Do I need to inoculate for malolactic fermentation?

After the completion of alcoholic fermentation, most red wines undergo malolactic fermentation, a process that stabilizes the wine against microbial attack during aging. This process also shifts the perception of acidity by lowering the TA (and increasing pH) of the wine¹. Malolactic fermentation itself includes the removal of carbon dioxide from malic acid, leaving lactic acid behind. Lactic acid bacteria (LAB) harness the energy released from breaking this bond and apply it to energy consuming processes in the cell. Many different genera of bacteria contain the enzymes needed to decarboxylate malic acid². In addition to malolactic fermentation, these microbes usually have many other impacts on the sensory properties of the wine, both positive and negative³.

Oenococcus oeni is the main malic acid consuming bacteria in winery settings. This bacteria is able to survive in the high alcohol, low pH, low nutrient conditions of wine, and still complete malolactic fermentation with a minimal level of spoilage characteristics¹. Though O. oeni is very prevalent in wineries, it is not the most common LAB found on grapes. Many different species of LAB, including Lactobacillus and Pediococcus, are found on grapes in higher population density than Oenococcus oeni. Oenococcus oeni is often so scarce in grape samples that enrichment techniques are needed to detect them⁴. As fermentation proceeds, alcohol accumulates and oxygen becomes scarce, leading to a decline in population level of other LAB whie O. oeni begins to dominate. By the end of alcoholic fermentation, the population level of Lactobacillus and Pediococcus is low while O. oeni is poised to drive malolactic fermentation³. 

If grapes are healthy (with limited LAB) and the winery has been through several vintages, there may be enough O. oeni resident to complete malolactic fermentation without inoculation. However, if grapes are compromised when they enter the winery, spoilage LAB will be more concentrated. Wine pH higher than 3.5 is more permissive/less inhibitory to LAB^3,4. If LAB other than O. oeni predominate, the wine is vulnerable to spoilage characteristics such as volatile acidity (acetic acid and ethyl acetate), mouse taint, biogenic amines, and ropy texture⁵. Inoculation help O. oeni to predominate and limits the impact of other LAB.

 

Does it matter which ML bacteria I use?

The genome of Oenococcus oeni contains around 1700 genes, roughly half of which are known to be variable within the species. This means that different strains may have different levels of tolerance, metabolic rate, and production levels of secondary metabolites³. One of the most familiar sensory outcomes of malolactic fermentation is the production of diacetyl, the compound that smells like movie theater popcorn (“butter”). Diacetyl production differs by strain (as well as environmental factors), as does production of fruity esters (red berry character), liberation of oak lactones (vanilla) from barrels, and degradation of vegetal aromas³. One only needs to look at recent versions of product catalogues to see full aroma profiles from different strains of bacteria.

The strain of LAB that completes malolactic fermentation may also alter the impact of future Brettanomyces infections. Some strains of Oenococcus oeni (and other LAB) have enzymes that release tartaric acid bound to hydroxycinnamic acids (TAE-HCA’s), releasing p-coumaric and ferulic acids. Brettanomyces bruxellensis use p-coumaric and ferulic acids as substrate to produce 4EP and 4EG, the compounds responsible for barnyard, spice and clove impacts in wine. Brettanomyces bruxellensis is unable to degrade TAE-HCAs on its own, so the production of this substrate by malic acid bacteria can greatly increase the concentration of 4EP and 3EG in Brett infected wines. Not all strains of Oenococcus oeni contain these enzymes⁶. Commercial strains have been tested and removed from the market if they have the enzymes⁶, but ambient strains do not have this guarantee.

 

Why is my malolactic fermentation so slow this year?

(...and how might co-inoculation help?)

Though Oenococcus oeni is (thankfully) better adapted to the harsh conditions of wine than other LAB, it is still more likely to be inhibited by environmental conditions than Saccharomyces cerevisiae. Stress is cumulative; though environmental conditions may be within the tolerance limits on their own, if two or three conditions are moderately stressful (close to the edge of the range), then bacteria may still be inhibited or even killed. Some manufacturers now offer a scorecard to add up potential stresses to determine if specialized bacteria may be needed.

Potential stressors include:

