truncations was constructed, and the sterol profiles of strains expressing these mutants were analyzed. Based on the phenotype data of the mutant strains and a structure prediction based on homology-based modeling, a highly conserved consensus motif (63LS/QARTFGT/LWT72) was identified within Erg28p, which is required for the sterol C-4 demethylation reaction involved in ergosterol biosynthesis. MATERIALS AND METHODS Strains, culture conditions and reagents The BY4741 wild-type (WT) strain of S. cerevisiae (ΔMAT, his3Δ, leu2Δ, met15Δ, ura3Δ) and the ERG28 knockout strain Y00177 (erg28) (BY4741; MATa; ura3Δ0; leu2Δ0; his3Δ1; met15Δ0; YER044c::kanMX4) were obtained from Euroscarf (Institute for Molecular Biosciences, Frankfurt, Germany) (http://www.euroscarf.de/), and were grown at 30°C for approximately 18–20 h in YPD medium consisting of 1% yeast extract (Oxoid Ltd, England), 2% Bacto peptone (Becton, Dickinson and Company, France) and 2% glucose (Aladdin Corp., China). Yeast transformants containing plasmids expressing either WT ERG28 or its truncated mutants were grown in complete synthetic uracil-deficient medium (CSM-URA), comprising 0.67% yeast nitrogen base, 2% glucose and 0.8 g/L uracil dropout amino acid mixture (amino acid and nitrogen base supplements; Aladdin Corp.). The yeast transformant carrying the EGFP-ERG26 plasmid was grown in complete synthetic histidine-deficient medium (CSM-HIS), comprising 0.67% yeast nitrogen base, 2% glucose and 0.8 g/L histidine dropout amino acid mixture. All yeast strains were preserved at 4°C on glucose-yeast peptone agar slants. Plasmid DNA was first used to transform Escherichia coli BL21 cells (Promega Corp., USA), which were grown in LB medium comprising 0.5% yeast extract, 1% Bacto peptone, 1% NaCl and 50 mg/L ampicillin at 30°C. The sequences of the oligonucleotide primers and of the ScERG28Δ175-204 mutant were chemically synthesized by Sangon Biotech (Shanghai, China). ExPfu DNA polymerase, restriction enzymes, Tango buffer and T4 ligase were purchased from Thermo-Fisher Scientific (USA). Authentic reference standards of the sterol substrates ergosterol and lanosterol, as well as the antibiotics G418 and ampicillin were purchased from Sigma-Aldrich Corporation (USA). All other chemicals used in this study were from commercial sources and of reagent grade. Molecular cloning of ERG28 and its truncated mutants The ERG28 gene and its truncated mutants were amplified from the chromosomal DNA of S. cerevisiae BY4741. Restriction sites for HindIII and NotI were introduced using the primers listed in Table S1, Supporting Information. The PCR procedure was performed using Taq DNA polymerase with a total number of 30 cycles. Each cycle included three steps: 94°C for 30 s, 57°C–61°C (according to the primers) for 30 s, and 72°C for 90 s. The PCR products were purified and ligated into the pGEM-T Easy vector, which was then digested using the corresponding restriction enzymes. The digested products were ligated into the pYES2 plasmid and used to transform E. coli BL21 competent cells. The positive colonies were confirmed both by bacterial colony-PCR and by restriction-enzyme double digestion. The confirmed recombinant plasmids were extracted from E. coli and used to transform the erg28 strain using the PEG-LiAc transformation method (Sheff and Thorn 2004), yielding ERG28-WT, ERG28-T1, ERG28-T2, ERG28-T3, ERG28-T4 and ERG28-T5, respectively (Fig. 4B). Construction of the EGFP-fusion plasmid and fluorescence microscopy All fluorescence microscopy procedures were performed as described previously (Niedenthal et al.1996). The ERG26 gene with its endogenous promoter was cloned by PCR using the corresponding primers listed in Table S1, Supporting Information. The fragments were gel-purified and ligated into the pKT128 expression vector to generate the EGFP-fusion plasmid pKT128-ERG26. For the microscopy assay, the yeast strains BY4741 and BY4741/erg28 were transformed with pKT128-ERG26 and grown to the exponential phase in liquid CSM-HIS medium or on CSM-HIS plates. For staining of mitochondria and nuclei in living cells, cells grown to the exponential phase were incubated with 4΄,6-diamidino-2-phenylindole dihydrochloride (1:100; Beyotime Institute of Biotechnology Jiangsu, China) and red ER tracker (1:1000; also from Beyotime) for 1 h at room temperature in the dark. Cells were fixed onto a polylysine-coated dish (Thermo Scientific Nunc®) for automated fluorescence imaging using a Zeiss Axioskop fluorescence microscope (Carl Zeiss Microscopy LLC, USA). Spot test analysis of cell growth Serial dilutions comprising 10 μL drops with 106–101 cells of the strains with different ERG28 genetic backgrounds in the early exponential growth phase were spotted onto YPD plates and incubated for 24 h at 30°C, after which the cell growth was analyzed visually and the plates photographed for documentation. The experiment was repeated three times. Sequence alignment, prediction of hydrophobic helices and homology-based structure modeling Alignment of the amino acid sequences encoded by the ERG28 genes from S. cerevisiae (NP_010962.1), Homo sapiens (NP_009107.1), Pan troglodytes (GenBank: JAA12605.1) and Candida albicans (XP_719465.1) was conducted using Vector NTI Advance 11.5 (Invitrogen, Darmstadt, Germany). Extensive BLASTn matrix analysis of a large number of putative Erg28p sequences was processed with Clustal Omega (www.clustal.org). Prediction of hydrophobic helices in Erg28p was conducted using the TMHMM server v. 2.0 (www.cbs.dtu.dk/services/TMHMM/) and the protein-fold recognition server Phyre 2 (Kelley and Sternberg 2009). Prediction of the tertiary structure in both Erg28p and the truncated Erg28p mutants was conducted using the I-TASSER on-line server (Fernandez-Leiro and Scheres 2016), using the protein structures 5H1Q and 5AEZ from the PDB database (www.wwpdb.org) as the templates for Erg28p and the mutant Erg28pΔ63-72, respectively. Extraction of sterols and gas chromatography/mass spectrometry (GC/MS) analysis Sterols were extracted after alkaline saponification in the presence of pyrogallol according to a classical method (Adams and Parks 1968). The yeast cells were collected by centrifugation at 3000× g and 4°C for 10 min, washed twice with distilled water and dried at 80°C to a constant weight. Samples comprising 300 mg of the dried biomass were resuspended in reaction mixtures comprising 5 mL of a 60% KOH solution, 7.5 mL of methanol and 7.5 mL of methanolic pyrogallol (0.5% w/v), and incubated at 80°C for 4 h under constant shaking at 180 rpm. The unsaponified matter was extracted with 10 mL of hexane. Different phases were formed after centrifugation at 3000× g for 10 min. The top n-hexane layer was removed to a clean tube and the residue was re-extracted twice with hexane. The n-hexane phase containing the extracted sterols was recovered, dried under nitrogen and the final obtained sterols were resuspended in 500 μL of chloroform/methanol (4:1) for GC/MS analysis. The sterols were separated and analyzed by GC-MS on an Agilent 3400 system (Agilent, USA) equipped with a fused silica column (DB-5, 15 m × 0.32 mm × 0.25 μm film thickness; J&W Scientific, USA), with helium at a flow rate of 1 mL/min as the carrier gas. The temperature program comprised a ramp from 40°C to 300°C at 10°C/min, and the injection volume was 1 μL in split mode (split ratio 1:20). The injection temperature was kept at 300°C for 4 min. Mass spectra were generated in electron impact mode at 70 eV, at an ion-source temperature of 150°C. The mass spectra were recorded from 40 to 700 u at 1 s intervals. Sterols were identified by searching against the metabolite mass spectra database, comparing their spectra to commercially available authentic reference standards, and on the basis of relative retention times, as reported previously (Hughes et al.2000; Yun et al.2014). Relative quantification of sterols between the strains expressing the WT ERG28 and its truncations was conducted by comparing the integrated peak areas of each sterol intermediate, and the specific sterol yields were calculated per gram of dry cell weight (DCW). Statistical analysis For each experiment, three biological replicates were performed. Mean values and standard curves calculated from three different batches of data were used to calculate the contents of the individual sterol components. Differences between strains were analyzed using Student's t-test, and a P-value < 0.05 was used as the threshold for statistical significance. RESULTS AND DISCUSSION Disruption of ERG28 significantly inhibited the growth and ergosterol accumulation of BY4741 Erg28p is a key component of the yeast sterol biosynthesis complex (Mo and Bard 2005a), and it interacts with both the sterol C-4 demethylase complex and the late-stage sterol biosynthesis proteins (Mo, Valachovic and Bard 2004). The biological role