Marie Curie Actions IRSES Mid-term Report

Marie Curie Actions IRSES Mid-term Report ADDITIONAL INFORMATION Project n : 612441 Project Acronym: YeSVitE Project Full Name: Yeasts for the Sustainability in Viticulture and Oenology Period covered: from 01.01.2014 to 31.12.2014 Period number: 1 Start date of project: 01.01.2014 Project coordinator name: ILEANA VIGENTINI Project coordinator organisation name: UNIVERSITY OF MILAN (UMIL) Date of preparation: 20.12.2014 Date of submission (SESAM): 04.03.2015 Duration: 1 YEAR Version: 1 Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 1/29

4. ADDITIONAL INFORMATION For each task a detailed Work Plan has been prepared by the corresponding participant in charge. The document indicate the specific activities done, on-going and pending to allow a useful assessment of the work carried out during the reporting period. At least a secondment has been linked to each activity, as reported in the Description of the Work of the Annex I. During the first year of the consortium activity, the project appeared well equilibrated from a socio-economic point of view (persons/resources). Although the participants have exploited only 50% of the eligible secondments, all the tasks are in line with the expected date of deliverables. This achievement is mainly due to the reason that most of the programmed tasks have been taken in charge by research groups that have been started an active experimentation on the specific topics before the beginning the project. Task 1.1: New molecular targets to determine taxonomic borders among species and type at strain level (Participant in charge: 2) The analysis of molecular markers enables a state of the art re-identification of strains at the species level, an aspect particularly important considering the systematic complexity of species groups such as that including Saccharomyces cerevisiae. This species has been employed for a long time in food and alcoholic beverage productions. Discrimination at strain level has become a strategic activity for the wine industry since it can link territory, environment and final products for wine valorisation. At the same time, it is useful to assay the origin of a strain, to define the cell linkages, to understand the possible relations between the genetic/physiological traits and the environmental origin, to protect the outcomes of the genetic improvement by a standardized characterization. Moreover, molecular typing improves knowledge on dynamics of microbial population during wine production, verifies the starter domination and allows a fast detection of potential spoiling microorganisms. Actually, no simple and fast method for strain characterization is available so far but the discrimination among yeast isolates is still achieved using approaches like the AFLP and ISS profiling, MLST, or REA-PFGE. Further, the accessibility and comparison of results by internet, representing the major advantage for the research field, it is not possible since databases are poorly represented. The activity of this task will consider: 1. The development of new mathematic models to verify/re-determine taxonomic borders between species; 2. The setup of new protocols for the yeast typing characterised by an easy, fast and cheap usage; 3. The creation of a collection of yeasts with oenological interest and a development of a web-free database collecting the yeast genetic profiles and phenotypes. Objective: To assess the evolutionary linkage among wine-related yeasts from ancient and new vine-growing areas The inter- and intra- specific relationships among wine yeasts are of paramount importance to understand the evolutionary dynamics of these species in different vine-growing areas with diverse types of technology, selective pressure etc. One major problem of the species included in the former sensu stricto group of the genus Saccharomyces is the overlap of one species distribution over the other. This produces serious problems that will be considered in this task in order to gain better understanding and gold standard procedures. Plan of the Activities and Timetable: A1.1/1 Development of a basal (version 1.1) system of species delimitation for oenological species - Participant 2 will produce the mathematical model and will introduce in user friendly environments such as R (www.cran.org) A1.1/2 Creation of a shared and redundant collection of yeasts of oenological interest Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 2/29

- Participant 2 will produce the database (this activity will be carried out in cooperation with the CBS) - All Participants will send their data to Participant 2 - The Participants will exchange isolated strains according to the established strain exchange table A1.1/3 Definition of the oenological species of interest - Participant 2 will collect the data from all Participants and will make a list of all species found within the first year of work - Participant 2 will collect sequence data from the type strains and from the least central strains of the species of interest (this activity will be carried out in cooperation with the CBS) - All Participants will furnish all available information on the species of interest available in their laboratories A1.1/4 Definition of the taxonomic variation found within the project - The variability found within the project and in all different areas will be compared with the currently known inter-specific variability A1.1/5 Search of markers more efficient than ITS and 26S - Participant 2 will test a set of six genes already defined in the Embarc Project in a set of different yeasts - Participant 1 will test four genes already used in the multi-locus sequencing in a set of different yeasts - Participant 5 will test NGS techniques for the massive analysis of strains with several markers simultaneously. - Participant 2 will analyse the statistical and phylogenetic performance of the different loci tested A1.1/6 Updates of the mathematical models - Participant 2 will develop one release per year as an average - Participant 2 will train the other Participants in the use of the software implementing the model A1.1/7 Updates of the biogeographic and taxonomic evaluations - Participant 2 will upgrade the analyses of the activity A1.1/4 every year based on the data provided by all Participants Activity Description Involved Participant Expected time A1.1/1. Definition of basal mathematical model 1-11 2 version 1.0 DONE A1.1/2. Creation of a shared and redundant 1-48 collection of yeasts of oenological 2, 1, 5, 7 interest A1.1/3. Definition of the oenological species of interest 2, 1, 7 8-16 A1.1/4. A1.1/5. Definition of the taxonomic variation found within the project Search of markers more efficient than ITS and 26S 2, 4, 1 A1.1/6. Updates of the mathematical models 2, 1 A1.1/7. Updates of the biogeographic and taxonomic evaluations 2 2, 1, 7 16-24 12-36 18-48 18-48 Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 3/29

Involved Participans: YeSVitE Consortium Secondments: No secondments for this task. Being largely a bioinformatics task, the exchanges will occur via the Internet Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 4/29

Task 1.2: Isolation and characterisation of new strains/species in wine producing areas (Participant in charge: 3) The international wine market is experiencing an ever increasing demand to improve the sensory and nutritional features or the final product. Currently, most wine production is based on the use of starter cultures consisting of selected strains of S. cerevisiae, in order to ensure quick and controlled fermentations. However, the limited real metabolic diversity among commercial S. cerevisiae wine yeast starters, together with the replacement of indigenous microbiota, can contribute to a wine standardization resulting in flattening of taste. In order to gain innovative characteristics of quality and safety, the research will focus on new Saccharomyces and non-saccharomyces yeast strains, isolated from different production areas. Concerning S. cerevisiae, strains with interesting features will be identified from a genetic and a physiological point of view. Data collected will help understanding evolutionary innovations that have enabled S. cerevisiae to competitively exclude other yeast species during fermentations. Concerning other yeast species (i.e. non-saccharomyces), even though they have been known for decades it is only recently (last five or ten years) that they have started to be generally considered as a good alternative in order to impact on wine features and quality, including for example their use to improve aroma complexity or specific aroma notes, reduce volatile acidity, reduce the use of sulfur dioxide or even reduce ethanol content of wines. The activities in this task will involve the isolation of wild yeasts from grapes, musts and wines, in vineyards and in the nearby environmental niches; identification of species diversity; and assessment of yeast performance in economically relevant settings. Objective: To select promising wild yeast strains for winemaking The consortium will evaluate the incidence of the collected yeast strains and their role in the fermentation process. Yeasts that will be discriminated by tasks 1.1 and 1.3 will be analysed for their growth, fermentative performance and for oenological/technological parameters of interest such aroma compounds, enzymatic activities, ability to reduce ethanol content. Plan of the Activities and Timetable: A3.2/2. The isolation of wild type yeasts from different vine-growing countries - Participants 3, 6 and 7, will isolate yeasts from grapes, musts and wines, in vineyards and in the nearby environmental niches. Both Saccharomyces and non-saccharomyces strains will be targeted. A3.2/3. Yeast species identification - Strains will be identified at the species level ITS RFLP, ITS sequencing or D1/Ds sequencing. A3.2/4. Physiological and technological characterization of yeast strains - Strains will be characterized for growth and respiratory capacity in synthetic grape must, production of volatile compounds, and production of enzymes of oenological interest. A3.2/5. Assessment of yeast strain performance - Strains selected on the basis of results from the activities described above will be employed for fermentations in economically relevant settings on the laboratory scale. A3.2/6. Optimization of yeast strain use - Optimal conditions for industrial usage of the most promising strains will be assessed in response to different process conditions, first on the laboratory scale and then directly on grape must under scale-up conditions. A3.2/7. The identification and quantification of species diversity - High-throughput sequencing technologies will be employed in order to assess species diversity and competitive interactions under ecologically and industrially relevant contexts (vineyard and spontaneous fermentation processes). Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 5/29

