Yeast plasma membrane Fungal Oligopeptide Transporters display distinct substrate preferences despite their high sequence identity

Fungi Journal of Article Yeast Plasma Membrane Fungal Oligopeptide Transporters Display Distinct Substrate Preferences despite Their High Sequence Identity Carmen Becerra-Rodríguez 1,2 , Géraldine Taghouti 3, Perrine Portier 3, Sylvie Dequin 1, Margarida Casal 2, Sandra Paiva 2and Virginie Galeote 1,*   Citation: Becerra-Rodríguez, C.; Taghouti, G.; Portier, P.; Dequin, S.; Casal, M.; Paiva, S.; Galeote, V. Yeast Plasma Membrane Fungal Oligopeptide Transporters Display Distinct Substrate Preferences despite Their High Sequence Identity. J. Fungi 2021,7, 963. https://doi.org/ 10.3390/ jof7110963 Academic Editor: Aaron Neiman Received: 4 October 2021 Accepted: 4 November 2021 Published: 12 November 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1SPO, Univ. Montpellier, INRAE, Institut Agro, F-34060 Montpellier, France; [email protected] (C.B.-R.); [email protected] (S.D.) 2Centre of Environmental and Molecular Biology, Department of Biology, Campus of Gualtar, University of Minho, 4710-057 Braga, Portugal; [email protected] (M.C.); [email protected] (S.P.) 3Univ. Angers, Institut Agro, INRAE, IRHS, SFR QUASAV, CIRM-CFBP, F-49000 Angers, France; [email protected] (G.T.); [email protected] (P.P.) *Correspondence: vir[email protected] Abstract: Fungal Oligopeptide Transporters (Fot) Fot1, Fot2 and Fot3 have been found in Saccharomyces cerevisiae wine strains, but not in strains from other environments. In the S. cerevisiae wine strain EC1118, Fot1 and Fot2 are responsible for a broader range of oligopeptide utilization in comparison with strains not containing any Fot. This leads to better fermentation efficiency and an increased production of desirable organoleptic compounds in wine. Despite the benefits associated with Fot activity in S. cerevisiae within the wine environment, little is known about this family of transporters in yeast. The presence of Fot1, Fot2 and Fot3 in S. cerevisiae wine strains is due to horizontal gene transfer from the yeast Torulaspora microellipsoides, which harbors Fot2Tm, FotX and FotY proteins. Sequence analyses revealed that Fot family members have a high sequence identity in these yeast species. In this work, we aimed to further characterize the different Fot family members in terms of subcellular localization, gene expression in enological fermentation and substrate specificity. Using CRISPR/Cas9, we constructed S. cerevisiae wine strains containing each different Fot as the sole oligopeptide transporter to analyze their oligopeptide preferences by phenotype microarrays. The results of oligopeptide consumption show that Fot counterparts have different di- /tripeptide specificities, suggesting that punctual sequence divergence between FOT genes can be crucial for substrate recognition, binding and transport activity. FOT gene expression levels in different S. cerevisiae wine strains during enological fermentation, together with predicted binding motifs for transcriptional regulators in nitrogen metabolism, indicate that these transporters may be under the control of the Nitrogen Catabolite Repression (NCR) system. Finally, we demonstrated that Fot1 is located in the yeast plasma membrane. This work contributes to a better understanding of this family of oligopeptide transporters, which have demonstrated a key role in the utilization of oligopeptides by S. cerevisiae in enological fermentation. Keywords: Fungal Oligopeptide Transporters (Fot); oligopeptide transport; Saccharomyces cerevisiae; phenotype microarrays; GFP labeling 1. Introduction Oligopeptides, often simply referred to as peptides, are short chains of two to nine amino acid residues linked by an amide type bond mainly present in nature as a result of proteolytic processes. They constitute a highly diverse source of nitrogen and carbon for microorganisms. As for other nutrient molecules, the ability to consume oligopeptides from the environment can represent a crucial competitive