of Erg28p in yeast was reconfirmed in our study. As shown in Fig. 1A, a deletion of erg28 resulted in a significant reduction of the specific growth rate under the same culture conditions, especially during the early period of the logarithmic growth stage. Correspondingly, the cell density of the erg28 strain was significantly decreased (75.4% of the WT in the late stationary phase up to 48 h). GC-MS analysis was further performed to analyze the differences of the free (nonsaponifiable) sterol profiles between the cell membranes of the erg28 and WT strains, since Erg28p is reported to be strongly co-regulated with yeast ergosterol biosynthesis (Hughes et al.2000). As shown in Fig. 1B, the accumulation of the end-product ergosterol was dramatically decreased in the erg28 strain (42.85% of the WT). Moreover, according to the respective retention times and structure predictions of the sterol intermediates analyzed by GC/MS, the main sterol fractions contained intermediates of sterol C4-demethylation such as lanosterol, 4α-methyl-zymosterol, 4α-methyl-zymosterone and 4,4-dimethylzymosterol (Table S2, Supporting Information). The sterol profile of the erg28 strain was slightly different in both variety and content from what was reported previously (Gachotte et al.2001), probably due to the different genetic background of the haploid BY4741 and the heterozygous BY4743 used to generate the two erg28 knockout strains. Furthermore, complementation of the erg28 strain with the ScERG28 plasmid restored the ergosterol biosynthesis pathway, as evidenced by the disappearance of all the biosynthetic precursors and the recovery of the end-product—ergosterol (Fig. 1B). Correspondingly, some studies demonstrated that a decrease of ergosterol synthesis in yeast may result in cell growth inhibition (Buurman et al.2004). Taken together, the results thus corroborate the pivotal role of Erg28p in the sterol biosynthesis pathway of yeast. Figure 1. View largeDownload slide Growth rate phenotypes and free-sterol profiles of the WT and erg28 strains. (A) The biomass concentration and specific growth rate (μ) of the WT and erg28 strains cultivated at 30°C and 250 rpm in YPD medium. (B) GC analysis comparing sterol accumulation between the erg28 strain complemented with a plasmid containing the ScERG28 gene, the WT and the erg28 strain. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, 4α-methyl-zymosterone; Peak 4, unidentified sterol (probably 4α-methylfecosterol); Peak 5, lanosterol; Peak 6, 4,4-dimethylzymosterol. Figure 1. View largeDownload slide Growth rate phenotypes and free-sterol profiles of the WT and erg28 strains. (A) The biomass concentration and specific growth rate (μ) of the WT and erg28 strains cultivated at 30°C and 250 rpm in YPD medium. (B) GC analysis comparing sterol accumulation between the erg28 strain complemented with a plasmid containing the ScERG28 gene, the WT and the erg28 strain. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, 4α-methyl-zymosterone; Peak 4, unidentified sterol (probably 4α-methylfecosterol); Peak 5, lanosterol; Peak 6, 4,4-dimethylzymosterol. As an ER transmembrane protein, Erg28p was shown to function as a scaffold to tether the C-4 demethylase complex to the ER (Mo et al.2002). Moreover, the direct interaction between Erg28p and Erg27p/Erg25p was confirmed by solubilized membrane protein co-immunoprecipitation as well as by a split-ubiquitin membrane protein yeast two-hybrid system (Mo and Bard 2005a). However, to the best of our knowledge, there is no direct evidence of an effect of Erg28p on the subcellular localization of Erg26p in the literature. Thus, Erg26p was visualized in both the WT and the erg28 strain by constructing a C-terminal EGPF fusion. The results of confocal laser microscopy demonstrated that the deletion of erg28 had no obvious effect on the ER localization of Erg26p, since it remained co-localized with the red ER-tracer in the erg28 strain as shown in Fig. 2. Previous reports demonstrated that Erg11p, Erg25p, Erg27p and Erg28p appear to form a core center that can interact with many other enzymes from the sterol biosynthesis pathway, and most, if not all, may be tethered to the ER as a large complex (Mo and Bard 2005b). New methodologies such as cryoelectron microscopy may provide a more feasible approach to unravel the mechanism by which the C-4 demethylase multienzyme system is assembled by Erg28p into ER membrane-localized structures (Fernandez-Leiro and Scheres 2016). Figure 2. View largeDownload slide Subcellular localization of Erg26p in the WT (A) and the erg28 (B) strain. Figure 2. View largeDownload slide Subcellular localization of Erg26p in the WT (A) and the erg28 (B) strain. C-terminal truncation of ERG28Δ136-273 leads to the accumulation of the intermediates of sterol C-4 demethylation ScErg28p is an ER-resident protein, and previous studies reported conflicting prediction results of its transmembrane domains based on four different databases (TMAP, MIPS, CBS and TSEG), and two or three transmembrane domains were proposed (Mo, Valachovic and Bard 2004). A recent report on an ERG28 orthologue from C. elegans, which was confirmed to regulate the synaptic function and alcohol response in addition to assisting the de novo sterol biosynthesis, predicated four transmembrane domains within the protein structure (Oh et al.2017). Although the speed of molecular evolution of ERG28 orthologues appears to vary, their positively charged character (isoelectric point > 9.0) is required for Erg28p protein function, and has consequently been maintained through a wide evolutionary span (Vinci, Xia and Veitia 2008). Since the 3D structure of Erg28p has not been resolved to date, its hydrophobic profile was analyzed in silico, and two independent predictions both indicated that ScErg28p is likely to possess four transmembrane domains (Fig. 3). In addition, a small signal peptide was also found at the N-terminus of ScErg28p (Fig. 4A), while it diverged in higher organisms, where it has an ER retention/retrieval motif at the C-terminus (Vinci, Xia and Veitia 2008). Accordingly, the truncation strategy was applied, and five sets of truncated strains were constructed (Fig. 4B). Significant differences in the sterol composition between the rescued strain and the ERG28-truncated strains were observed by GC/MS. As shown in Fig. 4C, relative accumulation of free ergosterol in the serially truncated ERG28 mutant strains decreased with the increasing length of the C-terminal truncations. Moreover, a deletion of the first 20 amino acids at the N-terminus also resulted in a significant reduction of ergosterol accumulation by nearly 50%. As shown in Fig. 4D, accumulation of typical ergosterol precursors including 3-hydroxysterols (4α-methyl-zymosterol and 4α-methylfecosterol), a 4-formyl hydroxysterol (4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol) and a carboxylic acid sterol (4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol) were observed in the erg28-ERG28Δ136-447 strain, all of which are well-known intermediates of C-4 demethylation (Fig. S1, Supporting Information; Parks and Casey 1995). Such intermediates are normally accumulated by ERG25 and ERG26 mutant strains (Bard et al.1996; Gachotte et al.1998), indicating that the transmembrane domain of Erg28p (46–90 residues) may play an important role in C4-demethylation. Figure 3. View largeDownload slide Prediction of transmembrane domains in ScErg28p based on the TMHMM (A) and Phyre 2 (B) databases. Figure 3. View largeDownload slide Prediction of transmembrane domains in ScErg28p based on the TMHMM (A) and Phyre 2 (B) databases. Figure 4. View largeDownload slide Analysis of sterol intermediates of ergosterol biosynthesis in a series of erg28 strains complemented with plasmids expressing serial C-terminal truncations of Erg28p. (A) ScErg28p is predicted to be an ER-resident protein comprising four transmembrane domains and a signal peptide at the N-terminus (red). (B) Scheme of the construction of five groups of ScErg28p mutants with serial deletions in the C-terminal domain based on the results from topology prediction. (C) Relative ergosterol accumulation in the WT strain (control) and erg28 strains complemented with the plasmids expressing serial C-terminal truncations of Erg28p. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. (D) Chromatographic profile of free sterols in the erg28 strain complemented with the plasmid expressing the truncation ERG28Δ136-447. