Utilizing malolactic fermentation as a tool to prevent Brettanomyces bruxellensis wine spoilage

Brettanomyces bruxellensis is considered one of the most problematic wine spoilage yeasts due to the difficulty of controlling it, the potential significant financial losses due to loss of wine quality, and the cost of prevention and remediation measures. Wine is particularly vulnerable to B. bruxellensis infection during and shortly after the malolactic fermentation (MLF) as SO2 cannot be added until this process is complete. It has been suggested that conducting a rapid MLF initiated by inoculation of Oenococcus oeni is a useful strategy to prevent B. bruxellensis spoilage as this minimizes the length of time the wine is not protected by SO2. This project investigates an additional benefit of conducting a rapid MLF, the prevention of B. bruxellensis growth due to inhibitory interactions with O. oeni. Pinot noir wine (no SO2 additions, no MLF) was produced and used to test the ability of a number of commercial O. oeni strains to inhibit B. bruxellensis growth at the end of MLF. Sterile filtered wine was inoculated with various O. oeni strains and growth and malic acid monitored. When MLF is complete, wines will be inoculated with a select strain of B. bruxellensis and growth and volatile phenol production monitored.

The sensitivity of a number of B. bruxellensis strains to O. oeni is also being determined. B. bruxellensis strains have been sourced representing B. bruxellensis isolates from a wide range of winemaking regions including Oregon. A model wine system was identified for use to improve the rate that B. bruxellensis strains can be tested for inhibition by O. oeni. Results from the model wine system will be used to select which strains will be used in wine experiments that take significantly longer to complete.

Development of a genome-scale metabolic model for Saccharomyces cerevisiae for use in understanding and modifying strain performance

Two key metabolic activities of yeast relevant to wine fermentations are nutrient utilization efficiency and wine aroma development. For nutrient utilization efficiency (NUE), variability in yeast cell metabolism results from modulation of cellular processes that include changes in membrane composition along with a range of other metabolic pathways that are not fully understood. This variability often affects the completeness of a fermentation (characterized as “dry”, ”sluggish” or ”stuck”). Moreover, variability in yeast species or strains used in wine production results in different concentrations of aroma compounds, which can lead to distinct sensory characteristics. Controlling factors affecting nutrient utilization efficiency and wine aroma profile and mouthfeel characteristics related to yeast requires a detailed understanding of cellular metabolism. To develop such understanding, studies often use large-scale data approaches (e.g. genomics and metabolomics), along with multivariate statistics, to identify key metabolic fluxes or metabolites whose presence favors a specific fermentation outcome.

Although these studies are useful in exploring variation between yeasts, they are often not comprehensive enough, especially considering that they are labor intensive and costly. An alternative method is to use genome-scale metabolic models combined with dynamic FBA (flux balance analysis) to predict the flux distribution of all the metabolites within the cell over the course of an entire fermentation. As a part of this grant, our goal is to show that this computational approach can be used to predict experimental wine fermentation data, to understand differences between commercial strains, and to suggest genetic modification strategies towards increasing strain performance and control aroma characteristics. To date, we have been able to simulate anaerobic, nitrogen-limited yeast fermentations with the latest genome-scale yeast model. Behavior predicted for changing initial nitrogen concentration matches qualitatively with experiment. We simulated fermentation of three commercial yeast strains with highly varied NUE. Utilizing multivariate statistics, we have used the simulation results to identify the metabolic pathways that differ the most between these strains. On first analysis, the results are in agreement with existing experimental data. It is also clear that having an accurate biomass composition will be critical to a good quantitative fit of the data. Therefore, we are currently pursuing measurement of these key parameters as a function of fermentation time and strain.