Activity Description Involved Participant Expected time The isolation of wild type yeasts 13-16; 20-23 A1.2/1. from different vine-growing 1, 3, 6, 7 countries A1.2/2. Yeast species identification 1, 3, 6, 7 13-26 A1.2/3. Physiological and technological characterization of yeast strains 3, 6 15-36 A1.2/4. Assessment of yeast strain performance 3, 6 A1.2/5. Optimization of yeast strain use 3, 6 15-38 25-44 A1.2/6 The identification and quantification of species diversity 6 18-33 Involved Participants: 1-UMIL 3-CSIC 6-SU 7-AUG Secondments: ER5, from Participant 1 to 7, involved in Task 1.2 (M2, milestone) for deliverables D0.2 and D1.2 Objective: Isolation of yeasts from Georgian vineyards Training: Managing of traditional technologies for wine production ER16, from Participant 3 to 7, involved in Task 1.2 (M1 and M2, milestones) for deliverables D1.1 and D1.2 Objective: Set up molecular methods to study the yeast biodiversity Training: To extract genomic DNA, to amplify genes of interest (PCR), to sequence DNA amplicons Training: Metagenomic approach, data elaboration ER26, from Participant 7 to 3, involved in Task 1.2 and Task 4.1 (M2 and M4, milestones) for deliverables D1.2 and D4.1 Objective: Isolation of Yeasts from Spanish vineyards (grape and insects) vintage 2015 Training: Cultural isolation techniques, set up of a Spanish wine-related yeast collection ER26, from Participant 7 to 3, involved in Task 1.2 and Task 4.1 (M2 and M4, milestones) for deliverables D1.2 and D4.1 Objective: Isolation of Yeasts from Spanish vineyards (grape and insects) vintage 2016 Training: Update of the wine-related Spanish yeast collection Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 6/29

Task 1.3: Study of phylogenetic relationships and genome structure (Participant in Charge: 4) Tapping the biodiversity is one of the approaches to make biotechnological processes more sustainable. This is also true for wine production, especially in the wineries that practice spontaneous fermentations. Wine production is by its nature dependent on ever-changing environmental conditions, and in case of spontaneous fermentations also on the environmental impact on the micro-organisms (yeasts) which are responsible for grape must fermentation. Besides environmental conditions, however, also the available genetic pool of the yeast strains is crucial. For instance, in the old vine growing areas such as Caucasus region or Europe, yeast strains with desirable fermentation characteristics can be isolated from the vineyards or wine cellars. In contrast, in new vine growing areas, autochthonous yeast strains often exhibit undesirable and potentially disruptive fermentation characteristics. Our understanding of fermentation and aroma-producing potential of indigenous yeast strains is however largely empirical or even anecdotal, and predicting an outcome of fermentation based on the identification of strains present in must is as yet impossible. Genomes of a dozen commercially used wine strains have been sequenced until now, and their divergence is relatively high, ranging from 2 to 7 SNPs per kilobase. Additional wine strains will be sequenced during the course of this project. The first step toward rational approaches for exploiting the biodiversity of fermentative yeast strains is therefore to understand their phylogenetic relationships and genomic makeup, which later on should enable identification of genetic factors upon which an outcome of fermentation could be predicted. The sequencing of the variable D1/D2 region of the large (26S) subunit of the ribosomal DNA locus is one of the cornerstones for strain genotyping and subsequent phylogenetic analyses. For the strains with known differences in the desired traits characteristics, also sequencing of the genes/chromosome regions known to be important for these characteristics, or alternatively whole genomes should be used to compare different genotypes when the causative genetic elements are unknown. To identify such elements, quantitative trait loci mapping has recently become the method of choice also in yeast, and this method has been used successfully in studying the polygenic traits important for wine fermentation, such as ethanol production, aroma compounds composition or ethanol tolerance. Objective: To determine phylogenetic relationships between collected strains For comparison of oenological interest of the collected strains, their intra- and interspecies relationships will be required. The analysis of fast evolving loci and the setup of new genotyping approaches will be instrumental in reaching this goal. Plan of the Activities and Timetable: A1.3/1. Dissemination of methods for genome analysis - Participants 4 and 5 will provide knowledge and expertise on genome analysis methods to other Participants A1.3/2. Genome analysis and phylogenetic comparison of collected strains - Participant 4, with the help of Participant 5 and other Participants, with will perform genome analysis and phylogenetic analysis of collected strains Activity Description Involved Participants Expected time Dissemination of methods for 6-48 A1.3/1. The YeSVitE consortium genome analysis A1.3/2. Genome analysis and phylogenetic comparison of collected strains Involved Participants: 1-UMIL 2-UNIPG 3-CSCI The YeSVitE consortium Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 7/29 6-48

4-JSI 5-UTO 6-SU 7-AUG Secondments: ER1, from Participant 1 to 5, involved in Task 1.3 (M6, milestone) for deliverable D1.3 Objective: Comparative genomics among yeasts Training: Next-generation sequencing, use of software for the genetic comparison of genomes ER7, from Participant 1 to 5, involved in Task 1.3 (M6, milestone) for deliverable D1.3 Objective: Exploration of yeasts phylogenetic relationships and sequencing of gene of interest Training: Use of software for clustering analysis and elaboration of SNPs ESR13 (not eligible secondment), from Participant 2 to 4, involved in Task 1.3 (M6, milestone) for deliverable D1.3 Objective: Yeast phylogenetic analysis Training: Next-generation sequencing, assembly, and analysis ESR14, from Participant 3 to 5, involved in Task 2.2 (M2 and M6, milestones) for deliverable D2.2 and D1.3 Objective: Development of new tools to increase or decrease the expression of certain yeast genes Training: Yeast genetics engineering ER18, from Participant 4 to 6, involved in Task 1.3 and Task 4.1 (M2, M4 and M6, milestones) for deliverables D1.3 and D4.1 Objective: Phylogenetic comparison of South African wine yeast strains. Determine the richness of winerelated microbial populations Training: Methods for metagenomic analysis of microbial populations on grapes and in wine. 26S rdnabased phylogenetic analysis methods ER18, from Participant 4 to 5, involved in Task 1.3 (M2 and M6, milestones) for deliverables D1.3 Objective: To determine the phylogenetic relationships between selected newly isolated/sequenced yeast wine strains Training: Custom-made methods for NGS yeast genome data assembly and annotation ER18, from Participant 4 to 5, involved in Task 1.3 (M2 and M6, milestones) for deliverables D1.3 Objective: To determine phylogenetic relationships between selected newly isolated/sequenced yeast wine strains Training: Elaboration of phylogenetic data ER19, from Participant 4 to 6, involved in Task 1.3 and Task 4.1 (M2, M4 and M6, milestones) for deliverables D1.3 and D4.1 ER19, from Participant 4 to 5, involved in Task 1.3 (M2 and M6, milestones) for deliverables D1.3 Objective: To determine the phylogenetic and phenotypic relationships between selected newly isolated/sequenced yeast wine strains Training: Novel high-throughput yeast phenotyping methods. Custom-made methods for NGS yeast genome data assembly and annotation. ESR21 (not eligible secondment), from Participant 5 to 4, involved in Task 1.3 (M6, milestone) for deliverable D1.3 Objective: Yeast phylogenetic analysis Training: Next-generation sequencing, assembly, and analysis ESR25, from Participant 6 to 4, involved in Task 1.3 (M2 and M6, milestones) for deliverable D1.3 Objective: Exploration of yeasts phylogenetic relationships collected in Spain, Georgia and South Africa; in silico identification of wine strain relevant genomic loci Training: Bioinformatic tools and use of software for clustering analysis ER27, from Participant 7 to 4, involved in Task 1.3 (M6, milestone) for deliverable D1.3 Objective: Exploration of yeasts phylogenetic relationships collected in vintages 2014-2015 in Spain and Georgia Training: Use of software for clustering analysis Task 2.1: Production of non GMO yeasts for the participative selection of strains by the wineries Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 8/29