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, unidentified sterol; Peak 4, 4α-methylfecosterol; Peak 5, 4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol; Peak 6, 4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol. Figure 4. View largeDownload slide Analysis of sterol intermediates of ergosterol biosynthesis in a series of erg28 strains complemented with plasmids expressing serial C-terminal truncations of Erg28p. (A) ScErg28p is predicted to be an ER-resident protein comprising four transmembrane domains and a signal peptide at the N-terminus (red). (B) Scheme of the construction of five groups of ScErg28p mutants with serial deletions in the C-terminal domain based on the results from topology prediction. (C) Relative ergosterol accumulation in the WT strain (control) and erg28 strains complemented with the plasmids expressing serial C-terminal truncations of Erg28p. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. (D) Chromatographic profile of free sterols in the erg28 strain complemented with the plasmid expressing the truncation ERG28Δ136-447. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, unidentified sterol; Peak 4, 4α-methylfecosterol; Peak 5, 4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol; Peak 6, 4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol. A consensus sequence motif (63LS/QARTFGT/LWT72) is required for ergosterol biosynthesis Amino acid sequence alignment of ScErg28p with its homologs from other organisms was carried out and the results are shown in Fig. S2, Supporting Information. A BLASTn matrix assay demonstrated that the alignment scores of ScErg28p to its homologs from H. sapiens, P. troglodytes and C. albicans corresponded to 18.9% identity and 64.1% sequence similarity. Especially, each amino acid sequence of the Erg28p homologs from the four species contained a motif at positions 63–72, which was almost identical to the LS/QARTFGT/LWT consensus, and was located in the third truncated domain (ERG28Δ136-273). Moreover, results from the extensive BLASTn matrix analysis further confirmed the presence of the consensus motif 63LS/QARTFGT/LWT72 in all homologs (data not shown), indicating that the consensus sequence may be located within a region that is important for the role of Erg28p in sterol C-4 demethylation. Thus, a targeted deletion of the putative consensus motif was carried out, and the resulting ERG28 mutant (ScERG28Δ175-204) was used to complement the erg28-disrupted strain. The growth phenotype and the profile of sterol metabolic intermediates, especially the relative accumulation of ergosterol, were investigated in strains with different ERG28 genetic backgrounds, including WT, disrupted (erg28), restored (erg28-ScERG28), and mutant (erg28-ScERG28Δ175-204) strains. As shown in Fig. 5A, the spot plate analysis revealed that the number of total viable cells of the disrupted and mutant strains in the unit area of the plate was significantly smaller than that of the WT and restored strains at dilutions of 10−5 and 10−6. Compared with the relatively higher yield of ergosterol in BY4741 cells (8.89 ± 0.33 mg/g DCW), the deletion of erg28 resulted in a significant decrease (3.81 ± 0.17 mg/g DCW). Moreover, the yield of ergosterol in the recovered strain erg28-ScERG28 reached 12.51 ± 0.53 mg/g DCW, while a further deletion of the consensus motif in the erg28-ScERG28Δ175-204 strain resulted in an obvious decrease to 1.25 ± 0.46 mg/g DCW. Considering the complex interactions between Erg28p and other components involved in C-4 demethylation, there may be a number of proteins and interfaces that are not limited to these 10 amino acids. Thus, studies providing direct evidence of multiprotein interactions, such as polyprotein mass spectrometry, are urgently needed. Figure 5. View largeDownload slide Comparison of the growth rate and relative ergosterol accumulation in BY4741 strains with different ERG28 genetic backgrounds, including WT, erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204. (A) Spot plate assay for growth of different strains on CSM-URA medium ate 30°C. Serial dilutions of 106–101 cells were spotted onto the indicated plate areas. (B) Relative ergosterol accumulation in the WT (control), erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204 strains. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. Figure 5. View largeDownload slide Comparison of the growth rate and relative ergosterol accumulation in BY4741 strains with different ERG28 genetic backgrounds, including WT, erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204. (A) Spot plate assay for growth of different strains on CSM-URA medium ate 30°C. Serial dilutions of 106–101 cells were spotted onto the indicated plate areas. (B) Relative ergosterol accumulation in the WT (control), erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204 strains. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. Disruption of the four parallel helices of ScErg28p by the deletion of the 63LS/QARTFGT/LWT72 sequence Analysis of protein structures with multiple transmembrane domains by classical crystallographic methods is fraught with significant difficulties (Lacapere et al.2007). Since the 3D structure of Erg28p had not been reported at the time of this study, homology-based modeling of both WT and mutant ScErg28p with the 63LS/QARTFGT/LWT72 sequence deletion was conducted based on the reported crystal structures of similar proteins. Based on the TM-align structure calibration procedure, a I-TASSER model was matched to all the structures in the PDB library, and 10 proteins with high structural similarity were identified. The protein structure with the highest TM-score for Erg28p is shown in Fig. 6A. The template (PDB ID: 5H1Q) is a transport protein from Caenorhabditis elegans that is located in the worm's INX-6 gap junction hemichannel (Oshima, Tani and Fujiyoshi 2016). It has been observed that Erg28p has four parallel α-helix structures with an additional short α-helix as a signal peptide (Oh et al.2017), which is consistent with the results of the prediction of hydrophobic helices. The structure of the erg28-ScERG28Δ175-204 mutant was also modeled in the same way (Fig. 6B and C), using the C. albicans transceptor Mep2, a protein that plays an important role in fungal development as an ammonium sensor, as the template (PDB ID: 5AEZ) (Van Den Berg et al.2016). As shown in Fig. 6B, the original parallel spiral structure of the four α-helixes was severely damaged, resulting in irregular random coils, leaving only a short region of α-helix within the third transmembrane domain. As shown in Fig. 6C, although the four α-helix structures were present in the ScERG28Δ175-204 mutant, the folding, stretching angle and length of the four domains were obviously altered. These results suggest that the deletion of the residues spanning the region from 63 to 72 seriously affected the topological structure of Erg28p, and in turn almost certainly affected its biological function. Figure 6. View largeDownload slide Three-dimensional structure prediction of ScErg28p by homology modeling. (A) Homology model of WT Erg28p generated using its homolog from C. elegans (PDB ID: 5H1Q) as template. (B) Models of two recommended 3D structures for the Erg28pΔ63-72 mutant constructed based on the template protein (PDB ID: 5AEZ) using the I-TASSER online server. The five parallel helix domains are shown in red (1–20), green (21–44), blue (45–90), yellow (91–113) and bright blue (113–148), respectively. Figure 6. View largeDownload slide Three-dimensional structure prediction of ScErg28p by homology modeling. (A) Homology model of WT Erg28p generated using its homolog from C. elegans (PDB ID: 5H1Q) as template. (B) Models of two recommended 3D structures for the Erg28pΔ63-72 mutant constructed based on the template protein (PDB ID: 5AEZ) using the I-TASSER online server. The five parallel helix domains are shown in red (1–20), green (21–44), blue (45–90), yellow (91–113) and bright blue (113–148), respectively. In summary, our study identified a novel consensus motif sequence (63LS/QARTFGT/LWT72) that significantly correlates with the phenotypes of yeast growth, ergosterol biosynthesis and correct folding of four parallel α-helixes in the three-dimensional structure of Erg28p, confirming its biofunctional role in sterol C-4 demethylation in the ergosterol biosynthesis pathway of S. cerevisiae. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 31400978) and the Natural Science Foundation of Zhejiang Province (No. LY18B020019). Conflict of interest. None declared. REFERENCES Adams BG, Parks LW. Isolation from yeast of a metabolically active water-soluble form of ergosterol. J Lipid Res  1968; 9: 8– 11. Google Scholar PubMed  Bard M, Bruner DA, Pierson CA et al.   Cloning and characterization of ERG25, the Saccharomyces cerevisiae gene encoding C-4 sterol methyl oxidase. P Natl Acad Sci USA  1996; 93: 186– 90. Google Scholar CrossRef Search ADS   Bloxham DP, Wilton DC, Akhtar M. Studies on the mechanism and regulation of C-4 demethylation in cholesterol biosynthesis. The role of adenosine 3':5'-cyclic monophosphate. Biochem J  1971; 125: 625– 34. Google Scholar CrossRef Search ADS PubMed  Buurman ET, Blodgett AE, Hull KG et al.   