• Ethanol – Bacterial cell membranes are less complex than Saccharomyces membranes. As ethanol concentration rises, membranes become less fluid, limiting entry of nutrients into the cell. They also become more permeable, leading to destabilization¹.
• Temperature - If the temperature is too low, metabolic processes occur very slowly. If temperature is too high, membranes become too permeable, adding to stress already exerted by high ethanol. Spoilage microbes are also more active at higher temperatures, leading to competitive inhibition for O. oeni¹.
• pH – If pH is too high (>3.6), spoilage microbes are less inhibited, leading to competitive inhibition. If the pH is too low, even small amounts of free SO2 leads to inhibitory concentration of molecular SO₂. Oenococcus can be inhibited by as little as 0.1 – 0.15 mg/L of molecular SO₂, while levels from 0.3 – 0.5 mg/L can be toxic⁷. At a pH of 3.6, free SO2 of 6-9 mg/L of meets the inhibition threshold while free SO2 of 19-31 mg/L would be toxic.
• Total SO₂ – When SO₂ is added to grapes or juice during processing, most of it is quickly bound up by sugars. As fermentation proceeds, yeast produce acetaldehyde, which binds SO₂ even more strongly. Between sugars and acetaldehyde binding, free SO₂ is nearly zero.  Oenococcus oeni metabolize acetaldehyde, releasing bound SO₂ to the free fraction. If the pH is low enough, the molecular SO₂ may each inhibitory concentration. In this way, all of the SO₂ has the potential to form toxic products for malic acid bacteria⁷.
• Yeast strain – The total SO₂ of a fermenting wine includes any SO₂ added at crush plus SO₂ produced by the yeast. Most Saccharomyces strains produce some SO₂ during fermentation, with the amount dependent on conditions. One survey of SO₂ production by four different strains of commercial Saccharomyces found a range of 13-42 ppm SO₂ produced under the same conditions⁸. In the product catalogue, high SO₂ producing yeast are often noted to be less permissive of malolactic fermentation.
• Nutrients – Like other microbes, Oenococcus oeni needs a source of nitrogen to build proteins. Unlike Saccharomyces, they do not generate amino acids on their own, so adding DAP does not help. They also cannot take up amino acids from the environment, but rather they take up small proteins to supply this need. For these reasons, YAN is not a good indicator if the juice/wine contains ample nutrients for malolactic bacteria to grow and metabolize. A good rule of thumb is that if the alcoholic fermentation was stressful for the yeast, then nutrients might be limiting. As yeast cells break down in the lees, they release needed peptides⁷. Options for supplementation specific to malolactic bacteria are also available from vendors.

Table 1: Favorable and unfavorable conditions for malolactic fermentation (adapted from AWRI: Achieving Successful Malolactic Fermentation¹)

 

What are the pros and cons of co-inoculation?

Co-inoculation allows malic acid bacteria to avoid many of the most inhibitory of the conditions listed above. Despite the name, co-inoculation of malic acid bacteria usually occurs 12-48 hours after yeast inoculation^1,9. If added at the same time as yeast inoculation, malic bacteria loose 99% of their viable cells due to free sulfites⁸. As fermentation begins, yeast produce compounds to detoxify SO₂, allowing a much more permissive environment for malic acid bacteria. When added early, bacteria experience a low ethanol, high nutrient environment with the lowest pH the wine will likely have. Bacterial action is usually suppressed during active fermentation, but cells still acclimatize their cell membranes to rising alcohol levels during this time. As the yeast start to die off at the end of fermentation, they release nutrients back into the wine just as malic bacteria transition from lag phase to log phase⁹. When malolactic fermentation occurs under the reductive conditions of fermentation, diacetyl production is low and lower pH leads to higher production of fruity esters.

Unfortunately, most malic acid bacteria, including Oenococcus oeni are heterofermentative, which means that they possess the metabolic machinery to utilize glucose and fructose to gain energy with acetic acid as a byproduct². Malic acid and citric acid are consumed preferentially over sugar, however, if these substrates have been depleted, malic acid bacteria will consume any sugar that remains. In a normal fermentation, sugar is consumed before malic acid is fully depleted. However, in a stuck or sluggish fermentation, or in whole cluster fermentation that releases sugar at pressing, residual sugar will result in production of high amounts of acetic acid. Careful fermentation monitoring is key to success with coinoculation³. Malic acid bacteria are also more sensitive to high temperature, so in highly extractive, hot fermentations, sequential inoculation will be more successful.

---

References

(1)            Achieving Successful Malolactic Fermentation, 2020. https://www.awri.com.au/wp-content/uploads/2011/06/Malolactic-fermentation.pdf.

(2)            Liu, S.-Q. Malolactic Fermentation in Wine - beyond Deacidification. J Appl Microbiol 2002, 92 (4), 589–601.

(3)            Krieger-Webe, S. Good Management of Malolactic Fermentation - Sensory Impacts, 2024.

(4)            Liu, Y.; Rousseaux, S.; Tourdot-Maréchal, R.; Sadoudi, M.; Gougeon, R.; Schmitt-Kopplin, P.; Alexandre, H. Wine Microbiome: A Dynamic World of Microbial Interactions. Critical Reviews in Food Science and Nutrition 2017, 57 (4), 856–873.

(5)            ETS Labs. Scorpions Bacteria Panel. https://www.etslabs.com/analyses/analysis/161987.

(6)            Chescheir, S.; Philbin, D.; Osborne, J. P. Impact of Oenococcus Oeni on Wine Hydroxycinnamic Acids and Volatile Phenol Production by Brettanomyces Bruxellensis. American Journal of Enology and Viticulture 2015, 66 (3), 357–362.

(7)            Malolactic Fermentation: Importance of Wine Lactic Acid Bacteria in Winemaking; Morenzoni, R., Ed.; Lallemand Inc.: Montreal, 2015.

(8)            Henick‐Kling, T.; Park, Y. H. Considerations for the Use of Yeast and Bacterial Starter Cultures: SO2 and Timing of Inoculation. American Journal of Enology and Viticulture 1994, 45 (3), 464–469.