advantage in a microbial ecosystem. In the eukaryotic model organism Saccharomyces cerevisiae, three systems for oligopeptide transport are well documented: Ptr2, a Proton-dependent Oligopeptide Transporter J. Fungi 2021,7, 963. https://doi.org/10.3390/jof7110963 https://www.mdpi.com/journal/jof J. Fungi 2021,7, 963 2 of 16 (Pot/Ptr, Transporter Classification (TC) number: 2.A.17) [ 1 ] is the best-known diand tripeptide transporter in yeast, together with the allantoate and ureidosuccinate permease Dal5 (TC 2.A.1.14.4), which also displays dipeptide transport activity [ 2 ]. Two members of the Oligopeptide Transport (Opt) family (TC 2.A.67), glutathione transporter Opt1 and its paralogue Opt2, can transport tetrapeptides and also pentapeptides in the specific case of Opt1 [ 3 , 4 ]. Finally, new oligopeptide transporters from eukaryotes have been identified from a soil sample. Phylogenetic analyses indicated that these transporter sequences only grouped with other fungal species and were not homologous to either Ptr2, Dal5 or Opt members; as a result, a novel family termed Fungal Oligopeptide Transporters (Fot) was defined and characterized as proton-driven membrane transporters in fungi able to import di-and tripeptides [ 5 ]. Fot proteins were related to fungal members of the Amino Acid/Auxin Permease family (TC 2.A.18.4.) [ 5 ]. Interestingly, S. cerevisiae FOT genes are only present in wine strains, and not in strains from other environments. The genes FOT1, FOT2 and FOT3, found in different S. cerevisiae wine strains, are the result of a horizontal gene transfer and subsequent gene conversions between FOTX and FOT2Tm from the yeast Torulaspora microellipsoides, which also contains FOTY [ 6 , 7 ]. Fot family members display a high sequence identity at both the protein and gene level, from 90% identity between FOTY and FOT2 genes to 98% between FOTX and FOT3 [7]. The deletion of FOT1 and FOT2 genes in the S. cerevisiae 59A strain, a haploid derivative of commercial wine strain Lalvin EC1118 ® , led to a 35% drop in oligopeptide-derived nitrogen consumption. Compared to the deletion mutant, wild-type strain 59A had higher biomass accumulation by the end of fermentation, which is consistent with a higher nitrogen consumption [ 7 ]. Strain 59A also showed two remarkable features as a consequence of the higher consumption of glutamate/glutamine-rich oligopeptides due to Fot1 and Fot2 activity: first, a higher cell viability by the end of fermentation in comparison to fot1fot2 ∆ [ 7 ], and secondly, a more positive wine organoleptic balance due to a lower production of acetate and higher levels of ester acetates and fusel alcohols [ 8 ]. This effect on the production of fermentative organoleptic compounds was later demonstrated to be dependent on the source of peptides present in the must [ 9 ]. With higher biomass and viability as evident fitness indicators, these results suggest that Fot acquisition by S. cerevisiae wine strains confers a competitive advantage in the wine environment [ 7 , 8 , 10 ]. Despite the important role that Fot family members play in the adaptation of S. cerevisiae to the wine fermentation environment, little is known still about this family of transporters. The most evident question that arises is the biological significance of the five different Fot members currently known, considering their high sequence identity. To answer this question, we followed a CRISPR/Cas9 strategy to construct S. cerevisiae wine strains containing single Fot members as the sole oligopeptide transporter. Analysis of oligopeptide preferences by these strains revealed that Fot1, Fot2, Fot3, FotX and FotY have distinct substrate specificities, highlighting the importance of the amino acid nature and position within the oligopeptide for their consumption as substrates. Moreover, we provide evidence that FOT expression is dependent on the strain, stage of enological fermentation and composition of the yeast assimilable nitrogen. Additionally, using fluorescence microscopy and co-localization studies with GFP labeling, we demonstrated that Fot1 is localized in the yeast plasma membrane. With this work we aimed at advancing the knowledge regarding the expression and function of Fot family members, which have a key role in the adaptation of S. cerevisiae to wine environments. 