Pyridines and pyrimidines mediating activity against an efflux-negative strain of Candida albicans through putative inhibition of lanosterol demethylase. Antimicrob Agents Ch  2004; 48: 313– 8. Google Scholar CrossRef Search ADS   Fernandez-Leiro R, Scheres SH. Unravelling biological macromolecules with cryo-electron microscopy. Nature  2016; 537: 339– 46. Google Scholar CrossRef Search ADS PubMed  Gachotte D, Barbuch R, Gaylor J et al.   Characterization of the Saccharomyces cerevisiae ERG26 gene encoding the C-3 sterol dehydrogenase (C-4 decarboxylase) involved in sterol biosynthesis. P Natl Acad Sci USA  1998; 95: 13794– 9. Google Scholar CrossRef Search ADS   Gachotte D, Eckstein J, Barbuch R et al.   A novel gene conserved from yeast to humans is involved in sterol biosynthesis. J Lipid Res  2001; 42: 150– 4. Google Scholar PubMed  Gachotte D, Sen SE, Eckstein J et al.   Characterization of the Saccharomyces cerevisiae ERG27 gene encoding the 3-keto reductase involved in C-4 sterol demethylation. P Natl Acad Sci USA  1999; 96: 12655– 60. Google Scholar CrossRef Search ADS   Herrmann JM, Westermann B, Neupert W. Analysis of protein-protein interactions in mitochondria by coimmunoprecipitation and chemical cross-linking. Methods Cell Biol  2001; 65: 217– 30. Google Scholar CrossRef Search ADS PubMed  Hughes TR, Marton MJ, Jones AR et al.   Functional discovery via a compendium of expression profiles. Cell  2000; 102: 109– 26. Google Scholar CrossRef Search ADS PubMed  Kelley LA, Sternberg MJ. Protein structure prediction on the Web: a case study using the Phyre server. Nat Protoc  2009; 4: 363– 71. Google Scholar CrossRef Search ADS PubMed  Lacapere JJ, Pebay-Peyroula E, Neumann JM et al.   Determining membrane protein structures: still a challenge! Trends Biochem Sci  2007; 32: 259– 70. Google Scholar CrossRef Search ADS PubMed  Mialoundama AS, Jadid N, Brunel J et al.   Arabidopsis ERG28 tethers the sterol C4-demethylation complex to prevent accumulation of a biosynthetic intermediate that interferes with polar auxin transport. Plant Cell  2013; 25: 4879– 93. Google Scholar CrossRef Search ADS PubMed  Mo C, Bard M. Erg28p is a key protein in the yeast sterol biosynthetic enzyme complex. J Lipid Res  2005a; 46: 1991– 8. Google Scholar CrossRef Search ADS   Mo C, Bard M. A systematic study of yeast sterol biosynthetic protein-protein interactions using the split-ubiquitin system. BBA-Mol Cell Biol L  2005b; 1737: 152– 60. Google Scholar CrossRef Search ADS   Mo C, Valachovic M, Bard M. The ERG28-encoded protein, Erg28p, interacts with both the sterol C-4 demethylation enzyme complex as well as the late biosynthetic protein, the C-24 sterol methyltransferase (Erg6p). BBA-Mol Cell Biol L  2004; 1686: 30– 6. Google Scholar CrossRef Search ADS   Mo C, Valachovic M, Randall SK et al.   Protein-protein interactions among C-4 demethylation enzymes involved in yeast sterol biosynthesis. P Natl Acad Sci USA  2002; 99: 9739– 44. Google Scholar CrossRef Search ADS   Niedenthal RK, Riles L, Johnston M et al.   Green fluorescent protein as a marker for gene expression and subcellular localization in budding yeast. Yeast  1996; 12: 773– 86. Google Scholar CrossRef Search ADS PubMed  Oh KH, Haney JJ, Wang X et al.   ERG-28 controls BK channel trafficking in the ER to regulate synaptic function and alcohol response in C. elegans. Elife  2017; 6: e24733. Google Scholar CrossRef Search ADS PubMed  Oshima A, Tani K, Fujiyoshi Y. Atomic structure of the innexin-6 gap junction channel determined by cryo-EM. Nat Commun  2016; 7: 13681. Google Scholar CrossRef Search ADS PubMed  Parks LW, Casey WM. Physiological implications of sterol biosynthesis in yeast. Annu Rev Microbiol  1995; 49: 95– 116. Google Scholar CrossRef Search ADS PubMed  Rahier A. Dissecting the sterol C-4 demethylation process in higher plants. From structures and genes to catalytic mechanism. Steroids  2011; 76: 340– 52. Google Scholar CrossRef Search ADS PubMed  Rahier A, Darnet S, Bouvier F et al.   Molecular and enzymatic characterizations of novel bifunctional 3beta-hydroxysteroid dehydrogenases/C-4 decarboxylases from Arabidopsis thaliana. J Biol Chem  2006; 281: 27264– 77. Google Scholar CrossRef Search ADS PubMed  Sheff MA, Thorn KS. Optimized cassettes for fluorescent protein tagging in Saccharomyces cerevisiae. Yeast  2004; 21: 661– 70. Google Scholar CrossRef Search ADS PubMed  Sipiczki M. Fission yeast phylogenesis and evolution. In: The Molecular Biology of Schizosaccharomyces Pombe . Berlin, Heidelberg: Springer, 2004, 431– 43. Google Scholar CrossRef Search ADS   Thaminy S, Auerbach D, Arnoldo A et al.   Identification of novel ErbB3-interacting factors using the split-ubiquitin membrane yeast two-hybrid system. Genome Res  2003; 13: 1744– 53. Google Scholar CrossRef Search ADS PubMed  Van Den Berg B, Chembath A, Jefferies D et al.   Structural basis for Mep2 ammonium transceptor activation by phosphorylation. Nat Commun  2016; 7: 11337. Google Scholar CrossRef Search ADS PubMed  Veitia RA, Hurst LD. Accelerated molecular evolution of insect orthologues of ERG28/C14orf1: a link with ecdysteroid metabolism? J Genet  2001; 80: 17– 21. Google Scholar CrossRef Search ADS PubMed  Vinci G, Xia XH, Veitia RA. Preservation of genes involved in sterol metabolism in cholesterol auxotrophs: facts and hypotheses. PLoS One  2008; 3: e2883. Google Scholar CrossRef Search ADS PubMed  Wilton DC, Akhtar M. The mechanism of C-4 demethylation during cholesterol biosynthesis. Biochem J  1975; 149: 233– 5. Google Scholar CrossRef Search ADS PubMed  Yun Y, Yin D, Dawood DH et al.   Functional characterization of FgERG3 and FgERG5 associated with ergosterol biosynthesis, vegetative differentiation and virulence of Fusarium graminearum. Fungal Genet Biol  2014; 68: 60– 70. Google Scholar CrossRef Search ADS PubMed  © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Letters Oxford University Press

Identification of a consensus motif in Erg28p required for C-4 demethylation in yeast ergosterol biosynthesis based on mutation analysis

Loading next page...
1
 
/lp/ou_press/identification-of-a-consensus-motif-in-erg28p-required-for-c-4-hdiQvyo5Qz

References (35)

Publisher
Oxford University Press
Copyright
© FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0378-1097
eISSN
1574-6968
DOI
10.1093/femsle/fny002
Publisher site
See Article on Publisher Site

Abstract

Abstract The Erg28p protein is localized to the endoplasmic reticulum, where it acts as a scaffold to tether the C-4 demethylase complex involved in the sterol biosynthesis pathway of Saccharomyces cerevisiae. However, due to the challenges involved in characterizing the interactions of membrane proteins, the precise region of Erg28p that is responsible for the assembly of this enzyme complex remains unknown. To address this question, mutants with serial truncations in the C-terminus of Erg28p were constructed based on a topology prediction of its transmembrane domain. Sterol profiles demonstrated that intermediates involved in the stepwise removal of the two C-4 methyl groups from the tetracyclic sterol ring were accumulated in the ERG28Δ135-447 strain. Homologous alignment of Erg28p further identified a highly conserved 10-amino acid sequence (63LS/QARTFGT/LWT72) within the truncated region of ERG28Δ136-273. Complementation of the BY4741/erg28 strain with the ScERG28Δ175-204 plasmid resulted both in a significant growth inhibition and a reduction of ergosterol biosynthesis compared with the plasmid without the Δ175-204 truncation. Furthermore, homology modeling of the Erg28p mutant indicated that the deletion of residues 63–72 significantly disrupted the 3D structure of the four parallel helices in Erg28p. Taken together, the data indicate that the region spanning amino acids 63–72 constitutes a key consensus motif within Erg28p that is required for sterol C-4 demethylation during ergosterol biosynthesis in S. cerevisiae. C-4 demethylation complex, Erg28p; ergosterol synthesis pathway, conserved motif INTRODUCTION Ergosterol is the predominant sterol constituent of the plasma membrane of S. cerevisiae, and its biosynthetic pathway comprising the ERG gene family members is now well understood (Veitia and Hurst 2001). The sterol molecule becomes functional only after removal of the two nuclear methyl groups at the C-4 position, which involves some of the most complicated processes that have been identified in the sterol biosynthetic pathways of eukaryotic organisms (Bloxham, Wilton and Akhtar 1971; Rahier et al.2006; Rahier 2011). In yeast, the C-4 demethylation process involves a series of three redox reactions, catalyzed by the C-4 oxidase (SMO) encoded by ERG25 (Bard et al.1996), C-3 decarboxylase (3βHSD/D) encoded by ERG26 (Gachotte et al.1998) and C-3 ketoreductase (3SR) encoded by ERG27 (Gachotte et al.1999), respectively. The C-4 methyl group is sequentially