Development of a prediction tool for phenolic extraction in red wines as a function of winemaking practices and fermentor design

Red wine fermentations are performed in the presence of grape skins and seeds to ensure
extraction of color and other phenolics. The presence of these solids results in two distinct phases
in the fermentor, as the solids float to the top to form a “cap.” Modeling of red wine
fermentations is, therefore, complex and must consider spatial heterogeneity to predict
fermentation kinetics. We have developed a reactor-engineering model for red wine
fermentations that includes the fundamentals of fermentation kinetics, heat transfer, diffusion,
and compressible fluid flow. To develop the heat transfer component of the model, the heat
transfer properties of grapes were experimentally determined as a function of fermentation
progression. COMSOL was used to solve all components of the model simultaneously utilizing a
Finite Elements Analysis (FEA) approach. Predictions from this model were validated using
prior experimental work. Model prediction and experimental data showed excellent agreement.
The model was then used to predict spatial profiles of active yeast cell concentration and ethanol
productivity, as well as liquid velocity profiles. The model was also used to predict how these
gradients would change with differences in initial nitrogen concentration, a key parameter in
predicting fermentation outcome in nitrogen-limited wine fermentations. After validation, this
model was applied to examine how fermentor design (e.g. scale and aspect ratio) would affect
fermentation mixing, temperature control, and chemical gradients. Along these lines,
temperature control and mixing were also evaluated for concrete eggs using the same model.
Finally, a preliminary model for phenolic extraction from skins and seeds was developed and
validated using experimental data. This led to an analysis of phenolic release of tannins from
grape seeds that we are currently pursuing. We are now in the process of the next step in
modeling—combining the two models for fermentation dynamics and phenolic extraction to be
able to predict and control phenolic profiles in finished red wines. In the seven months since this
grant began, we have been highly productive having published two papers, submitted a third that
is under review, and will be submitting a fourth paper within the next month. We have also
presented this work at various extension venues around the state.

Impact of Pre-Fermentation Cold Soak Conditions on Microbial Populations and Consequences for Wine Aroma

Wine aroma is one of the most important components of wine quality and can be impacted by grape variety, viticultural practices, and winemaking procedures. One particular practice that is employed during Pinot noir production to impact wine aroma is cold soaking. In this process grapes are held at cold temperatures to prevent growth of Saccharomyces cerevisiae and delay the beginning of alcoholic fermentation. Recent research has demonstrated that yeast naturally present during the cold soak can impact wine aroma and flavor (Hall et al. 2017). This research builds off these results and investigates how cold soak conditions could be manipulated to encourage or discourage growth of certain yeast and the consequences for wine aroma. Specifically, ways a winemaker may manage a cold soak (temperature, SO2, yeast diversity) were investigated for their impact on yeast populations and production of volatile aromas. Pinot noir wines were produced where the grapes were cold soaked for six days at two different temperatures (6 or 10?C) with the addition of 0, 50, or 100 mg/L SO2. Six non-Saccharomyces yeast species commonly isolated from grapes were inoculated and their populations monitored throughout the cold soak. Wine was also produced from grapes that did not undergo cold soak. Temperature and SO2 concentration impacted the growth of non-Saccharomyces yeast during the six day cold soak in a species specific manner. The highest populations observed were in the cold soak at 10?C when no SO2 addition had been made. Here H. uvarum increasing in population from approx. 103 cfu/mL to almost 108 cfu/mL by the end of the cold soak. As increasing concentrations of SO2 were added to the grapes prior to cold soak the growth of the non-Saccharomyces yeast including H. uvarum decreased. When 50 mg/L SO2 was added only low populations of H. uvarum, T. delbrueckii, and L. thermotolerans were detected at the end of the 6?C cold soak while at 10?C only H. uvarum was detected. When 100 mg/L SO2 was added there were few culturable yeast present in the cold soaks at either 6 or 10?C. Overall, increasing SO2 was more effective at minimizing H. uvarum growth than decreasing the temperature as there was still significant growth of H. uvarum at 6?C when no SO2 addition was made.