2. Materials and Methods 2.1. Yeast Strains and Fermentation Conditions In this study, we worked with S. cerevisiae strains 59A and MTF2533, haploid derivatives of the commercial wine strains Lalvin EC1118 ® and LMD1, respectively [ 11 , 12 ]. We used a 59A version in which gene AMN1 has been deleted to avoid cell aggregation [ 7 , 13 , 14 ]. Fermentations were carried out in 1.2 L glass fermenters inoculated with 10 6 cells/mL. We used natural Colombard grape must (Caussens, France, 2019) containing 183 g/L of sugars J. Fungi 2021,7, 963 3 of 16 and 257 mg/L of yeast assimilable nitrogen (ammonium and amino acids). To mimic the conditions and composition of the natural must, a synthetic must (SM) was prepared with 183 g/L glucose/fructose, 257 mg/L yeast assimilable nitrogen (ammonium and free amino acids), 1.12 mg/L oleic acid and 3.75 mg/L ergosterol, at pH 3.3 [ 15 ]. Cells were grown on double overnight pre-cultures, first on YPD (1% yeast extract, 2% bactopeptone, 2% glucose) at 28 ◦ C in 10 mL flasks with shaking (180 rpm) and then on SM in the same conditions before inoculation of fermenters. Strains with antibiotic resistance were selected on YPD medium supplemented with 300 µ g/mL hygromycin B (Sigma, 31282–04–9, Saint Louis, MO, USA), 200 µ g/mL G418 (Sigma A-1720) and/or 100 µ g/mL nourseothricin (Werner, 96736–11–7, Meisenweg, Germany). 2.2. Strain and Plasmid Construction Tandem genes FOT1–FOT2 were replaced by a KANMX4 cassette in strain 59A, obtaining the strain fot1fot2 ∆ . The genes of the non-Fot oligopeptide transporters in 59A, i.e., OPT1,OPT2 and DAL5, were then deleted from 59A and fot1fot2 ∆ using the CRISPR/Cas9 system with two plasmids and one repair fragment [ 16 ]. This system requires first a transformation with plasmid pCfB2513 for the expression of Cas9; in a second transformation, a repair fragment and plasmid pMEL15 [ 17 ] containing the guide RNA (gRNA) cassette for the target gene are introduced. Genes DAL5,OPT1 and OPT2 were sequentially deleted, with the consequent pMEL15 loss in between to ensure the correct selection of transformants. pMEL15 vectors containing the gRNA cassettes were generated by PCR, whereas the repair fragments consisting of disrupted versions of the target genes were designed and amplified from a pEX-A128 plasmid (Eurofins Genomics, Ebersberg, Germany). PTR2, which is the gene coding the best-known dipeptide transporter in S. cerevisiae, is not functional in 59A, and its deletion was therefore not required [ 5 ]. Deletion of non-FOT oligopeptide transporter genes in fot1fot2 ∆ resulted in the strain opt1 ∆ opt2 ∆ dal5 ∆ fot1fot2 ∆ ::KANMX4, termed PepKO, which was a knockout strain for oligopeptide transport and constituted a platform strain for the insertion of single FOT genes. Using CRISPR/Cas9, each FOT gene was inserted in a substitution of the KANMX4 cassette in PepKO. In this way, all FOT genes were individually located in the original FOT1–FOT2 locus and therefore were under the regulation of the FOT2 promoter and FOT1 terminator. All these transformations were carried out using the lithium acetate method for yeast transformation [ 18 ]. The different strains containing single FOT genes were confirmed by sequencing. Plasmids containing the GFP gene fused to FOT1 at its 5 0 or 3 0 ends (pGFP–Fot1 and pFot1–GFP, respectively) were constructed by Gibson assembly (New England Biolabs, Ipswich, MA, USA) and confirmed by digestion with restriction enzymes. The strains used and constructed in this study are listed in Table 1; primers and plasmids used in this study are listed in Tables S1 and S2, respectively. 