oxidized, decarboxylated, ortho-ketonized and reduced to a hydroxyl group, so that after two rounds of demethylation, the C-4 dimethyl is finally removed (Wilton and Akhtar 1975; Rahier et al.2006). In addition to the well-defined catalytic mechanisms of the S. cerevisiae ergosterol biosynthesis pathway, the endoplasmic reticulum (ER) transmembrane protein Erg28p, the only known member of the ERG pathway lacking enzymatic activity, is highly conserved across eukaryotes, and homologs of the ERG28 gene have also been observed in Schizosaccharomyces pombe (Sipiczki 2004), Arabidopsis (Mialoundama et al.2013) and C. elegans (Oh et al.2017). This protein was firstly found to be strongly co-regulated with ergosterol biosynthesis, after which its protein–protein interactions with C-4 demethylation enzymes were confirmed (Mo et al.2002; Mialoundama et al.2013). All four components of the sterol C-4 demethylase complex contain potential transmembrane domains, and the role of Erg28p may be either to tether Erg25p, Erg26p and Erg27p to the ER, or to facilitate the interaction between these proteins, acting as a scaffold promoting the co-localization of the ERG enzymes (Gachotte et al.2001). In addition to the critical role in anchoring the C-4 demethylase complex to the ER, Erg28p was also confirmed to be associated with other ERG members such as Erg11p, Erg6p and Erg1p, demonstrating its crucial role in the ergosterol-biosynthetic enzyme complex (Mo, Valachovic and Bard 2004; Mo and Bard 2005a,b). The identification and characterization of protein–protein interactions between transmembrane proteins is highly challenging because of the physicochemically diverse properties of such proteins and the complexity of the enzymatic reactions associated with the ER membrane (Lacapere et al.2007). Typical biochemical methods, such as co-immunoprecipitation, co-purification or cross-linking, have been used to investigate protein complexes (Herrmann, Westermann and Neupert 2001). Alternative genetic methods such as the split-ubiquitin membrane yeast two-hybrid system also offer powerful tools to understand the protein–protein interactions involved in ergosterol synthesis (Mo and Bard 2005b). However, due to the hydrophobic nature of transmembrane proteins as well as the fact that integral and membrane-associated proteins undergo complex post-translational modifications or oligomerize via interactions between their transmembrane domains, there still are significant challenges in determining the interactions of transmembrane proteins (Thaminy et al.2003). Thus, the precise interacting region in the protein complex organized by Erg28p remains to be investigated further. In the present study, we proposed the sequence extent of the hydrophobic transmembrane domain of S. cerevisiae Erg28p (ScErg28p). Correspondingly, a series of ERG28 mutants with sequential truncations was constructed, and the sterol profiles of strains expressing these mutants were analyzed. Based on the phenotype data of the mutant strains and a structure prediction based on homology-based modeling, a highly conserved consensus motif (63LS/QARTFGT/LWT72) was identified within Erg28p, which is required for the sterol C-4 demethylation reaction involved in ergosterol biosynthesis. MATERIALS AND METHODS Strains, culture conditions and reagents The BY4741 wild-type (WT) strain of S. cerevisiae (ΔMAT, his3Δ, leu2Δ, met15Δ, ura3Δ) and the ERG28 knockout strain Y00177 (erg28) (BY4741; MATa; ura3Δ0; leu2Δ0; his3Δ1; met15Δ0; YER044c::kanMX4) were obtained from Euroscarf (Institute for Molecular Biosciences, Frankfurt, Germany) (http://www.euroscarf.de/), and were grown at 30°C for approximately 18–20 h in YPD medium consisting of 1% yeast extract (Oxoid Ltd, England), 2% Bacto peptone (Becton, Dickinson and Company, France) and 2% glucose (Aladdin Corp., China). Yeast transformants containing plasmids expressing either WT ERG28 or its truncated mutants were grown in complete synthetic uracil-deficient medium (CSM-URA), comprising 0.67% yeast nitrogen base, 2% glucose and 0.8 g/L uracil dropout amino acid mixture (amino acid and nitrogen base supplements; Aladdin Corp.). The yeast transformant carrying the EGFP-ERG26 plasmid was grown in complete synthetic histidine-deficient medium (CSM-HIS), comprising 0.67% yeast nitrogen base, 2% glucose and 0.8 g/L histidine dropout amino acid mixture. All yeast strains were preserved at 4°C on glucose-yeast peptone agar slants. Plasmid DNA was first used to transform Escherichia coli BL21 cells (Promega Corp., USA), which were grown in LB medium comprising 0.5% yeast extract, 1% Bacto peptone, 1% NaCl and 50 mg/L ampicillin at 30°C. The sequences of the oligonucleotide primers and of the ScERG28Δ175-204 mutant were chemically synthesized by Sangon Biotech (Shanghai, China). ExPfu DNA polymerase, restriction enzymes, Tango buffer and T4 ligase were purchased from Thermo-Fisher Scientific (USA). Authentic reference standards of the sterol substrates ergosterol and lanosterol, as well as the antibiotics G418 and ampicillin were purchased from Sigma-Aldrich Corporation (USA). All other chemicals used in this study were from commercial sources and of reagent grade. Molecular cloning of ERG28 and its truncated mutants The ERG28 gene and its truncated mutants were amplified from the chromosomal DNA of S. cerevisiae BY4741. Restriction sites for HindIII and NotI were introduced using the primers listed in Table S1, Supporting Information. The PCR procedure was performed using Taq DNA polymerase with a total number of 30 cycles. Each cycle included three steps: 94°C for 30 s, 57°C–61°C (according to the primers) for 30 s, and 72°C for 90 s. The PCR products were purified and ligated into the pGEM-T Easy vector, which was then digested using the corresponding restriction enzymes. The digested products were ligated into the pYES2 plasmid and used to transform E. coli BL21 competent cells. The positive colonies were confirmed both by bacterial colony-PCR and by restriction-enzyme double digestion. The confirmed recombinant plasmids were extracted from E. coli and used to transform the erg28 strain using the PEG-LiAc transformation method (Sheff and Thorn 2004), yielding ERG28-WT, ERG28-T1, ERG28-T2, ERG28-T3, ERG28-T4 and ERG28-T5, respectively (Fig. 4B). Construction of the EGFP-fusion plasmid and fluorescence microscopy All fluorescence microscopy procedures were performed as described previously (Niedenthal et al.1996). The ERG26 gene with its endogenous promoter was cloned by PCR using the corresponding primers listed in Table S1, Supporting Information. The fragments were gel-purified and ligated into the pKT128 expression vector to generate the EGFP-fusion plasmid pKT128-ERG26. For the microscopy assay, the yeast strains BY4741 and BY4741/erg28 were transformed with pKT128-ERG26 and grown to the exponential phase in liquid CSM-HIS medium or on CSM-HIS plates. For staining of mitochondria and nuclei in living cells, cells grown to the exponential phase were incubated with 4΄,6-diamidino-2-phenylindole dihydrochloride (1:100; Beyotime Institute of Biotechnology Jiangsu, China) and red ER tracker (1:1000; also from Beyotime) for 1 h at room temperature in the dark. Cells were fixed onto a polylysine-coated dish (Thermo Scientific Nunc®) for automated fluorescence imaging using a Zeiss Axioskop fluorescence microscope (Carl Zeiss Microscopy LLC, USA). Spot test analysis of cell growth Serial dilutions comprising 10 μL drops with 106–101 cells of the strains with different ERG28 genetic backgrounds in the early exponential growth phase were spotted onto YPD plates and incubated for 24 h at 30°C, after which the cell growth was analyzed visually and the plates photographed for documentation. The experiment was repeated three times. Sequence alignment, prediction of hydrophobic helices and homology-based structure modeling Alignment of the amino acid sequences encoded by the ERG28 genes from S. cerevisiae (NP_010962.1), Homo sapiens (NP_009107.1), Pan troglodytes (GenBank: JAA12605.1) and Candida albicans (XP_719465.1) was conducted using Vector NTI Advance 11.5 (Invitrogen, Darmstadt, Germany). Extensive BLASTn matrix analysis of a large number of putative Erg28p sequences was processed with Clustal Omega (www.clustal.org). Prediction of hydrophobic helices in Erg28p was conducted using the TMHMM server v. 2.0 (www.cbs.dtu.dk/services/TMHMM/) and the protein-fold recognition server Phyre 2 (Kelley and Sternberg 2009). Prediction of the tertiary structure in both Erg28p and the truncated Erg28p mutants was conducted using the I-TASSER on-line server (Fernandez-Leiro and Scheres 2016), using the protein structures 5H1Q and 5AEZ from the PDB database (www.wwpdb.org) as the templates for Erg28p and the mutant Erg28pΔ63-72, respectively. Extraction of sterols and gas chromatography/mass spectrometry (GC/MS) analysis