(Participant in charge: 2) The rationale of this approach is based on the ability to sporulate, retained by many (although not by all), wine strains of teleomorphic species such as S. cerevisiae, Hanseniaspora uvarum, etc. In the case of homothallic strains the spores can undergo a switch of the mating type, producing roughly equal numbers of haploid cells of either type in the same colony, giving rise by conjugation to diploid cells homozygous at all loci. These cells represent extreme genetic combinations, especially for the quantitative trait loci so important in the fermentative performance of strains. Alternative strategies have been developed for hemihomothallic and heterothallic strains. In order to meet the concerns of the European consumers, oenologists are interested to develop non-gmo improved strains for the production of wines with specific sensory characteristics and low ethanol content. The task will consider: 1-The possibility to subject cells to direct high density selection, allows to get the best possible strains for the designed conditions and/or to produce a set of interesting new strains for more detailed analyses that cannot be carried on plates, but rather require growth in liquid media of each single improved strain. Objective: To develop improved strains for the production of wines with specific sensory characteristics and/or low ethanol content The scientific purpose of the WP2 will be achieved through the use of classic genetic tools and recombinant DNA technology. The tasks will generate and investigate new strains with improved metabolic traits that could have an impact in terms of the wine sustainability. In particular, the main outcomes of this WP will be the selection of yeasts that produce specific aromas or that have a modified carbon metabolism useful for the production of low-alcohol wines. Plan of the Activities and Timetable: A2.1/1 Preparation to the Production of new strains with classical Genetics tools - All Participants will send to participant 2 a list of desired selectable oenological relevant characters - All Participants will send to participant 2 their best (still sporulating) yeasts of oenological interest A2.1/2 Production of new strains with classical Genetics tools - Participant 2 will produce new strains of oenological interest - Participant 1 will test their level of homozygosity by multilocus SSR analysis A2.1/3 Primary technological tests of the new strains - Participant 2 and 6 will test the relevant technological traits of the new strains at the laboratory scale A2.1/4 Metabolomic testing of the new strains (to be merged in 3.1) - Participant 2 will test the new strains new means of FTIR - Participant 5 will test the new strains with metabolomics tools A2.1/5 Technological testing of the new strains (to be merged in 3.3) - Participant 2 will prepare the strains and carry out the first preliminary tests - Participant 6 will carry out pilot fermentations and metabolomics evaluations Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 9/29

Activity Description Involved Participants Expected time A2.1/1. A2.1/2. Preparation to the Production of new strains with classical Genetics tools Production of new strains with classical Genetics tools 2 and all Participants for the requested information 6-12 DONE 1, 2 12-36 A2.1/3. Primary technological tests of the new strains 1, 2, 4, 6 IN PROGRESS A2.1/4. Metabolomic testing of the new strains (to be merged in 3.1) 2, 5 20-40 A2.1/5. Technological testing of the new strains (to be merged in 3.3) 2, 6 30-46 Involved Participants: 1-UMIL 2-UNIPG 4-JSI 5-UTO 6-SU Secondments: ESR2 (not eligible secondment), from Participant 1 to 2, involved in Task 2.1 (M7, milestone) for deliverable D2.1 Objective: To improve yeasts by classic genetic manipulation Training: Ascus Dissection, Hybridisation spore-spore, Tetrad analysis, Fusion of spheroplasts ESR9 (not eligible secondment), from Participant 1 to 2, involved in Task 2.1 (M5, milestone) for deliverable D2.1 Objective: To improve yeasts by classic genetic manipulation Training: Sporulation of yeasts, Hybridisation spore-spore ESR23, from Participant 6 to 2, involved in Task 2.1 (M5, milestone) for deliverable D2.1 Objective: To improve yeast by traditional methodologies Training: Sporulation of yeasts, Ascus Dissection, Hybridisation spore-spore, Tetrad analysis, Fusion of spheroplasts Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 10/29

Task 2.2: Development of new tools to increase or decrease the expression of certain yeast genes (Participant in Charge: 3) The Saccharomyces/Kluyveromyces sensu lato contains, apart from the most usual wine yeast S. cerevisiae, also the so called non-conventional Saccharomyces species which play a role in certain food applications, such as in lager beer fermentation for S. pastorianus. S. cerevisiae x S. kudriavzevii hybrids produce wine with a significantly higher amount of higher alcohols (ester precursors). In addition, non-saccharomyces yeast strains, always present during the first stages of alcoholic fermentation, are gaining enological interest, due to their positive contribution to several wine features (also to their potential spoilage activities). Hundreds of these yeasts have recently been analyzed for the sequence of their genomes (mostly Saccharomyces). So far only for a few yeast species molecular genetic tools have been developed, like a transformation system, directed gene knock-out protocols, etc. Surprisingly, very few wine yeasts can be manipulated in the lab using genetic or molecular biology approaches. Molecular biology studies should help understanding (and further biotechnological approaches) the contribution of these not so well known yeast species to wine features. In principle, putative novel aroma or carbon metabolism associated genes obtained should be either deleted or their expression diminished in the target strains. Afterwards the mutated strains can be tested for their phenotype profiles. If the mutated gene is indeed responsible for a certain aroma compound, then the aromatic profile should be changed. This approach requires development of molecular genetic tools for poorly characterized yeasts. The bacterial CRISPR-Cas immune systems have recently been developed for genome engineering in S. cerevisiae. It shows apparent advantages over RNAi approaches in yeast. The possibility of extending the use of this system to alternative yeast species will be explored in this task. Objective: To develop molecular tools to understand the contribution of different yeast species to wine quality The tasks will generate transformation protocols for non-conventional Saccharomyces species or non- Saccharomyces yeast strains. Systems to direct homologous recombination, especially for gene knockout will also be developed. Plan of the Activities and Timetable: A3.2/1. Selection of yeast strains for proof of concept - Participants 1, 3 and 5, will design the three most interesting oenological yeast species to by target by genetic engineering. A3.2/2. Identification of suitable selection markers - Selected trains will be transformed with either autonomously replicating (at least in S. cerevisiae) plasmids or integrative plasmids carrying dominant selectable markers. Standard S. cerevisiae transformation protocols will be used in first instance. Basal tolerance of the host strains will be stablished in advance for each antimicrobial or selection method. The results will allow to stablish marker suitability for each strain. A3.2/3. Optimization of transformation conditions - For the more suitable strain/marker pairs, transformation parameters will be adjusted in order to have the better possible transformation efficiency. At least two markers per strain will be required for further steps. A3.2/4. Constitutive Cas9 expression - Construct allowing for the constitutive expression of Cas9 in yeast (with endogenous or S. cerevisiae promoters) will be transformed in each strain. A3.2/5. Homologous recombination - Transient gdna cassettes will be constructed and transformed in the above mentioned strains. Model target genes will be chosen taking into account assay simplicity. After functional verification, recombinant strains will be analysed in order to confirm the viability of the system at the molecular level. Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 11/29