2.3. Phenotype Microarray Assays for Diand Tripeptide Consumption The di/tripeptide utilization profile of the different strains was performed using the Biolog (Hayward, CA, USA) Phenotype MicroArrays (PM) system combined with the OmniLog reader (Biolog, Hayward, CA, USA) at the COMIC facility of SFR Quasay (University of Angers, France). This technique enables the monitoring of the consumption of a range of substrates over time through a colorimetric method. We used four PM plates (PM3B for nitrogen sources, PM6, PM7 and PM8 for peptide nitrogen sources) containing a total range of 270 dipeptides and 14 tripeptides, with a negative control and L-glutamine as a positive control per plate. Cultures were incubated on yeast nitrogen base plates without amino acids (YNB, Difco BD 91940, 6.7 g/L; 2% glucose, 2% agar) at 28 ◦ C in duplicate and prepared for incubation on PM plates as in [ 19 ]. PM assays were performed at 30 ◦ C over 72 h. After this time, data from the Omnilog system were retrieved using Kinetics software v1.30. Area under the curve was calculated and assigned as consumption values. We considered a strain to have utilized a di/tripeptide when the consumption value after J. Fungi 2021,7, 963 4 of 16 72 h on this nitrogen source represented more than 20% of L-glutamine consumption; this threshold was taken as the positive control and therefore the glutamine consumption value for each strain in each plate represented 100% consumption. Below 20%, the growth signal would possibly be mistaken with spontaneous reduction of the tetrazolium dye or reduction by the remaining yeast inoculum [ 5 ]. Levels of consumption from 0 to 5 were established based on the consumption value of L-glutamine, the positive control in each plate. Level 0, 0–20% of consumption on L-Gln; 1, 21–40%; 2, 41–60%; 3, 61–80%; 4, 81–100%; 5, >100%. Table 1. Strains used in this study. Strain Genotype Source/Reference 59A MATa ho amn1∆::LOXP [7] fot1fot2∆MATa ho amn1∆::LOXP fot1fot2∆::KANMX4 This study opt1∆opt2∆dal5∆MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆This study PepKO MATa ho amn1∆::LOXP fot1fot2∆::KANMX4 opt1∆opt2∆dal5∆This study PepKO-Fot1 MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOT1 This study PepKO-Fot2 MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOT2 This study PepKO-Fot3 MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOT3 This study PepKO-FotX MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOTX This study PepKO-FotY MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOTY This study PepKO-Fot2Tm MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOT2Tm This study PepKO-Fot1Fot2 MATa ho amn1∆::LOXP opt1∆opt2∆dal5∆fot1fot2∆::FOT1–FOT2 This study 59A-GFP MATa ho amn1∆::TEFp-GFP-ADH1-NATMX4 [7] MTF2533 MATa ho::LOXP [12] 2.4. Gene Expression Analysis Gene expression was analyzed in three independent cultures for each medium and yeast strain. Cells were sampled at 10% (growth phase) and 40% (stationary phase) of the total CO 2 produced from each must during fermentation (see Figure S3). A total of 10 9 cells were collected for each strain/must/fermentation stage, washed with diethylpyrocarbonate (DEPC)-treated water, frozen in methanol and stored at − 80 ◦ C for RNA extraction. RNA was isolated using Trizol reagent (Gibco BRL, Life Technologies, Waltham, MA, USA) and purified by isopropanol precipitation using a RNeasy kit (Qiagen, Hilden, Germany). RNA samples were retro-transcribed into cDNA and used for Quantitative Polymerase Chain Reaction (qPCR). Following the recommendations of the Real-Time PCR system manufacturer (Applied Biosystems, Waltham, MA, USA), FOT gene expression was quantified by a relative standard curve method using genomic DNA of the yeast strain. Expression values were subsequently normalized with those of the house-keeping gene SCR1 (Small Cytoplasmatic RNA 1), which is commonly used as reference gene in S. cerevisiae due to its high expression stability. Primers were designed for specific amplification of FOT1,FOT2, FOT3 and SCR1 (Table S1). 