Sterols were extracted after alkaline saponification in the presence of pyrogallol according to a classical method (Adams and Parks 1968). The yeast cells were collected by centrifugation at 3000× g and 4°C for 10 min, washed twice with distilled water and dried at 80°C to a constant weight. Samples comprising 300 mg of the dried biomass were resuspended in reaction mixtures comprising 5 mL of a 60% KOH solution, 7.5 mL of methanol and 7.5 mL of methanolic pyrogallol (0.5% w/v), and incubated at 80°C for 4 h under constant shaking at 180 rpm. The unsaponified matter was extracted with 10 mL of hexane. Different phases were formed after centrifugation at 3000× g for 10 min. The top n-hexane layer was removed to a clean tube and the residue was re-extracted twice with hexane. The n-hexane phase containing the extracted sterols was recovered, dried under nitrogen and the final obtained sterols were resuspended in 500 μL of chloroform/methanol (4:1) for GC/MS analysis. The sterols were separated and analyzed by GC-MS on an Agilent 3400 system (Agilent, USA) equipped with a fused silica column (DB-5, 15 m × 0.32 mm × 0.25 μm film thickness; J&W Scientific, USA), with helium at a flow rate of 1 mL/min as the carrier gas. The temperature program comprised a ramp from 40°C to 300°C at 10°C/min, and the injection volume was 1 μL in split mode (split ratio 1:20). The injection temperature was kept at 300°C for 4 min. Mass spectra were generated in electron impact mode at 70 eV, at an ion-source temperature of 150°C. The mass spectra were recorded from 40 to 700 u at 1 s intervals. Sterols were identified by searching against the metabolite mass spectra database, comparing their spectra to commercially available authentic reference standards, and on the basis of relative retention times, as reported previously (Hughes et al.2000; Yun et al.2014). Relative quantification of sterols between the strains expressing the WT ERG28 and its truncations was conducted by comparing the integrated peak areas of each sterol intermediate, and the specific sterol yields were calculated per gram of dry cell weight (DCW). Statistical analysis For each experiment, three biological replicates were performed. Mean values and standard curves calculated from three different batches of data were used to calculate the contents of the individual sterol components. Differences between strains were analyzed using Student's t-test, and a P-value < 0.05 was used as the threshold for statistical significance. RESULTS AND DISCUSSION Disruption of ERG28 significantly inhibited the growth and ergosterol accumulation of BY4741 Erg28p is a key component of the yeast sterol biosynthesis complex (Mo and Bard 2005a), and it interacts with both the sterol C-4 demethylase complex and the late-stage sterol biosynthesis proteins (Mo, Valachovic and Bard 2004). The biological role of Erg28p in yeast was reconfirmed in our study. As shown in Fig. 1A, a deletion of erg28 resulted in a significant reduction of the specific growth rate under the same culture conditions, especially during the early period of the logarithmic growth stage. Correspondingly, the cell density of the erg28 strain was significantly decreased (75.4% of the WT in the late stationary phase up to 48 h). GC-MS analysis was further performed to analyze the differences of the free (nonsaponifiable) sterol profiles between the cell membranes of the erg28 and WT strains, since Erg28p is reported to be strongly co-regulated with yeast ergosterol biosynthesis (Hughes et al.2000). As shown in Fig. 1B, the accumulation of the end-product ergosterol was dramatically decreased in the erg28 strain (42.85% of the WT). Moreover, according to the respective retention times and structure predictions of the sterol intermediates analyzed by GC/MS, the main sterol fractions contained intermediates of sterol C4-demethylation such as lanosterol, 4α-methyl-zymosterol, 4α-methyl-zymosterone and 4,4-dimethylzymosterol (Table S2, Supporting Information). The sterol profile of the erg28 strain was slightly different in both variety and content from what was reported previously (Gachotte et al.2001), probably due to the different genetic background of the haploid BY4741 and the heterozygous BY4743 used to generate the two erg28 knockout strains. Furthermore, complementation of the erg28 strain with the ScERG28 plasmid restored the ergosterol biosynthesis pathway, as evidenced by the disappearance of all the biosynthetic precursors and the recovery of the end-product—ergosterol (Fig. 1B). Correspondingly, some studies demonstrated that a decrease of ergosterol synthesis in yeast may result in cell growth inhibition (Buurman et al.2004). Taken together, the results thus corroborate the pivotal role of Erg28p in the sterol biosynthesis pathway of yeast. Figure 1. View largeDownload slide Growth rate phenotypes and free-sterol profiles of the WT and erg28 strains. (A) The biomass concentration and specific growth rate (μ) of the WT and erg28 strains cultivated at 30°C and 250 rpm in YPD medium. (B) GC analysis comparing sterol accumulation between the erg28 strain complemented with a plasmid containing the ScERG28 gene, the WT and the erg28 strain. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, 4α-methyl-zymosterone; Peak 4, unidentified sterol (probably 4α-methylfecosterol); Peak 5, lanosterol; Peak 6, 4,4-dimethylzymosterol. Figure 1. View largeDownload slide Growth rate phenotypes and free-sterol profiles of the WT and erg28 strains. (A) The biomass concentration and specific growth rate (μ) of the WT and erg28 strains cultivated at 30°C and 250 rpm in YPD medium. (B) GC analysis comparing sterol accumulation between the erg28 strain complemented with a plasmid containing the ScERG28 gene, the WT and the erg28 strain. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, 4α-methyl-zymosterone; Peak 4, unidentified sterol (probably 4α-methylfecosterol); Peak 5, lanosterol; Peak 6, 4,4-dimethylzymosterol. As an ER transmembrane protein, Erg28p was shown to function as a scaffold to tether the C-4 demethylase complex to the ER (Mo et al.2002). Moreover, the direct interaction between Erg28p and Erg27p/Erg25p was confirmed by solubilized membrane protein co-immunoprecipitation as well as by a split-ubiquitin membrane protein yeast two-hybrid system (Mo and Bard 2005a). However, to the best of our knowledge, there is no direct evidence of an effect of Erg28p on the subcellular localization of Erg26p in the literature. Thus, Erg26p was visualized in both the WT and the erg28 strain by constructing a C-terminal EGPF fusion. The results of confocal laser microscopy demonstrated that the deletion of erg28 had no obvious effect on the ER localization of Erg26p, since it remained co-localized with the red ER-tracer in the erg28 strain as shown in Fig. 2. Previous reports demonstrated that Erg11p, Erg25p, Erg27p and Erg28p appear to form a core center that can interact with many other enzymes from the sterol biosynthesis pathway, and most, if not all, may be tethered to the ER as a large complex (Mo and Bard 2005b). New methodologies such as cryoelectron microscopy may provide a more feasible approach to unravel the mechanism by which the C-4 demethylase multienzyme system is assembled by Erg28p into ER membrane-localized structures (Fernandez-Leiro and Scheres 2016). Figure 2. View largeDownload slide Subcellular localization of Erg26p in the WT (A) and the erg28 (B) strain. Figure 2. View largeDownload slide Subcellular localization of Erg26p in the WT (A) and the erg28 (B) strain. C-terminal truncation of ERG28Δ136-273 leads to the accumulation of the intermediates of sterol C-4 demethylation ScErg28p is an ER-resident protein, and previous studies reported conflicting prediction results of its transmembrane domains based on four different databases (TMAP, MIPS, CBS and TSEG), and two or three transmembrane domains were proposed (Mo, Valachovic and Bard 2004). A recent report on an ERG28 orthologue from C. elegans, which was confirmed to regulate the synaptic function and alcohol response in addition to assisting the de novo sterol biosynthesis, predicated four transmembrane domains within the protein structure (Oh et al.2017). Although the speed of molecular evolution of ERG28 orthologues appears to vary, their positively charged character (isoelectric point > 9.0) is required for Erg28p protein function, and has consequently been maintained through a wide evolutionary span (Vinci, Xia and Veitia 2008). Since the 3D structure of Erg28p has not been resolved to date, its hydrophobic profile was analyzed in silico, and two independent predictions both indicated that ScErg28p is likely to possess four transmembrane domains (Fig. 3). In addition, a small signal peptide was also found at the N-terminus of ScErg28p (Fig. 4A), while it diverged in higher organisms, where it has an ER