Activity Description Involved Participants Expected time Selection of yeast strains for proof 13-16 A2.2/1. 1, 3, 5 of concept A2.2/2. Identification of suitable selection markers 1, 3, 5 18-24 A2.2/3. Optimization of transformation conditions 1, 3, 5 18-36 A2.2/4. A2.2/5. Constitutive Cas9 expression Homologous recombination 1, 3, 5 1, 3, 5 30-38 35-48 Involved Participants: 1-UMIL 3-CSIC 5-UTO Secondments: ESR9 (not eligible secondment), from Participant 1 to 3, involved in Task 2.2 (M5, milestone) for deliverable D2.2 Objective: knock-down the expression of a certain gene Training: Use of CRISPR-Cas system or alternative approaches to silence genes ESR14, from Participant 3 to 5, involved in Task 2.2 (M2 and M6, milestones) for deliverable D2.2 and D1.3 Objective: Development of new tools to increase or decrease the expression of certain yeast genes Training: Yeast genetics engineering Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 12/29

Task 3.1 Metabolomic assessment of new yeast strains under oenological conditions (Participant in Charge: 5) The FTIR (Fourier Transform InfraRed Spectroscopy) technique is a chemical analytical procedure based on the different response of diverse chemical compounds to the irradiation with Infrared light. This procedure has been used to analyse the metabolome (i.e. the whole set of metabolic compounds in a cell) of whole cells, with particular emphasis on microbes, without any need of previous preparatory steps. The technique output is a spectrogram reporting the wavenumbers (cm -1 ) on the X axis and the intensity on the ordinates. The mathematical and statistical analysis of the spectrogram is used to process raw data and gain insight on the physiological state of the cell. The technique is characterized by a very high level of reproducibility (each spectrum is the average of 256 reads), further improved by the possibility of carrying out several independent and machine replicas simultaneously. The very low cost in terms of consumables allows to development of very complex designs of experiment with several variables. The metabolome of the yeast will also be profiled using full scan LC-MS metabolomics to identify altered compounds in the small molecule metabolome. One of the open questions regarding the impact of fermentative yeast on wine aroma is which intermediates in metabolism are the precursors for specific aroma compounds. We will therefore combine the results of aroma compound profiling and yeast cells metabolomics to identify potentially originator intermediates. Additionally, the metabolomics profiles will be used as phenotypic traits and by superimposing their similarities with the genotype-based phylogeny (Task 1.3), we should be able to estimate which metabolic pathways have evolved because of the local environmental constraints, and which have gradually evolved during the separation of the European and North American strains. Sugar metabolism intermediates profiles of a number of wine strains under the same conditions, mimicking wine fermentation, will be generated. The full scan metabolomics approach measures the steady-state level of hundreds of known compounds, including glycolytic and TCA intermediates, and additionally can identify novel compounds specific to one or several strains. The measurement of metabolite levels will be complemented with kinetic flux profiling analysis to measure the rate of glycolysis, the TCA cycle, and the pentose phosphate pathway. Since also genetic/genomic data will be available for these selected strains, we will be able to address the question whether all wine strains with different genetic makeup use a single metabolic strategy to rapidly produce high concentration of ethanol, or do different scenarios, leading to a common final result, exist. Objectives: To exploit metabolomic profiles for yeast discrimination at species/strains level (FTIR approach) The present objective will compare metabolomic data with genetic profiles of different species/strains to test the FT-IR approach as rapid technique to discriminate among yeasts; To study the impact of fermentative yeast on wine aroma using metabolomics (LC-MS approaches) This goal will elaborate data from aroma compound profiling and yeast cells metabolomics to identify the similarities with the genotype-based phylogeny. This activity will be useful to estimate which metabolic pathways have evolved because of the local environmental constraints, and which have gradually evolved during the separation of the European and North American strains. Plan of the Activities and Timetable: A3.1/1. Yeast discrimination at species/strain level by FT-IR Approach - Participant 2 will analyse and evaluate the metabolomic fingerprint at the species/strain level of yeasts grown in highly standardized conditions using the FT-IR approach that returns a very detailed fingerprint (920 continuous descriptors). Aim of this activity is to establish an effective way of metabolomics discrimination at the geographic, strain or species level. A3.1/2. Definition of the physiological/stress state of the cells (FT-IR and LC-MS approaches) - On the basis of the clustering inherent to the activity A3.1/1, Participant 2 will select a limited number of strains for further metabolomic analysis which will be carried out by Participant 5. Participant 5 will evaluate the similarities of fermentative/aroma profiles with the genotype-based phylogeny. This activity will be useful to estimate which metabolic pathways have evolved because Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 13/29

of the local environmental constraints, and which have gradually evolved during the separation of the European and North American strains. A3.1/3. Advanced clustering - Participant 4 will gather FTIR spectra and LC-MS data from Participant 2 and Participant 5 respectively. Clustering of FTIR profiles with different statistical approaches will be tested along with Participant 2 in order to define the best analytical pipeline for massive metabolomics analyses. LC-MS and FTOR data from the selected strains will be compared with various statistical tools, in order to data-mine both series of data and find useful correlations that will allow in future to predict LC-MS data from the corresponding FTIR profiles. Activity Description Involved Participants Expected time A3.1/1. Yeast discrimination at species/strain 12-40 2 level by FT-IR Approach A3.1/2. Definition of the physiological/stress 24-36 state of the cells (FT-IR and LC-MS 2, 4 approaches) A3.1/3. Advanced clustering 4, 5 12-40 Involved Participants: 2-UNIPG 4-JSI 5-UTO Secondments: ER10, from Participant 2 to 5, involved in Task 3.1 (M5, milestone) for deliverable D3.1 Objective: Definition of the physiological/stress state of the cells Training: Data elaboration of metabolomics results ER11, from Participant 2 to 5, involved in Task 3.1 (M5, milestone) for deliverable D3.1 Objective: Definition of the physiological/stress state of the cells Training: Data elaboration of metabolomics results ER12, from 2 to 5, involved in Task 3.1 (M5, milestone) for deliverable D3.1 Objective: Definition of the physiological/stress state of the cells in modelled wine fermentations Training: LC-MS metabolomics ESR13, from Participant 2 to 5, involved in Task 3.1 (M5, milestone) for deliverable D3.1 Objective: Definition of the physiological/stress state of the cells Training: Analysis of the yeast metabolome by LC-MS ESR17, from Participant 4 to 5, involved in Task 3.1 (M5, milestone) for deliverable D3.1 Objective: Definition of the physiological/stress state of the cells and kinetic flux profiling analysis of wine yeast strains in modelled wine fermentations Training: LC-MS metabolomics Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 14/29