2.5. Epifluorescence and Confocal Microscopy N-(3-Triethylammoniumpropyl)-4-(6-(4-(Diethylamino) Phenyl) Hexatrienyl) Pyridinium Dibromide or FM4–64 dye (Invitrogen, Waltham, MA, USA) was used as a fluorescence differential marker for plasma membrane (excitation 558 nm, emission 734 nm). A total of 100 µ L of cells was collected at the exponential phase and incubated with 80 µ M FM4–64 at 4 ◦ C with agitation (1500 rpm). After a 1-h incubation, cells were spun at 700 g for 3 min at 4 ◦ C and prepared for immediate visualization (adapted from [ 20 ]). Cells were visualized with a 100 × /1.3 oil objective under an Axio Imager Cam MRM A2 microscope (Carl Zeiss, White Plains, NY, USA) equipped with an excitation source and a range of filters. For this study, we used filter 38 (excitation BP 470/40 and emission BP 525/50) for visualizing GFP (488 nm excitation and 530 nm emission) and filter 20 (excitation: BP 546/12, emission: BP 575–640) for visualizing FM4–64 staining. Images of epifluorescence J. Fungi 2021,7, 963 5 of 16 microscopy were captured and processed with ZEN 2012 software, version 1.1.2.0 (blue edition; Carl Zeiss, New York, NY, USA). Confocal microscopy was performed with a Confocal Leica 8 (Leica Microsystems CMS, Germany) with a 40 × /1.1 oil objective. Fluorescence emission was collected at 500–540 nm for GFP and at 600–650 nm for FM4–64 by sequential acquisition. Images from confocal microscopy were treated and analyzed with LAS X (Leica) and Fiji [21] software. 2.6. Analysis of Promoter Regions Characterization of promoter regions was carried out using the YeTFaSCo database v1.02 [ 22 ] to identify binding sites for transcription factors. A 500 bp region upstream the start codon was analyzed against the expert-curated, non-dubious set of transcriptional factors. Results were filtered for a value of maximum score higher than 95% identity. 2.7. Data Treatment and Statistical Analysis Data were treated and analyzed using R v4.1.0 (R Core Team 2021) and RStudio (RStudio Team 2020). Dedicated packages were used depending on the purpose: tidyverse [ 23 ] for data manipulation and visualization; heatmaply [ 24 ] for heatmap construction and agricolae package (Mendiburu and Yaseen, 2020) for statistical analysis. 3. Results 3.1. Evaluation of Substrate Specificity in Fot Family Members by Phenotype Microarrays We sought to characterize the substrate preference of each Fot by phenotype microarrays. For this purpose, we first generated a complete knockout strain for oligopeptide transport by deleting OPT1,OPT2 and DAL5 with CRISPR/Cas9 and substituting FOT1–FOT2 tandem genes by a KANMX4 cassette (Table 1). This knockout strain, denominated PepKO, was unable to consume any of the 284 oligopeptides in the microarray, except for a weak consumption of His–Pro, Thr–Ser and γ -Glu–Gly dipeptides (Figure 1, second to last row in the heatmap; Table S3). The consumption of γ -Glu–Gly was probably due to the activity of the general amino acid transporter Gap1, which has also been reported to transport γ -glutamyl dipeptides [ 25 ]. In comparison, the wild-type strain 59A consumed 195 dipeptides and 12 tripeptides from a total range of 270 dipeptides and 14 tripeptides (Figure 1, first row in the heatmap). This result confirms that we had deleted all significant diand tripeptide transporters in strain 59A. Using CRISPR/Cas9, each FOT gene was inserted into the PepKO strain, substituting the KANMX4 cassette. In this way, we generated S. cerevisiae strains containing single Fot family members as sole oligopeptide transporters (Table 1), with FOT genes individually located in the original FOT1–FOT2 locus and therefore were under the regulation of the FOT2 promoter and FOT1 terminator. This allowed the characterization of independent transporters that were under the same gene regulation. Strains containing single Fot members showed different peptide specificities. In the cladogram to the right in Figure 1, 59A is grouped with strains expressing Fot1 and Fot2 but not the non-Fot oligopeptide transporters, which are opt1 ∆ opt2 ∆ dal5 ∆ and PepKO–Fot1Fot2. Strain opt1 ∆ opt2 ∆ dal5 ∆ resulted from DAL5,OPT1 and OPT2 deletion in the wild-type strain 59A, while PepKO–Fot1Fot2 originates from the re-insertion of FOT1–FOT2 