retention/retrieval motif at the C-terminus (Vinci, Xia and Veitia 2008). Accordingly, the truncation strategy was applied, and five sets of truncated strains were constructed (Fig. 4B). Significant differences in the sterol composition between the rescued strain and the ERG28-truncated strains were observed by GC/MS. As shown in Fig. 4C, relative accumulation of free ergosterol in the serially truncated ERG28 mutant strains decreased with the increasing length of the C-terminal truncations. Moreover, a deletion of the first 20 amino acids at the N-terminus also resulted in a significant reduction of ergosterol accumulation by nearly 50%. As shown in Fig. 4D, accumulation of typical ergosterol precursors including 3-hydroxysterols (4α-methyl-zymosterol and 4α-methylfecosterol), a 4-formyl hydroxysterol (4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol) and a carboxylic acid sterol (4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol) were observed in the erg28-ERG28Δ136-447 strain, all of which are well-known intermediates of C-4 demethylation (Fig. S1, Supporting Information; Parks and Casey 1995). Such intermediates are normally accumulated by ERG25 and ERG26 mutant strains (Bard et al.1996; Gachotte et al.1998), indicating that the transmembrane domain of Erg28p (46–90 residues) may play an important role in C4-demethylation. Figure 3. View largeDownload slide Prediction of transmembrane domains in ScErg28p based on the TMHMM (A) and Phyre 2 (B) databases. Figure 3. View largeDownload slide Prediction of transmembrane domains in ScErg28p based on the TMHMM (A) and Phyre 2 (B) databases. Figure 4. View largeDownload slide Analysis of sterol intermediates of ergosterol biosynthesis in a series of erg28 strains complemented with plasmids expressing serial C-terminal truncations of Erg28p. (A) ScErg28p is predicted to be an ER-resident protein comprising four transmembrane domains and a signal peptide at the N-terminus (red). (B) Scheme of the construction of five groups of ScErg28p mutants with serial deletions in the C-terminal domain based on the results from topology prediction. (C) Relative ergosterol accumulation in the WT strain (control) and erg28 strains complemented with the plasmids expressing serial C-terminal truncations of Erg28p. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. (D) Chromatographic profile of free sterols in the erg28 strain complemented with the plasmid expressing the truncation ERG28Δ136-447. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, unidentified sterol; Peak 4, 4α-methylfecosterol; Peak 5, 4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol; Peak 6, 4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol. Figure 4. View largeDownload slide Analysis of sterol intermediates of ergosterol biosynthesis in a series of erg28 strains complemented with plasmids expressing serial C-terminal truncations of Erg28p. (A) ScErg28p is predicted to be an ER-resident protein comprising four transmembrane domains and a signal peptide at the N-terminus (red). (B) Scheme of the construction of five groups of ScErg28p mutants with serial deletions in the C-terminal domain based on the results from topology prediction. (C) Relative ergosterol accumulation in the WT strain (control) and erg28 strains complemented with the plasmids expressing serial C-terminal truncations of Erg28p. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. (D) Chromatographic profile of free sterols in the erg28 strain complemented with the plasmid expressing the truncation ERG28Δ136-447. Peak 1, ergosterol; Peak 2, 4α-methyl-zymosterol; Peak 3, unidentified sterol; Peak 4, 4α-methylfecosterol; Peak 5, 4α-carboxy-4β-methyl-5α-cholesta-8,24-dien-3β-ol; Peak 6, 4α-formyl-4β-methyl-5α-cholesta-8,24-dien-3β-ol. A consensus sequence motif (63LS/QARTFGT/LWT72) is required for ergosterol biosynthesis Amino acid sequence alignment of ScErg28p with its homologs from other organisms was carried out and the results are shown in Fig. S2, Supporting Information. A BLASTn matrix assay demonstrated that the alignment scores of ScErg28p to its homologs from H. sapiens, P. troglodytes and C. albicans corresponded to 18.9% identity and 64.1% sequence similarity. Especially, each amino acid sequence of the Erg28p homologs from the four species contained a motif at positions 63–72, which was almost identical to the LS/QARTFGT/LWT consensus, and was located in the third truncated domain (ERG28Δ136-273). Moreover, results from the extensive BLASTn matrix analysis further confirmed the presence of the consensus motif 63LS/QARTFGT/LWT72 in all homologs (data not shown), indicating that the consensus sequence may be located within a region that is important for the role of Erg28p in sterol C-4 demethylation. Thus, a targeted deletion of the putative consensus motif was carried out, and the resulting ERG28 mutant (ScERG28Δ175-204) was used to complement the erg28-disrupted strain. The growth phenotype and the profile of sterol metabolic intermediates, especially the relative accumulation of ergosterol, were investigated in strains with different ERG28 genetic backgrounds, including WT, disrupted (erg28), restored (erg28-ScERG28), and mutant (erg28-ScERG28Δ175-204) strains. As shown in Fig. 5A, the spot plate analysis revealed that the number of total viable cells of the disrupted and mutant strains in the unit area of the plate was significantly smaller than that of the WT and restored strains at dilutions of 10−5 and 10−6. Compared with the relatively higher yield of ergosterol in BY4741 cells (8.89 ± 0.33 mg/g DCW), the deletion of erg28 resulted in a significant decrease (3.81 ± 0.17 mg/g DCW). Moreover, the yield of ergosterol in the recovered strain erg28-ScERG28 reached 12.51 ± 0.53 mg/g DCW, while a further deletion of the consensus motif in the erg28-ScERG28Δ175-204 strain resulted in an obvious decrease to 1.25 ± 0.46 mg/g DCW. Considering the complex interactions between Erg28p and other components involved in C-4 demethylation, there may be a number of proteins and interfaces that are not limited to these 10 amino acids. Thus, studies providing direct evidence of multiprotein interactions, such as polyprotein mass spectrometry, are urgently needed. Figure 5. View largeDownload slide Comparison of the growth rate and relative ergosterol accumulation in BY4741 strains with different ERG28 genetic backgrounds, including WT, erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204. (A) Spot plate assay for growth of different strains on CSM-URA medium ate 30°C. Serial dilutions of 106–101 cells were spotted onto the indicated plate areas. (B) Relative ergosterol accumulation in the WT (control), erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204 strains. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. Figure 5. View largeDownload slide Comparison of the growth rate and relative ergosterol accumulation in BY4741 strains with different ERG28 genetic backgrounds, including WT, erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204. (A) Spot plate assay for growth of different strains on CSM-URA medium ate 30°C. Serial dilutions of 106–101 cells were spotted onto the indicated plate areas. (B) Relative ergosterol accumulation in the WT (control), erg28, erg28-ScERG28 and erg28-ScERG28Δ175-204 strains. The relative amounts were measured in three separate experiments and expressed as means ± standard deviations. Disruption of the four parallel helices of ScErg28p by the deletion of the 63LS/QARTFGT/LWT72 sequence Analysis of protein structures with multiple transmembrane domains by classical crystallographic methods is fraught with significant difficulties (Lacapere et al.2007). Since the 3D structure of Erg28p had not been reported at the time of this study, homology-based modeling of both WT and mutant ScErg28p with the 63LS/QARTFGT/LWT72 sequence deletion was conducted based on the reported crystal structures of similar proteins. Based on the TM-align structure calibration procedure, a I-TASSER model was matched to all the structures in the PDB library, and 10 proteins with high structural similarity were identified. The protein structure with the highest TM-score for Erg28p is shown in Fig. 6A. The template (PDB ID: 5H1Q) is a transport protein from Caenorhabditis elegans that is located in the worm's INX-6 gap junction hemichannel (Oshima, Tani and Fujiyoshi 2016). It has been observed that Erg28p has four parallel α-helix structures with an additional short α-helix as a signal peptide (Oh et al.2017), which is consistent with the results of the prediction of hydrophobic helices. The structure of the erg28-ScERG28Δ175-204 mutant was also modeled in the same way (Fig. 6B and C), using the C. albicans transceptor Mep2, a protein that plays an important role in fungal development as an ammonium sensor, as the template (PDB ID: 5AEZ) (Van Den Berg