Task 3.2: Molecular mechanisms operated by spoilage yeasts to tolerate sulphur dioxide exposure (Participant in Charge: 1) Yeast spoilage in beverages has become of increasing importance because of the use of less-severe processing, the great variety of new formulations and the tendency to reduce the use of preservatives. Sulphur dioxide is widely used as an antioxidant agent and an antimicrobial compound in oenology. Nevertheless, it is known that it has allergenic effects on humans; however carcinogenicity has been not found. It would be important to reduce the quantity of SO 2 in wine to avoid a cumulative effect since it is a very common additive in many foods. Winemaking process can involve dozen different yeast species some of which usually produce a partial fermentation of sugars, increasing in volatile acidity and releasing of unpleasant aromas. The spoilage species in wines include Dekkera/Brettanomyces, Pichia, Schizosaccharomyces and Zygosaccharomyces. A common feature of these genera is their ability to tolerate the concentration of sulphite normally used during wine production. For example, the capability of Dekkera/Brettanomyces bruxellensis to survive and to grow in barrel-aged wine can be partially ascribed to its high resistance to SO 2, but no data are available on the mechanism that controls this metabolic trait. A better control in the sulphur dioxide manage during winemaking could avoid both the yeast spoilage and, in a sustainable perspective, to limit sulphite in bottled wines. As concern S. cerevisiae, sulphite is also a normal metabolite produced during reductive sulphate assimilation pathway. The question of how endogenous toxicity is avoided can be linked to an efficient regulation system that would minimize pools of intermediates, sulphite among them. At high SO 2 concentrations S. cerevisiae shows an increased level in the expression of several genes, among them SSU1, FZF1, ALD6, MET16. Several reports proved that a high biodiversity is present in D./B. bruxellensis species both from the genetic and physiological point of view. This polymorphism was also observed in a previous work by the Participant 1 of the proposed network where sensitive, low, medium and high SO 2 resistant strains were recognised. The RNA-Seq is a recently developed approach to transcriptome profiling that uses deep-sequencing technologies. RNA-Seq provides a far more precise measurement of levels of transcripts and their isoforms than other methods. The resulting sequence reads will be aligned with the reference genome or transcriptome, and classified (exonic reads, junction reads and poly(a) end-reads) to gain the ORFs annotation. These three types will be used to generate a base-resolution expression profile for each gene. Objective: To improve the managing of sulphur dioxide addition during winemaking to reduce spoilage yeasts and decrease the level of sulphite in wines The capability of spoilage yeasts to survive and to grow in barrel-aged wine can be partially ascribed to their high resistance to SO 2, but no data are available on the mechanism that controls this metabolic trait. A better control in the sulphur dioxide manage during winemaking could avoid both the yeast spoilage and, in a sustainable perspective, to limit sulphite in bottled wines. Plan of the Activities and Timetable: A4.4/1. Selection of sensitive and resistance strains to SO 2 - Participant 1 will transfer to Participant 4 a collection of D./B. bruxellensis strains that will be classified for their capability to growth under different concentrations of sulphite. Participant 4 research unit will perform the analysis using a custom-made robotic manipulator for highthroughput phenotyping of yeast strains. A4.4/2. Yeast biomass for transcriptome analysis by RNA-Seq - Strains will be grown in bioreactor under controlled oenological conditions (ph, temperature, oxygenation, etc.) to obtain the yeast biomass that will be further analysed in RNA-Seq methodology. The yeast growth will carried out in a wine-like medium; for resistance strains, sulphur dioxide will be added to the medium at a useful concentration. A4.4/3. Production of cdna libraries from yeast biomasses collected at point 2. of the Plan of the Activities Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 15/29

- Conversion into a library of cdna fragments of the population of RNA (such as poly(a)+) obtained from through either RNA fragmentation (RNA hydrolysis or nebulation) or DNA fragmentation (DNasiI treatment of sonication). A4.4/4. Analysis of the transcriptomes using the RNA-Seq approach - cdna reads will be assembled and compared to the available whole genomes of D./B. bruxellensis to obtain a full picture of the main genes involved in the SO 2 stress response. A4.4/5. Study of the molecular mechanisms involved in SO 2 tolerance in D./B. bruxellensis - Overexpression and gene deletion of genes involved in SO 2 tolerance will be carry out to study the molecular pathways activated by D./B. bruxellensis to counteract sulphite exposure. When necessary, molecular tools for D./B. bruxellensis will be make available by Participant 3 of the consortium. A4.4/6. SO 2 management in winemaking - The understanding of the molecular mechanisms involved in SO 2 tolerance in D./B. bruxellensis will be exploited to set up innovative and sustainable protocols for SO 2 addiction during winemaking. With the aim to reduce sulphite in wines improving SO 2 managing from grape ripening to wine aging, resistant strains to sulphur dioxide will be used to contaminate microvinifications that will be carried out in experimental cellar using. Cell survival, Viable but Not Culturable state (VBNC) of cell, production of off-flavours and sulphite levels in final products will be follow during winemaking and aging to assay the effectiveness of the procedures. Activity Description Involved Participants Expected time 1-6 A3.2/1. Selection of sensitive and resistance strains to SO 2 1, 6 DONE A3.2/2. A3.2/3. A3.2/4. A3.2/5. Yeast biomass for transcriptome analysis by RNA- Seq Production of cdna libraries from yeast biomasses collected during the activity A3.2/2 Analysis of the transcriptomes using the RNA-Seq approach Study of the molecular mechanisms involved in SO 2 tolerance in D./B. bruxellensis Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 16/29 1, 6 1, 6 6-12 12-15 4, 1 15-18 1, 6 18-33 A3.2/6. SO 2 management in winemaking 1, 6 33-36 Involved Participants: 1-UMIL 4-JSI 6-SU 7-AUG Secondments: ER4, from Participant 1 to 7, involved in Task 3.2 (M3, milestone) for deliverable D3.2, Activity A3.2/1 Objective: Selection of sensitive and resistance strains to SO2 Training: Management of B. bruxellensis fermentations in oenological condition ER27, from Participant 7 to 1, involved in Task 3.2 (M3, milestone) for deliverable D3.2, Activity A3.2/2 Objective: Setup of controlled fermentations of D. bruxellensis in bioreactor (B. bruxellensis CBS2499) Training: Fermentation technology in oenological conditions (batch cultivation) ESR22, from Participant 6 to 1, involved in Task 3.2 (M3, milestone) for deliverable D3.2, Activities A3.2/2 and A3.2/3 Objective: Controlled fermentations of D. bruxellensis in bioreactor (B. bruxellensis AWRI1499)

Training: Fermentation technology in oenological conditions (batch cultivation) ESR8, from Participant 1 to 4, involved in Task 3.2 (M3, milestone) for deliverable D3.2, Activities A3.2/4 Objective: To study the transcriptional response of D. bruxellensis under SO 2 stress by RNA-Seq approach Training: Expression analysis Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 17/29

Task 3.3: Reduction of ethanol content in wines by use of non-saccharomyces yeasts (Participant in Charge: 6) Over the past twenty years, alcohol levels in wines have increased significantly. This trend, observed in many producing areas, is linked to various factors, including the selection of grapes with a high sugar yield, the global warming and the evolution of winemaking practices. The demand for reducing the alcohol level has led the industry and research to consider different strategies that rely on interventions at various phases such as the selection of vine-plant or physical interventions for removing sugar in musts or alcohol in wines, or genetic manipulation of yeast. In the past, a strategy was represented by the classic genetics (spore dissection or hybridisation), nowadays discarded because of its laborious and time-consuming character. An option is the use of metabolic engineering of yeast strains that produce less alcohol for the same amount of consumed sugar, redirecting the carbons towards other metabolites. An example is glycerol that represents the most abundant by-product of alcoholic fermentation after ethanol and CO 2 and that contributes to the body of wines. The overexpression of glycerol-3-phosphate dehydrogenase has allowed efficient reduction of the ethanol yield, up to 15-20%, however the glycerol overproduction is accompanied by major changes in the level of other metabolites, including some of which undesirable in wine, notably acetate and acetoin. Although other pathways were engineered to adjust these defects (i.e. conversion of acetoin into 2,3- butanediol, a sensory-neutral compound), too many efforts are needed to redirect the carbon fluxes without altering yeast properties and the sensory traits of wine. An attractive possibility could be the use of selected S. cerevisiae strains combined with non-saccharomyces yeasts,, which are poorer ethanol producers, that at an appropriate ratio or fermentation time can drive the carbon metabolism through a decreased ethanol yield. Actually, some producer of starter cultures have introduced on the market non-conventional yeasts for wine production but their first aim is referred to the modification/improving of the wine quality. Indeed, several yeasts, which separated from the Saccharomyces lineage just before the whole genome duplication event, are known to produce much less ethanol (are less Crabtree positive), among them are Torulaspora, Lachancea and Zygosaccharomyces yeasts. Accurate phenotyping of the strains is crucial for the evaluation of their fermentation capabilities. To rapidly identify the strains with potential reduced ethanol production capability, a custom-built robotic manipulator that enables replica plating of up to 4000 yeast strains per day will be used in combination with a novel method for growth rate measurements of a large number of strains simultaneously. The tolerance to higher concentration of ethanol in the medium will be used as a proxy to identify among the studied strains those that themselves produce less ethanol. In the second step, an enzyme-based assay to measure the produced ethanol and glycerol in a semi-high-throughput manner will be used to identify the strains with the desired trait. Objective: To reduce ethanol content in wines by the setup of innovative protocols for the inoculation of musts with new starter cultures The demand for reducing the alcohol level has led the industry and research to consider different strategies that rely on interventions at various phases such as the selection of vine-plant or physical interventions for removing sugar in musts or alcohol in wines, or genetic manipulation of yeast. An attractive possibility could be the use of selected S. cerevisiae strains combined with non-saccharomyces yeasts, which are poorer ethanol producers, that at an appropriate ratio or fermentation time can drive the carbon metabolism through a decreased ethanol yield. Plan of the Activities and Timetable: A3.3/1. Selection of non-conventional yeasts of oenological origin - Strains will be screened for: (1) their sugar to ethanol yield and (2) their volatile compound production. The aim of this task is to select strains with a lower sugar to ethanol yield than Saccharomyces cerevisiae, the traditional wine yeast, and with the ability to produce aroma compounds of oenological relevance while producing no or low amounts of off-flavour compounds. A3.3/2 Testing the influence of the selected non-conventional yeasts in co-inoculation with S. cerevisiae Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 18/29