in PepKO. Both strains consumed, respectively, 189 and 191 oligopeptides from a total of 207 oligopeptides consumed by the wild-type atrain, only differing in Gly–Leu and Lys–Trp, which were weakly consumed (consumption level = 1) by PepKO–Fot1Fot2 (Table S3). This result confirms that the genome insertion of FOT genes on the platform strain PepKO did not perturb the oligopeptide consumption phenotype. Strain fot1fot2 ∆ was able to consume 12 dipeptides and 2 tripeptides that were also consumed by the wild-type and, conversely, not consumed by opt1 ∆ opt2 ∆ dal5 ∆ or PepKO–Fot1Fot2. Therefore, this 14-oligopeptide fraction, characterized by the presence of glycine at the oligopeptide N-terminus, is specifically consumed by strains expressing the non-Fot oligopeptide transporters, Dal5, Opt1 or Opt2. Consequently, these results confirm J. Fungi 2021,7, 963 6 of 16 that Fot1 and Fot2 are the main diand tripeptide transporters in the wine strain 59A, as we previously reported under different experimental conditions [7]. Among the strains containing single Fot, the strain with the widest range of oligopeptide utilization was the one expressing FotX, consuming 152 diand tripeptides (Figure 1, forthth row in the heatmap). Consumption profiles for cells expressing Fot2Tm and Fot2 from S. cerevisiae were highly similar (Figure 1, sixth–seventh rows in the heatmap), an expected result considering they share the same protein sequence. Fot1 peptide preferences were closer to those of Fot2 than to the ones of FotX or Fot3, which was an interesting result considering that Fot1 shares 96.21% protein sequence identity with Fot2 and, respectively, 98.11% and 97.26% with FotX and Fot3 (Figure S1) ([ 7 ]; reviewed in [ 10 ]). Fot3 and FotY were sub-grouped together in the dendrogram (Figure 1) since they had a comparatively shorter range of oligopeptide utilization, i.e., 110 and 99 out of 284 diand tripeptides, respectively. The fact that the Fot3 strain had a peptide utilization closer to the FotY strain than to the Fot1 or FotX strains was a priori unexpected, considering that Fot3 displays 97.26% and 98.8% amino acid sequence identity, respectively, with these transporters, but only 92.21% with FotY. This result may indicate that the localized sequence differences between Fot3 and FotX can result in a dramatic change in transport capabilities. It was remarkable that Fot1 and Fot2 as sole transporters were not able to match the transport specificity of strains containing both transporters, PepKO–Fot1Fot2 and opt1 ∆ opt2 ∆ dal5 ∆ . Strains expressing Fot1–Fot2 consumed 59 oligopeptides more than strains with Fot1 (PepKO–Fot1) or Fot2 (PepKO–Fot2) only (Figure S2A). This difference did not simply come from an expected additive effect of Fot1 and Fot2 acting together, since the level of consumption of 21 out of these 59 oligopeptides was 4 in strains harboring Fot1–Fot2 (high consumption) versus level 0 in strains with Fot1 or Fot2 (Figure S2B). The group of 59 oligopeptides transported by Fot1 and Fot2 together but not singly were rich in Gly and, to a lesser extent, in Trp, Glu, Pro and Asp, with the three latter particularly abundant at the C-terminal position (Figure S2C). This result suggests a possible interaction between Fot1 and Fot2 that causes a modification in their transport abilities and specificities. Figure 1. Fot have different peptide specificities. The heatmap represents the consumption of diand tripeptides (columns) by Fot-containing strains (rows). Strains are sorted by consumption preferences similarity, represented by the cladogram to the right. The cladogram on top of the heatmap ranks the diand tripeptides according to their preferential consumption. J. Fungi 2021,7, 963 7 of 16 3.2. Specificity of Fot Members Depends on the Type of Amino Acid Located in the Oligopeptide N-Terminus The results of the diand tripeptide consumption from the Phenotype Microarrays (PM) plates were categorized based on the type of amino acid located at the N-terminal position (Figure 2). In this way, we observed how, in