et al.2016). As shown in Fig. 6B, the original parallel spiral structure of the four α-helixes was severely damaged, resulting in irregular random coils, leaving only a short region of α-helix within the third transmembrane domain. As shown in Fig. 6C, although the four α-helix structures were present in the ScERG28Δ175-204 mutant, the folding, stretching angle and length of the four domains were obviously altered. These results suggest that the deletion of the residues spanning the region from 63 to 72 seriously affected the topological structure of Erg28p, and in turn almost certainly affected its biological function. Figure 6. View largeDownload slide Three-dimensional structure prediction of ScErg28p by homology modeling. (A) Homology model of WT Erg28p generated using its homolog from C. elegans (PDB ID: 5H1Q) as template. (B) Models of two recommended 3D structures for the Erg28pΔ63-72 mutant constructed based on the template protein (PDB ID: 5AEZ) using the I-TASSER online server. The five parallel helix domains are shown in red (1–20), green (21–44), blue (45–90), yellow (91–113) and bright blue (113–148), respectively. Figure 6. View largeDownload slide Three-dimensional structure prediction of ScErg28p by homology modeling. (A) Homology model of WT Erg28p generated using its homolog from C. elegans (PDB ID: 5H1Q) as template. (B) Models of two recommended 3D structures for the Erg28pΔ63-72 mutant constructed based on the template protein (PDB ID: 5AEZ) using the I-TASSER online server. The five parallel helix domains are shown in red (1–20), green (21–44), blue (45–90), yellow (91–113) and bright blue (113–148), respectively. In summary, our study identified a novel consensus motif sequence (63LS/QARTFGT/LWT72) that significantly correlates with the phenotypes of yeast growth, ergosterol biosynthesis and correct folding of four parallel α-helixes in the three-dimensional structure of Erg28p, confirming its biofunctional role in sterol C-4 demethylation in the ergosterol biosynthesis pathway of S. cerevisiae. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 31400978) and the Natural Science Foundation of Zhejiang Province (No. LY18B020019). Conflict of interest. None declared. REFERENCES Adams BG, Parks LW. Isolation from yeast of a metabolically active water-soluble form of ergosterol. J Lipid Res  1968; 9: 8– 11. Google Scholar PubMed  Bard M, Bruner DA, Pierson CA et al.   Cloning and characterization of ERG25, the Saccharomyces cerevisiae gene encoding C-4 sterol methyl oxidase. P Natl Acad Sci USA  1996; 93: 186– 90. Google Scholar CrossRef Search ADS   Bloxham DP, Wilton DC, Akhtar M. Studies on the mechanism and regulation of C-4 demethylation in cholesterol biosynthesis. The role of adenosine 3':5'-cyclic monophosphate. Biochem J  1971; 125: 625– 34. Google Scholar CrossRef Search ADS PubMed  Buurman ET, Blodgett AE, Hull KG et al.   Pyridines and pyrimidines mediating activity against an efflux-negative strain of Candida albicans through putative inhibition of lanosterol demethylase. Antimicrob Agents Ch  2004; 48: 313– 8. Google Scholar CrossRef Search ADS   Fernandez-Leiro R, Scheres SH. Unravelling biological macromolecules with cryo-electron microscopy. Nature  2016; 537: 339– 46. Google Scholar CrossRef Search ADS PubMed  Gachotte D, Barbuch R, Gaylor J et al.   Characterization of the Saccharomyces cerevisiae ERG26 gene encoding the C-3 sterol dehydrogenase (C-4 decarboxylase) involved in sterol biosynthesis. P Natl Acad Sci USA  1998; 95: 13794– 9. Google Scholar CrossRef Search ADS   Gachotte D, Eckstein J, Barbuch R et al.   A novel gene conserved from yeast to humans is involved in sterol biosynthesis. J Lipid Res  2001; 42: 150– 4. Google Scholar PubMed  Gachotte D, Sen SE, Eckstein J et al.   Characterization of the Saccharomyces cerevisiae ERG27 gene encoding the 3-keto reductase involved in C-4 sterol demethylation. P Natl Acad Sci USA  1999; 96: 12655– 60. Google Scholar CrossRef Search ADS   Herrmann JM, Westermann B, Neupert W. Analysis of protein-protein interactions in mitochondria by coimmunoprecipitation and chemical cross-linking. Methods Cell Biol  2001; 65: 217– 30. Google Scholar CrossRef Search ADS PubMed  Hughes TR, Marton MJ, Jones AR et al.   Functional discovery via a compendium of expression profiles. Cell  2000; 102: 109– 26. Google Scholar CrossRef Search ADS PubMed  Kelley LA, Sternberg MJ. Protein structure prediction on the Web: a case study using the Phyre server. Nat Protoc  2009; 4: 363– 71. Google Scholar CrossRef Search ADS PubMed  Lacapere JJ, Pebay-Peyroula E, Neumann JM et al.   Determining membrane protein structures: still a challenge! Trends Biochem Sci  2007; 32: 259– 70. Google Scholar CrossRef Search ADS PubMed  Mialoundama AS, Jadid N, Brunel J et al.   Arabidopsis ERG28 tethers the sterol C4-demethylation complex to prevent accumulation of a biosynthetic intermediate that interferes with polar auxin transport. Plant Cell  2013; 25: 4879– 93. Google Scholar CrossRef Search ADS PubMed  Mo C, Bard M. Erg28p is a key protein in the yeast sterol biosynthetic enzyme complex. J Lipid Res  2005a; 46: 1991– 8. Google Scholar CrossRef Search ADS   Mo C, Bard M. A systematic study of yeast sterol biosynthetic protein-protein interactions using the split-ubiquitin system. BBA-Mol Cell Biol L  2005b; 1737: 152– 60. Google Scholar CrossRef Search ADS   Mo C, Valachovic M, Bard M. The ERG28-encoded protein, Erg28p, interacts with both the sterol C-4 demethylation enzyme complex as well as the late biosynthetic protein, the C-24 sterol methyltransferase (Erg6p). BBA-Mol Cell Biol L  2004; 1686: 30– 6. Google Scholar CrossRef Search ADS   Mo C, Valachovic M, Randall SK et al.   Protein-protein interactions among C-4 demethylation enzymes involved in yeast sterol biosynthesis. P Natl Acad Sci USA  2002; 99: 9739– 44. Google Scholar CrossRef Search ADS   Niedenthal RK, Riles L, Johnston M et al.   Green fluorescent protein as a marker for gene expression and subcellular localization in budding yeast. Yeast  1996; 12: 773– 86. Google Scholar CrossRef Search ADS PubMed  Oh KH, Haney JJ, Wang X et al.   ERG-28 controls BK channel trafficking in the ER to regulate synaptic function and alcohol response in C. elegans. Elife  2017; 6: e24733. Google Scholar CrossRef Search ADS PubMed  Oshima A, Tani K, Fujiyoshi Y. Atomic structure of the innexin-6 gap junction channel determined by cryo-EM. Nat Commun  2016; 7: 13681. Google Scholar CrossRef Search ADS PubMed  Parks LW, Casey WM. Physiological implications of sterol biosynthesis in yeast. Annu Rev Microbiol  1995; 49: 95– 116. Google Scholar CrossRef Search ADS PubMed  Rahier A. Dissecting the sterol C-4 demethylation process in higher plants. From structures and genes to catalytic mechanism. Steroids  2011; 76: 340– 52. Google Scholar CrossRef Search ADS PubMed  Rahier A, Darnet S, Bouvier F et al.   Molecular and enzymatic characterizations of novel bifunctional 3beta-hydroxysteroid dehydrogenases/C-4 decarboxylases from Arabidopsis thaliana. J Biol Chem  2006; 281: 27264– 77. Google Scholar CrossRef Search ADS PubMed  Sheff MA, Thorn KS. Optimized cassettes for fluorescent protein tagging in Saccharomyces cerevisiae. Yeast  2004; 21: 661– 70. Google Scholar CrossRef Search ADS PubMed  Sipiczki M. Fission yeast phylogenesis and evolution. In: The Molecular Biology of Schizosaccharomyces Pombe . Berlin, Heidelberg: Springer, 2004, 431– 43. Google Scholar CrossRef Search ADS   Thaminy S, Auerbach D, Arnoldo A et al.   Identification of novel ErbB3-interacting factors using the split-ubiquitin membrane yeast two-hybrid system. Genome Res  2003; 13: 1744– 53. Google Scholar CrossRef Search ADS PubMed  Van Den Berg B, Chembath A, Jefferies D et al.   Structural basis for Mep2 ammonium transceptor activation by phosphorylation. Nat Commun  2016; 7: 11337. Google Scholar CrossRef Search ADS PubMed  Veitia RA, Hurst LD. Accelerated molecular evolution of insect orthologues of ERG28/C14orf1: a link with ecdysteroid metabolism? J Genet  2001; 80: 17– 21. Google Scholar CrossRef Search ADS PubMed  Vinci G, Xia XH, Veitia RA. Preservation of genes involved in sterol metabolism in cholesterol auxotrophs: facts and hypotheses. PLoS One  2008; 3: e2883. Google Scholar CrossRef Search ADS PubMed  Wilton DC, Akhtar M. The mechanism of C-4 demethylation during cholesterol biosynthesis. Biochem J  1975; 149: 233– 5. Google Scholar CrossRef Search ADS PubMed  Yun Y, Yin D, Dawood DH et al.   Functional characterization of FgERG3 and FgERG5 associated with ergosterol biosynthesis, vegetative differentiation and virulence of Fusarium graminearum. Fungal Genet Biol  2014; 68: 60– 70. Google Scholar CrossRef Search ADS PubMed  © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Microbiology LettersOxford University Press

Published: Mar 1, 2018

There are no references for this article.