- As non-conventional yeasts cannot ferment grape juice to dryness, co-inoculation with S. cerevisiae is prescribed. This task aimed at checking whether co-inoculating the yeasts selected in task A3.3/1 with S. cerevisiae will result in a lower ethanol content than inoculating with a pure culture of S. cerevisiae. A3.3/3 Testing the influence of biotic and abiotic factors on the production of ethanol - The influence of various factors, especially dissolved oxygen level and inoculum ratios, on ethanol yield and cell survival, will be investigated. A3.3/4 Micro-vinifications - To gain this point micro-vinifications will be carried out in controlled fermenters at the experimental cellar of Participant 6 (Stellenbosch University) and Participant 1 (University of Milan in collaboration with the Vine and Wine Research Centre, Riccagioia, Pavia, Italy). A3.3/5 Investigating acetyl metabolism as a means to lower ethanol yield - Activated acetyl, either in the form of acetyl-coa or acetyl carnitine, is a core metabolite of central carbon metabolism, and controls the essential juncture between catabolic energy metabolism and most major anabolic reactions. Understanding the regulation of this metabolite will directly inform methods to redirect carbon-flux in yeast to lower ethanol yields. Activity Description Involved Participants Expected time A3.3/1 Selection of non-conventional yeasts 3, 6 1-12 DONE A3.3/2 Testing the influence of the selected 1-12 non-conventional yeasts in coinoculation with S. 3, 6 DONE cerevisiae A3.3/3 Testing the influence of biotic and abiotic factors on the production of ethanol 3, 6 A3.3/4 Microvinifications 1, 3, 6, 7 A3.3/5 Involved Participants: P1-UMIL P3-CSIC P4-JSI P6-SU P7-AUG Investigating acetyl metabolism as a means to lower ethanol yield 4, 6 1-24 1-48 13-18 Secondments: ESR17, from Participant P4 to P6, involved in Task 3.3 (M4 and M7, milestones) for deliverables D3.3 and D4.4 Objective: Yeast Population dynamics in controlled fermentation with non-conventional yeasts Training: Metagenomic approach ER27, from Participant P7 to P1, involved in Task 3.3 (M7, milestone) for deliverable D3.3 Objective: Must fermentation trials to produce low alcohol wines Training: Winemaking procedures Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 19/29

Task 4.1: Interplay between grape and yeasts/microbes in the vineyard/nature (Participant in charge: 6) While during the last decades wine fermentations were run by commercially added yeast, now there is a progressive tendency to promote spontaneous fermentations, which rely on microbes present on the grapes and those transmitted around by insects, and leading to re-establishment of natural interactions in the complex plant-microbe-insect. Some preliminary studies show that there is a lively interaction between the three Participants but a detailed inventory of the biodiversity potential has so far not been determined. In nature, yeasts and other microbes occupy a commonly overlooked trophic level between plants and insects. Plants and especially their fruits provide a rich carbon source (sugars) which represent food for a number of microbes. Often yeasts, vectored by insects, win in this competition and represent at the end of fermentations a major part of the microbe biomass. We have recently shown that there is an enormous number of yeast species associated with fermentations. It is now believed that yeast odours represent a critical signal to establish the insect-fruit-microbe relationship and communities. Objective To study the role of yeasts in the vineyard ecosystem While wine fermentations were run by selected commercial starters (ADY) during the last decades, today there is a progressive tendency to promote spontaneous fermentations, which rely on microbes present in vineyards, and re-establishing natural interactions between plants-microbes-insects. This objective wants to study the role of yeasts in vine-growing areas as potential attractor for insect to grapes and to analyse different landscapes to understand if there is a correlation between microbial environment and product. Plan of the activities and Timetable: A4.1/1 Yeast isolation from Georgian grapevine cultivars - Yeast isolates obtained from Georgian grapevine cultivars by Participant 7 will be identified through various molecular techniques by Participants 1 and 7; and analyzed for different phenotypic characteristics. Participants 1, 5 and 7 will jointly employ various statistical packages to evaluate the biogeography of the Georgian grapevine microbiome. A4.1/2 Quantification of yeast biodiversity - Participants 2 and 6 will use data generated from South African, Italian and Georgian vineyard microbiome analysis to determine microbial community structures in the vineyards and how these are influence by agricultural practices. A4.1/3 To assess the dynamic of microbial population in real grape must - Participants 3 and 6 will evaluate the dynamics and yeast populations in spontaneous fermentation of real must and how the dynamics are influenced by standard winemaking practices. The trends of non-saccharomyces yeast populations will be compared. A4.1/4 Phylogenetic comparison of South African wine yeast strains. Determine the richness of wine-related microbial populations. - Participant 6 will transfer to Participant 4 a collection of Saccharomyces cerevisae and several non- Saccharomyces yeast species for molecular typing and phenotypic characterization using highthroughput robotic phenotyping methods. A4.1/5 Identification of the genetic pool of South African wine-related microbial populations - Participant 6 will determine the microbial diversity associated with different grape cultivars in South Africa using metagenomic approaches as well as culture dependent methods. Isolates obtained will be transferred to Participant 4 for further typing. A4.1/6 Isolation and characterization of and yeasts associated with grapevines in various countries Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 20/29

- Participants 3 and 6 will employ multivariate data analysis as well as network algorithms to evaluate the relationship between grape associated microbiomes from different countries. Data will be obtained from various participating Participants. A4.1/7 Isolation of Yeasts from Spanish vineyards (grape) vintage 2015 and 2016 - Participants 7 and 3 will be involved in isolation of yeast associated with grapevine over two consecutive harvest seasons. Activity Description Involved Participants Expected time A4.1/1 Yeast isolation from Georgian grapevine cultivars 1, 5, 7 1-6 DONE A4.1/2. Quantification of yeast biodiversity 2, 6 19-22 A4.1/3 A4.1/4. A4.1/5. A4.1/6. A4.1/7. A4.1/7 To assess the dynamic of microbial population in real grape must Phylogenetic comparison of South African wine yeast strains. Determine the richness of wine-related microbial populations Identification of the genetic pool of South African wine-related microbial populations Isolation and characterization of and yeasts associated with grapevines in various countries Isolation of Yeasts from Spanish vineyards (grape) vintage 2015 Isolation of Yeasts from Spanish vineyards (grape) vintage 2016 3, 6 30-33 4, 6 16-19 4, 6 26-29 6, 3 28-31 7, 3 21-24 7, 3 33-37 Involved Participants P1-UMIL P3-CSIC P5-UTO P6-SU P7-AUG Secondments ER6, from 1 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: To asses growth performance of mixed starter cultures Training: Controlled wine fermentation, Metagenomics on must and wine samples ESR2, from Participant 1 to 7, Objectives: Yeast isolation from Georgian grapevine cultivars Training: Microbial protocols for yeast isolation ER12, from Participant 2 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: Quantification of yeast biodiversity Training: Metagenomics on must and wine samples ER16, from Participant 3 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: To assess the dynamic of microbial population in real grape must Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 21/29