general, Fot preferably transport peptides with hydrophobic amino acids at the N-terminus (Figure 2A), with the general exception of Gly-X and Pro-X oligopeptides. Within the group of oligopeptides with hydrophobic amino acids at the N-terminus, those containing Glu, Asp, Lys, Gly or Pro at the C-terminus were weakly or not consumed by Fot-expressing strains, with the exceptions of Phe–Glu, Phe–Asp or Phe–Gly, moderately consumed in the presence of FotX. Peptides with Trp at the N-terminus were less consumed, although Trp-Tyr was highly consumed by strains harboring FotX, Fot1 and Fot2, but moderately so by strains expressing the other transporters. Except for Cys–Gly, peptides with polar, uncharged amino acids in the N-terminus were also generally consumed by strains harboring Fot members (Figure 2B), particularly if these oligopeptides contained Tyr and/or a hydrophobic amino acid at the C-terminus. Peptides containing charged amino acids were generally not consumed by strains with single Fot, either in N- (Figure 2C) or C-terminal positions, with the exception of His-containing peptides, which were moderately to highly consumed when any Fot was present. A general exception to all these observations regarded peptides containing an Arg in the C-terminal position, which were generally well consumed by all strains expressing any single Fot member, while oligopeptides with Arg in the N-terminus were not consumed by strains containing single Fot or Fot1–Fot2 (except for Arg–Trp, Arg–Lys and Arg–Tyr, consumed in a level range of 2 to 4 by opt1 ∆ opt2 ∆ dal5 ∆ and PepKO–Fot1Fot2; Table S4). Exceptions to this observation were Pro–Arg, highly consumed by FotX and Fot1–Fot2 strains only, and transporter FotY, which did not allow any Met–Arg, Trp–Arg or Pro–Arg consumption. The case of Arg suggests that localization of amino acids within the oligopeptide is also important for oligopeptide recognition as a substrate. Gln was also more frequent at the C-terminal position of peptides consumed by single Fot-containing strains rather than at the N-terminus (Table S4). Moreover, peptides with D-, β - or γ -amino acids were not consumed by any strain, with the exception of γ -Glu-Gly which was weakly consumed by all strains including PepKO. J. Fungi 2021,7, 963 8 of 16 Figure 2. Oligopeptide preferences depend on the character of the amino acid residues. In this figure, diand tripeptides are classified based on the amino acid type at the N-terminal position. Levels of consumption are expressed in comparison to the consumption value of L-glutamine, as a positive control in each plate; level 0, 0–20% of consumption on L-Gln; level 1, 21–40%; level 2, 40–60%; level 3, 60–80%; level 4, 80–100%; level 5, >100%. ( A ) Peptides with hydrophobic amino acids. (B) Peptides with polar, uncharged amino acids; (C) Peptides with charged amino acids. J. Fungi 2021,7, 963 9 of 16 3.3. FOT Genes Expression Depends on the S. cerevisiae Strain Background, Type of Nitrogen Source and Stage of Enological Fermentation Two S. cerevisiae wine strains were selected to analyze the expression of FOT1,FOT2 and FOT3. Strain 59A contains FOT1 and FOT2 tandem genes, while strain MTF2533 only has FOT3. Two conditions of enological fermentation were evaluated: on one hand, a natural Colombard grape must (Caussens, France, 2019) containing 257 mg/L of assimilable nitrogen and 183 g/L of sugars; on the other hand, a synthetic must with the same concentrations of assimilable nitrogen and sugars. Additionally, FOT expression was analyzed at two key points of enological fermentation (Figure S3): 10% of fermentation, point at which cells were at mid-log phase of growth and had not yet reached the maximum fermentation rate or V max , and 40% of fermentation, which corresponded to a time point of fermentation after V max , when cells were in stationary phase. FOT expression was quantified by qPCR and normalized with expression values of the housekeeping gene SCR1. Relative expression values of FOT genes were low in both strains (Figure 3), although particularly low for