All wines made from grapes that underwent cold soak had significantly higher color and polymeric pigment content than wine made from grapes that did not undergo cold soak with only small differences in color and polymeric pigment content being noted between wines made from grapes cold soaked under different SO2 and temperature conditions. Initial volatile aroma analysis demonstrated significant differences between the concentration of a number of esters in wines made from grapes that were or were not cold soaked. In particular, wines made from grapes cold soaked with no SO2 additions had lower concentrations of certain ethyl esters.

Identifying Compound(s) Responsible for Off-flavors associated with “Stressed Vine Syndrome” in Pinot Noir

One of the increased concerns of wine industry is related to vine stress. Although the off-flavor descriptors vary from winery to winery, the frequent descriptors used in the wineries include “tequila”, “shellfish”, “peanut”, “ashtray”, “dry weed”, “herbaceous’, “flint” and other descriptors. In young wine, the taint smells like “bay leave”, and the wines do not age well. There were observations from wineries that taint could be related to compromised or nutritionally imbalanced fruits from stressed vines, induced by drought, nitrogen deficit, or a combination of many factors, but the exact cause(s) has never been studied or documented. This research is aimed at identifying the chemical nature of these off-flavors using gas chromatography/olfactometry, GC-MS and sensory analysis. Once the chemical nature of the off-flavor is identified, viticulture and enology remedies could be further investigated. The objective for the first year is to identify wine sensory characteristics that define “stress vine syndrome” through sensory evaluation and to identify wines for further analysis.

Several wines have been identified from industrial collaborators. Rollin Soles from ROCO winery identified a matured vineyard with full cover crop and dry farming practice to give a wine with bay leaf aromatic (2012 vintage) and the taint intensified with aging. Another vineyard with young vine also repeated to give dried herb tequila aromatics. Gary Horner from Erath winery also identified several wines with stressed vine aromatics including Pinot noir, Pinot gris and Chardonnay. More off-flavor wines are being recruited.

The “stressed vine” off-flavor was characterized by a sensory panel consisted of six winemakers from the Willamette Valley who have been involved with “stressed vine” off flavors previously and five OSU researchers involved in the project. The panel was able to identify the “stressed vine” off-flavor from the wines collected from industry. The off-flavor was also compared with wine standards comprised of tequila, peanut shell, bay leaf, agave and herbal characteristics. The off-flavored wine and the standards will be used for further sensory training and analysis.

Some preliminary chemical analyses were also performed in these wines including higher alcohols, esters, volatile phenols and TDN (kerosene aroma). The analyses were also performed on 20 normal Pinot noir wines. Data are being analysis to compare the off-flavored wine with the normal wines. GC/olfactometry will be performed next to identify the off-flavor compound(s) in “vine stressed” wines.

Investigating Fruitiness Perception in Red and White wines

This report details activities that occurred from August 2017 to January 2017. Several accomplishments were achieved during these first 6 months. We successful made wine, Pinot noir and Pinot gris, that contained no aroma compounds. This was achieved by altering some winemaking productions and an addition of resin that specifically absorbs aroma compounds. There is enough wine to serve as our base wine for the entirety of the study.  Fruit standards for red wines were developed and used for three sensory panels that investigated fruitiness in red wine. After 3 training sessions panelists were found to be consistent with their standard evaluations. The first sensory panel validated previous research focused on the impact of β-­damascenone, lactones, furaneols and red and black-berry associated esters to fruitiness in red wine. Our results for furaneols and β-dsamscenone do agree with previous work. The 2nd and 3rd sensory panels investigated the same compounds as the 1st sensory panel but at different concentrations and combinations. We have found that when norisoprenoids, β-damascenone and β-ionone, are at low concentrations and when furaneol compounds are at high concentrations, there is an impact to perception of red fruit aromas. However this only occurs if all other compounds are at lower concentrations. Once these compounds are in combination with higher concentrations of other compounds, the tested esters and lactones, there is a shift from red fruit aromas to dark fruit aromas. As we anticipated dark fruit aromas appear to be due to a combination of many compounds, with no one compound class dominating over another. We still have 1 more sensory panel to conduct to complete the investigation into the impact of norisoprenoids, furaneols, lactones and red and black berry associated esters on fruity aroma perception. It is our intent to begin focusing on the effect of acetate esters and volatile fatty acids to fruity perception of red wine. We also are prepared to begin investigating fruitiness in white wine, focusing on terpenes and esters.