ER18, from Participant 4 to 6, involved in Task 1.3 and Task 4.1 (M2, M4 and M6, milestones) for deliverables D1.3 and D4.1 Objective: Phylogenetic comparison of South African wine yeast strains. Determine the richness of winerelated microbial populations Training: Methods for metagenomic analysis of microbial populations on grapes and in wine. 26S rdnabased phylogenetic analysis methods ER19, from Participant 4 to 6, involved in Task 1.3 and Task 4.1 (M2, M4 and M6, milestones) for deliverables D1.3 and D4.1 Objective: Identification of the genetic pool of South African wine-related microbial populations Training: 26S rdna-based phylogenetic analysis methods. Methods for metagenomics analysis of microbial populations on grapes and in wine ESR24, from Participant 6 to 3, involved in Task 4.1 (M4 and M7, milestones) for deliverable D4.1 Objective: Isolation and characterization of and yeasts associated with grapevines in various countries Training: Molecular characterisation of yeasts and bacteria in co-fermentation ER26, from Participant 7 to 3, involved in Task 1.2 and Task 4.1 (M2 and M4, milestones) for deliverables D1.2 and D4.1 Objective: Isolation of Yeasts from Spanish vineyards (grape) vintage 2015 Training: Cultural isolation techniques, set up of a Spanish wine-related yeast collection ER26, from Participant 7 to 3, involved in Task 1.2 and Task 4.1 (M2 and M4, milestones) for deliverables D1.2 and D4.1 Objective: Isolation of Yeasts from Spanish vineyards (grape) vintage 2016 Training: Update of the wine-related Spanish yeast collection Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 22/29

Task 4.2: Bio-protection of grape quality in vineyard and post-harvest withering The control of fungal diseases and mycotoxins contamination during grape maturation and post-harvesting is currently based on treatments with chemical fungicides. However the environmental dispersion, the progressive loss of effectiveness, the emergence of resistant strains and the increasing level of residues in wine, overall straw wines, have led the European Union to the restrict the use of these compounds, addressing the researchers towards innovative and eco-friendly protocols to face the problem. A sustainable approach consists of exploiting the natural antagonistic potential of various yeasts against moulds that occurs by different mechanisms such as nutrient competition, killer toxins character, iron depletion, ethyl-esters production or inducing host-plant resistance. Inhibition capabilities on mycelial growth or conidia germination of Botrytis and Penicillium spp. have been reported by some strains of species living in vineyard and cellar ecosystem, like Criptococcus laurentii, Metschnikowia fructicola, Pichia guilliermondii, P. membranifaciens, P. ohmeri, Rhodotorula glutinis. In this background, a deepen investigation on antagonism patterns can constitute a promising source of new knowledge to set biological control strategies in order to prevent or reduce harvested commodity damages. Objective: To investigate the natural antagonistic potential of various yeasts against moulds to develop a sustainable bio-protection approach According to the UE Directive 128/2009 dealing with the plant pathogens that can be controlled by pesticides, this goal will focus on the theme of the environmental defence towards the treatments with chemical fungicides. The approach will exploit the natural antagonistic potential of various yeasts against moulds that occurs by different mechanisms such as nutrient competition, killer toxins character, iron depletion, ethyl-esters production or inducing host-plant resistance. In this background, a deepen investigation on antagonism patterns can constitute a promising source of new knowledge to set biological control strategies in order to prevent or reduce harvested commodity damages. Plan of the Activities and Timetable: A4.2/1. Isolation and Selection of yeast strains showing antagonistic effects towards moulds - Yeast collection already isolated by previous investigations by participants 1 and 7 will be screened for their ability to counteract the growth of vineyard moulds (i.e. Botrytis and Penicillium spp.) in laboratory conditions. - Yeast isolation will be organized by participants 1 and 7 to increase the probability to find out interesting strains with the above mentioned characters; in particular, Vitis silvestris grapevine species and vineyards treated with different agronomic practices (organic, biodynamic and conventional) will be exploited as a source of novel yeasts. A4.2/2. Challenge tests in oenological conditions - To assess the antagonistic potential effect of selected yeast strains against moulds, Participant 1 will carry out challenge tests by inoculating yeasts on grapes contaminated with mycelium of different mould species. The antagonistic potential effect will be compared to commercialised products used as chemical fungicides. A4.2/3. Vineyard and post-harvest withering experiments - Selected yeast strains, potentially useful for the bio-protection of grape quality, will be tested in real oenological conditions (vineyard and post-harvest withering) by Participants 7 and/or 1 and 6. Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 23/29

Activity Description Involved Participans Expected time Isolation and selection of yeast strains 1-24 A4.2/1. showing antagonistic effects towards 1, 7 moulds A4.2/2. Challenge tests in oenological conditions 1, 6, 7 24-36 A4.2/3. Involved Participans: 1-UMIL 6-SU 7-AUG Vineyard and post-harvest withering experiments 7, 1, 6 36-48 Secondments: ESR22, from Participant 6 to 1, involved in Task 4.2 (M7, milestone) for deliverable D4.2 Objective: To select yeasts with a natural antagonistic potential on moulds and/or spoilage yeasts Training: identification of yeast strains that show antagonistic effects towards growth of moulds in lab conditions ER27, from Participant 7 to 1, involved in Task 4.2 (M7, milestone) for deliverable D4.2 Objective: To select yeasts with a natural antagonistic potential on moulds and/or spoilage yeasts Training: To select yeasts with a natural antagonistic potential on moulds and/or spoilage yeasts; Challenge testing by inoculating the selected strains on grapes contaminated with mycelium of different mould species Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 24/29

Task 4.3 Wine optimization targeting on different grape cultivars through yeast selection (aroma profiling, antioxidant compounds) (Participant in Charge: 1) The varietal character of a wine is due to a wide range of traits which expression is modulated by several factors including grape production and transformation. While the environmental and cultural practices that affect the grape quality potential are well recognised and summarised by the French term of terroir, the consistency of the practical role of the yeast biodiversity on the exploitation of the grape varietal expression is still matter of research activities. In the framework of this task, we would like to approach this subject rely on a large range of different grapevine varieties and yeast strains already available in the institutions involved in the task. To face the market, the wine industry should consider the development of innovative wine styles, suitable for the modern consumer needs. Wines with peculiar, distinguishable and intense varietal aroma should be projected with specific grapes varieties and yeast strain, able to amplify the varietal quality. Actually, the varietal aroma profiling of a wine is due to the volatile compounds already present in the grapevine berry as well as to the precursors of molecules that will became volatile during the fermentation and the wine aging processes, following different biochemical and chemical mechanisms. Therefore, in broad sense, also some wine peculiar fermentative aroma compounds, and compound profiling, should be considered varietal aroma, for their direct linkage to specific grapes cultivar composition like for example the amino acidic profiling. Peculiar enzymatic activities of yeast strain, such as β-glycosidase, may therefore augment the varietal aroma of a wine by matching with the grapes aroma potential. Another compositional aspect of the wine that can be affected by the yeast strain involves polyphenols. This wide chemical family of antioxidant compounds affect a large range of wine sensorial attributes including colour, taste notes like bitterness and astringency, as well as aroma features. Sustainable oenology objectives should give priority to the use of biological tools in the production of high quality wines. Objective: To select specific yeasts strains for the different cultivar for the optimization of sensory characteristics or to show up nutraceutical features The present activity should allow the development of innovative wine styles, suitable for the modern consumer needs. Wines with peculiar, distinguishable and intense varietal aroma should be projected with specific grapes varieties and yeast strain, able to amplify the varietal quality. Moreover, the selection of specific strains able to transform precursors to compounds with nutraceutical effect (natural antioxidants, amino acid derivatives, etc.) could represent a benefit to the research. Plan of the Activities and Timetable: A4.3/1. Cultivar collection and Characterisation - Participants 1 and 7 will characterise from chemical point of view the poll of grape precursors for flavour-active compounds, responsible for the varietal character of a wine, in different cultivars. The grape-derived compounds, many of which are non-volatile and flavourless before fermentation, can then contribute to the sensory properties of wine. A4.3/2. Yeast selection for specific enzymatic activity - Participants 1, 3 and 6, using their own collections, will carry out a screening to select yeasts that show specific enzymatic activities (i.e. glycosidases, lyases, esterases, etc.) useful for the biotransformation of precursor flavour-active compounds in laboratory conditions. These compounds are either hydrolysed (glyco- and cysteine-conjugates), biotransformed (monoterpenes, phenolic acids) or react (anthocyanins) with yeast metabolites (carbonyls). A4.3/3. Crossing different cultivars and yeast strains - Participants 1, 3 and 6 will perform laboratory grape fermentations suitable to enhance the varietal character potential using selected yeasts and cultivars. Chemical analysis will be used to monitoring the aroma profiles of the product of fermentation. Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 25/29