FOT3 in strain MTF2533. Generally, these expression values were even lower in natural must than in synthetic must, although only statistically significant for FOT1 and FOT2 at 40% fermentation. Contrary to this result, we expected a higher expression of FOT genes in natural must since grape juice contains oligopeptides as nitrogen source, while synthetic must does not. In addition, FOT1 and FOT2 showed higher expression values at 40% of fermentation in both synthetic and natural must; however, the most notable expression was observed for FOT1 at 40% of fermentation in synthetic must, with a 10-fold expression compared to that at 10%; FOT1 was also 3.46-fold more expressed than FOT2 in the same medium and at the same stage of fermentation. Finding higher expression values at 40% of fermentation, i.e., during stationary phase (Figure S3) was also unexpected, since nitrogen is no longer consumed by yeast at this point of fermentation [ 15 ]. To predict how FOT expression is regulated, we analyzed the promoter regions of FOT1,FOT2 and FOT3, and those of FOTX,FOTY and FOT2Tm from T. microellipsoides using the scanner tool of the YeTFaSCo database (Table S5). Several binding motifs for transcriptional factors involved in the Nitrogen Catabolite Repression (NCR) such as Gln3 or Cup9 were found in the promoter regions of all FOT genes. Gln3 is a transcriptional activator of genes under the regulation of the NCR system, and Cup9 is a transcriptional repressor of both the dipeptide transporter gene PTR2 and tetrapeptide transporter gene OPT2 in S. cerevisiae [ 26 , 27 ]. Therefore, these findings suggest that FOT genes may be under NCR system repression in the presence of preferred nitrogen sources such as ammonium and certain amino acids, which is the case during the growth phase, and be are only expressed when these sources are scarce, which would agree with FOT1 overexpression at the stationary phase. J. Fungi 2021,7, 963 16 of 16 35. Lubkowitz, M.A.; Barnes, D.; Breslav, M.; Burchfield, A.; Naider, F.; Becker, J.M. Schizosaccharomyces pombe isp4 encodes a Transporter Representing a Novel Family of Oligopeptide Transporters. Mol. Microbiol. 1998,28, 729–741. [CrossRef] 36. Magasanik, B.; Kaiser, A.C. Nitrogen Regulation in Saccharomyces cerevisiae.Gene 2002,290, 1–18. [CrossRef] 37. Island, M.D.; Naider, F.; Becker, J.M. Regulation of Dipeptide Transport in Saccharomyces cerevisiae by Micromolar Amino Acid Concentrations. J. Bacteriol. 1987,169, 2132–2136. [CrossRef] 38. Rai, R.; Genbauffe, F.; Lea, H.Z.; Cooper, T.G. Transcriptional Regulation of the DAL5 Gene in Saccharomyces cerevisiae.J. Bacteriol. 1987,169, 3521–3524. [CrossRef] [PubMed] 39. Bleve, G.; Zacheo, G.; Cappello, M.S.; Dellaglio, F.; Grieco, F. Subcellular Localization and Functional Expression of the Glycerol Uptake Protein 1 (GUP1) of Saccharomyces cerevisiae tagged with Green Fluorescent Protein. Biochem. J. 2005 ,390, 145–155. [CrossRef] 40. Feilmeier, B.; Iseminger, G.; Schroeder, D.; Webber, H.; Phillips, G.J. Green Fluorescent Protein Functions as a Reporter for Protein Localization in Escherichia coli. J. Bacteriol. 2000,182, 4068–4076. [CrossRef] 41. Hyde, R.; Cwiklinski, E.L.; MacAulay, K.; Taylor, P.M.; Hundal, H.S. Distinct Sensor Pathways in the Hierarchical Control of SNAT2, a Putative Amino Acid Transceptor, by Amino Acid Availability. J. Biol. Chem. 2007 ,282, 19788–19798. [CrossRef] [PubMed] 42. Zhang, Z.; Albers, T.; Fiumera, H.L.; Gameiro, A.; Grewer, C. A Conserved Na+ Binding Site of the Sodium-coupled Neutral Amino Acid Transporter 2 (SNAT2). J. Biol. Chem. 2009,284, 25314–25323. [CrossRef] [PubMed] 43. Gueldener, U. A Second Set of loxP Marker Cassettes for Cre-Mediated Multiple Gene Knockouts in Budding Yeast. Nucleic Acids Res. 2002,30, 23e. [CrossRef] 44. Breslow, D.; Cameron, D.M.; Collins, S.; Schuldiner, M.; Stewart-Ornstein, J.; Newman, H.W.; Braun, S.; Madhani, H.; Krogan, N.J.; Weissman, J.S. A Comprehensive Strategy Enabling High-Resolution Functional Analysis of the Yeast Genome. Nat. Methods 2008,5, 711–718. [CrossRef]