Assessments of Difficult to Ferment Juices

The major goal of this project is to uncover the causes of chronically difficult to ferment juices. These juices are defined as those not due to fermentation management failures and inattention to nutritional needs of the yeast and maintenance of permissive growth and fermentation conditions. These juices are often derived from the same vineyard or block of a vineyard, and other similarly managed vineyards and blocks display normal fermentation kinetics. One class of these difficult to ferment juices is characterized by a high proline to arginine ratio. We have confirmed in several yeast strains that mannitol accumulates within the yeast in these juices and that this is associated with the presence of oxidative stress. To date all of the commercial strains tested are sensitive to these juices and reduce fermentation capacity. We have also confirmed the inhibitory role of previously identified lactic acid bacteria in yeast fermentation but have also discovered that these bacteria are efficient at inducing the establishment of the [GAR+] prion in wine strains. This prion is a protein conformational change that is inherited by progeny cells during cell division, thus once cells in the population have changed to establish the prion, subsequent generations will also be in the prion state without the need for continued induction. We have identified several other genera of lactic acid bacteria as well as acetic acid bacteria from arrested wines that are also capable of inducing the [GAR+] prion in wine yeast. We have received samples from over 53 wineries that have suspected bacterial inhibition of fermentation and have been able to isolate [GAR+] pion-inducing bacteria from many of these wines. Further we have shown that bacteria isolated from stuck wines, when grown in growth-permissive media then removed from the medium via filtration the inducer is still present in the filtrate and capable of  inducing the prion in wine strains of S. cerevisiae. Thus the inducer may be present in the absence of viable bacteria if the fermentation had bacteria present at some point. Controlling the presence of these bacteria is therefore important under production conditions.

Assessments of Difficult to Ferment Juices

The goal of this project is to uncover the causes of chronically difficult to ferment juices. These juices are often derived from the same vineyard or block of a vineyard and othersimilarly managed vineyards and blocks displaying normal fermentation kinetics. We haveconfirmed the inhibitory role of previously identified lactic acid bacteria in yeast fermentation buthave also discovered that these bacteria are efficient at inducing the establishment of the GAR+prion in wine strains. This prion is a protein conformational change that is inherited by progenycells during cell division, thus once cells in the population have changed to establish the prion subsequent generations will also be in the prion state without need for continued induction. We have also discovered that commercial wine strains that rapidly induce the prion as this induction occurred at a rapid rate in five of the 11 genetically unrelated commercial wine strains evaluated. The prevalence of this ability to rapidly induce this state suggests the prion state plays an important role in survival during wine fermentation. We are continuing to work out the metabolic changes that occur under these conditions to help identify or genetically construct via breeding strains that would be insensitive to the bacteria but also able to grow and ferment normally. In addition to inhibitory lactic acid bacteria, we discovered three species of acetic acid bacteria that can be found on grape berries at harvest that lead to arrest of fermentation. One of these bacteria, Gluconobacter cerinus, is as efficient as the lactic acid bacteria in inducing formation of the GAR+ prion. The other two acetic acid bacteria, Acetobacter malorum and Acetobacter ghanensis are inhibitory to yeast growth, showing similar levels of inhibition as Acetobacter aceti but without making the high concentrations of acetic acid found with A. aceti infection of wine. The inhibitor in this case is as yet unknown. Under certain metabolite conditions of low nitrogen or low vitamins we have shown that the high proline can be inhibitory to yeast metabolism.