A4.3/4. Micro-vinification experiments - The selection of yeasts with different metabolic aptitudes with a range of different grapes, selected for their diverse expected wine aroma will be useful to develop micro-vinification experiment at the experimental cellar of Participants 1, 7 and/or 6. - To evaluate the final products, experimental wines will be submitted to chemical and sensory analysis. To analyse the yeast effect, a comparison will be carry out using wines produced with the same cultivar and in the same conditions but inoculating commercial strains. Activity Description Involved Participant Expected time A4.3/1. Cultivar collection and Characterisation 1, 7 1-12 DONE A4.3/2. Yeast selection for specific enzymatic activity 1, 3, 6 12-24 A4.3/3. Crossing different cultivars and yeast strains 1, 3, 6 A4.3/4. Micro-vinification experiments 1, 6, 7 24-36 36-48 Involved Participants: 1-UMIL 3-CSIC 6-SU 7-AUG Secondments: ESR2, from Participant 1 to 7, involved in Task 4.3 (M4, milestone) for deliverable D4.3, Activity A4.3/1 Objectives: To collect Georgian cultivars for yeast fermentations, Micro-vinification experiments Training: Fermentation management ER3, from Participant 1 to 7, involved in Task 4.3 (M7, milestone) for deliverable D4.3, A4.3/4. Objective: To collect Georgian cultivars for yeast fermentations; Micro-vinification experiments Training: Fermentation management Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 26/29

Task 4.4. Interaction of yeast and bacterial species during fermentation (Participant in charge: 6) A large number of yeast and bacterial species are present in the vineyard and cellar. While most of the species disappear early during fermentation, many persist or even prosper during a significant part of the fermentation process, and have the ability to impact on final wine character. Currently, the ability of wine makers to exploit the genetic potential of such diverse microbial assemblies is very limited. Indeed, we lack even a basic understanding about interactions between many of these organisms, be it interactions between different yeast species or between yeast and bacteria. This task will focus on providing baseline data on the ecological, physiological and molecular dynamic of interactions in complex multi-species cultures. The data will open new avenues for a better exploitation of natural biodiversity, and provide wine makers with novel tools to direct wine styles in the desired direction, be it towards more varietal or regional typicity or to better respond to global market trends. The task will follow this steps: 1. Using model fermentations with known species assemblies (Saccharomyces and non-saccharomyces yeast, as well as Lactic Acid Bacteria) to be selected by Participant institutions according to ongoing projects) to follow fermentation kinetics, species interaction and metabolite production; To assess the impact of similar specie assemblies in real grape must. Objective: To select specific yeast strains and combination of yeast strains for different winestyles and the optimization of sensory characteristics or nutraceutical impacts The present objective will allow the development of innovative wine styles, suitable for the modern consumer needs. Wines with peculiar, distinguishable and intense varietal aroma should be projected with specific grapes varieties and yeast strain, able to amplify the varietal quality. Moreover, the selection of specific strains able to transform precursors to compounds with nutraceutical effect (natural antioxidants, amino acid derivatives, etc.) could represent a benefit to the research. Plan of the Activities and Timetable: A4.4/1. Selection of a representative wine yeast consortium of at least 8 different species, based on continuous metagenomic analysis of SA grape musts. - Strains will be selected for: (1) their frequent presence at significant levels in natural grape musts (2) their potential to contribute to fermemtation dynamics and wine aroma (3) their molecular characteristics with regard to population monitoring through ARISA. The aim of this task is to select strains with oenological potential to serve as a model consortium. A4.4/2 Assessing the fermentation behaviour of the consortium and of individual strains that are part of the consortium in standard synthetic must. - Comparisons of molecular data with standard cultivation based techniques to calibrate the molecular population monitoring. A4.4/3 Testing the influence of biotic and abiotic factors on the population dynamics - The influence of various factors on fermentation dynamics, including dissolved oxygen level, temperature, ph, presence or absence of S. cerevisiae and others will be investigated. A4.4/4 Analysis of the impact of changes to the population dynamics throughout fermentation on the volatile chemical composition and aromatic features of the product - Standard analytical methods (GC-FID, GC-MS, LC-MS) will be used to assess the impact of changes in the yeast consortium on aromatic outputs. A4.4/5 Assessment of the consortiums impact in real wine fermentations and its impact on different wine styles and wine varietals. - Real wine fermentations will be inoculated with the same consortium, and similar treatments will be applied to assess whether the data obtained in synthetic musts apply to real grape must. Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 27/29

Activity Description Involved Participants Expected time Selection of consortium of nonconventional yeasts 1-48 A4.4/1 1, 3, 4, 6 A4.4./2 Assessing the fermentation behaviour of the consortium 1, 6 6-18 A4.4/3 Testing the influence of biotic and abiotic factors on the population dynamics 3, 6 12-24 A4.4/4 A4.4/5 Analysis of the impact of changes to the population dynamics throughout fermentation on the volatile chemical composition and aromatic features of the product Assessment of the consortiums impact in real wine fermentations. 1, 3, 4, 6 1, 4, 6 12-48 24-48 Involved Participants: P1-UMIL P3-CSIC P4-JSI P6-SU Secondments: ER4, from Participant 1 to 6, involved in Task 4.4 (M7, milestone) for deliverable D4.4 Objective: To study the interactions between microorganisms at physiological level Training: GC/MSMS and LC/MSMS ER6, from 1 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: To asses growth performance of mixed starter cultures Training: Controlled wine fermentation, Metagenomics on must and wine samples ER8, from Participant 6 to 3, involved in Task 4.1 and Task 4.4 (M7, milestone) for deliverables D4.1 and D4.4 Objective: To assess the dynamic of microbial population in real grape must Training: Metagenomic approach, data elaboration ER10, from Participant 2 to 6, involved in Task 4.4 (M7, milestone) for deliverable D4.4 Objective: To study the interactions between microorganisms at genetic level Training: Metagenomics ER12, from 2 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: Quantification of yeast biodiversity Training: Metagenomics on must and wine samples ESR15, from Participant 3 to 6, involved in Task 4.4 (M4, milestone) for deliverable D4.4 Objective: Study of interactions between different microbial species and strains in winemaking Training: Techniques for the assessment of microbial interactions ER16, from Participant 3 to 6, involved in Task 4.1 and Task 4.4 (M4 and M7, milestones) for deliverables D4.1 and D4.4 Objective: To assess the dynamic of microbial population in real grape must Training: Metagenomic approach, data elaboration ESR17, from Participant 4 to 6, involved in Task 3.3 and 4.4 (M4 and M7, milestones) for deliverables D3.3 and D4.4 Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 28/29

Objective: Yeast Population dynamics in controlled fermentation with non-conventional yeasts Training: Metagenomic approach Person in charge of the project for the beneficiary/consortium Name: Ileana Vigentini Date: 03.03.2015 Signature Mid-term Report - Period 1 ADDITIONAL INFORMATION: Page 29/29