The metabolomics analyses performed this past year confirmed the presence of mannitol in sluggish fermentations from this fruit and confirmed the induction of mannitol accumulation in juices by treatment with oxidizing agents. Our working model is that the phenolic profile of these juices may have changed due to environmental conditions and the accumulation of proline in the berry is done to minimize an inhibitory aspect of this compound or compounds and that yeast cells are similarly sensitive to these inhibitors and accumulate mannitol to minimize inhibition. We further propose that high proline might interfere with the functionality of the mannitol or lead to changes in the cell membrane that decrease ability to take up key nutrients like vitamins thus requiring higher vitamin supplementation.

Assessments of Difficult to Ferment Juices

The goal of this project is to uncover the causes of chronically difficult to ferment juices. These juices often derive from the same vineyard or block of a vineyard and other similarly managed vineyards and blocks display normal fermentation kinetics. These difficult to ferment juices do not appear to respond to nitrogen or other commercial nutrient addition, occur regardless of strain used, and are challenging to restart. Our aims for the first year of this project were to evaluate the nutritional content of these juices to determine if a nutritional deficiency or presence of toxic compound was the cause of inhibition of fermentation. In this first year we were able to narrow down the possible issues with these juices. Although low nitrogen is a factor, addition of nitrogen to the fermentation seems to partially address the nutritional limitation. In addition these juices appear to impose a high vitamin demand on the yeast which we will explore in detail in this second year. This past year we were able to demonstrate that the accumulation of mannitol observed from the metabolomics analysis of the yeast strains grown in difficult to ferment juices is an indicator of oxidative stress in the juice.

It is not clear what juice compositional factors are leading to the oxidative stress and why this is not alleviated by use of Sulfur dioxide, and this will be explored further in this coming year. The low arginine and high proline of these juices suggests a problem with development, activity or destruction of the fine roots of the vine. Addressing this problem by adjustment of either yeast strain or nutrient supplementation would prevent the need for more invasive vineyard management strategies.

In this past year several commercial wineries sent us examples of chronic to ferment juices and we discovered that these juices either resembled the J. Lohr juice in having a deficient nutrient profile or were characterized by the presence of four rare (for grape) acetic acid bacteria species that seem to retain viability throughout the winemaking process and that appear to be inhibitory in ways not due to simple acetic acid production. An added goal for this coming year’s grant is to evaluate the role(s) of these bacteria in fermentation progression and to assess their sulfite sensitivities and persistence during fermentation.

Understanding Alcohol Tolerance in Wine Yeasts

Understanding alcohol tolerance in wine yeasts is important in order to develop tools to rectify sluggish and stuck wine fermentations and new commercial yeast strains with desirable sensory or process characteristics. In previous work, we have identified two potential mechanisms for alcohol tolerance including ones related to the composition of the yeast cell membrane and to the nutrient utilization efficiency of the strain. To understand these mechanisms, we further examined both potential effects. Specifically, we were able to find that the composition of the cell membrane changes significantly as fermentation temperature is raised or lowered. Using our high resolution methods for measuring membrane composition, we were also able to identify specific types of lipids in the cell membrane that are associated with lower ethanol tolerance (e.g. several phosphatidylinositol lipids) and higher ethanol tolerance (e.g. several phosphatidylcholine lipids). In this period, we also developed and used a method for measuring a wide range of metabolites inside and outside of yeast cells during fermentation. Our analysis shows that certain metabolic pathways like the pentose phosphate pathway and glycerol secretion are negatively correlated with cell growth and ethanol tolerance of these strains. These major results have allowed us to begin to identify targets for genetic manipulation of wine yeasts that should increase or decrease alcohol tolerance. We hope to actively initiate this part of the project prior to the completion of funding this year. The work completed as part of this grant has resulted in two papers (one published and the other in final revisions after review) with two more in preparation and a number of presentations at local and national meetings.