(9)            du Toit, M. The Wine Expert: Co-Inoculation of Selected Wine Bacteria, n.d.

Abstract

The purpose of this experiment was to compare cellar efficiency, chemical and sensory outcomes of sequential vs. co-inoculation of malolactic fermentation in Ingleside red wines. In 2023, wines completed malolactic fermentation in a timely manner (by December) regardless of inoculation time, with no notable differences in wine chemistry or sensory characteristics between treatments. Though there were no differences in this year, co-inoculation may speed up the process in other years with slower rate of malolactic fermentation. As long as fermentations are monitored carefully to avoid sluggish or stuck conditions (and consumption of residual sugar by Oenococcus), co-inoculation may be a good option to improve cellar efficiency at Ingleside.

 

Experimental Questions

Like many Virginia wineries, the standard protocol for malolactic fermentation at Ingleside has historically been sequential inoculation with a commercial strain of Oenococcus oeni. After inoculation, barrels were set in a single layer with frequent stirring to encourage the completion of malolactic fermentation. In 2022, malolactic fermentation was slow to finish, leading to space and labor inefficiency. One solution to this inefficiency may be the co-inoculation of malolactic fermentation 24 – 48 hours post inoculation with Saccharomyces. Coupled with extended maceration (also standard protocol at Ingleside), co-inoculation may allow wines to complete malolactic fermentation before they are transferred to barrel, allowing for addition of SO₂ to topped barrels with solid bungs before the cellar gets cold for the winter. The purpose of this experiment was to compare cellar efficiency, chemical and sensory outcomes of sequential vs. co-inoculation of malolactic fermentation in Ingleside red wines.

In this experiment, two treatments were employed:

• Sequential inoculation of red wines
• Co-inoculation of red wines 24-48 hours after yeast inoculation

 

Methods

Petit Verdot fruit was hand harvested, chilled overnight, then destemmed to TBins with the addition of FT Rouge Berry (30 g/hL) and oak chips. Saignée of 5-8% of the volume was done post de-stemming, with an equal proportion removed for each treatment within a lot. Fermentation was inoculated with 25 g/hL commercial yeast (MT yeast (PV1) or ICV D254 (PV2). Sugar (22 g/L) and acid additions (2 g/L) were also done on day 2 of fermentation. Stimula Cab Sauv (40 g/hL) was added on day 3. Fermentations were allowed to macerate for 3 weeks prior to pressing. Wine settled in tank for one day, then was transferred to barrels. At the completion of malolactic fermentation, 80 ppm SO2 (14 g/hL KMBS) was added to the wine. Wine was racked in December. Oenococcus oeni (Silka, Lallemand) was added to the co-inoculated treatments 2 days post inoculation and to the sequential inoculated treatments after the completion of fermentation.

Similar treatments were done in Merlot and Sangiovese. Results for all four lots are reported below.

 

Results

Table 1: Juice chemistry for three lots used for experimentation (Imbibe Solutions)

 

Fruit came in with relatively low Brix. Malic acid values ranged from relatively low (Sangiovese) to relatively high (Petit Verdot) (Table 1). Fermentation kinetics in Petit Verdot were very similar regardless of yeast strain or treatment group (Figure 1). Fermentation temperature peaked between 84-86°F. All wines finished fermentation with RS < 1.00 (ICV Labs, November 2024), regardless of variety, yeast, or treatment condition.

 

Figure 1: Fermentation kinetics of sequential and co-inoculated Petit Verdot using two commercial yeast strains (PV1 = MT yeast, PV2 = D254)

Figure 2: Malic acid concentration at pressing (November 17, 2023; Imbibe solutions)

At the end of extended maceration, each of the wines contained less than 0.2 g/L of malic acid, regardless of inoculation timing (Figure 2). None of the co-inoculated wines had finished malolactic fermentation at this time while one of the sequential inoculation wines had (Sangiovese).

 

Table 2: Wine chemistry after the completion of malolactic fermentation for four lots of experimental wines (ICV Labs, December 2024)

Each lot had slightly higher lactic acid and pH in the co-inoculated wine vs. the sequentially inoculated wine. There were no consistent differences in alcohol conversion rates or levels of acetic acid production.

 

Figure 3: Free and bound SO₂ for four treatment lots. The ratio of free:bound SO₂ is labeled on the endcaps.

 

Acetaldehyde is a byproduct of fermentation that is eventually consumed by Saccharomyces at the end of alcoholic fermentation and Oenococcus at the end of malolactic fermentation. Any acetaldehyde not consumed, or produced later by the oxidative conversion of ethanol, will be bound by SO₂. Acetaldehyde is the primary binder of SO₂ in the wine, so the amount of bound SO₂ gives a snapshot of overall acetaldehyde in the wine. Early completion of malolactic fermentation may lead to less acetaldehyde metabolism and, theoretically, shift the fraction of free:bound SO₂. That does not appear to be the case, here. There is not consistent pattern of SO₂ binding relative to the timing of malolactic fermentation (Figure 3).

 

Figure 4: Color Intensity for four treatment lots (ICV Labs Dec 2023). Hue (A420:A520) is labeled on the endcaps.

Color is a complex measure that is influenced by several factors including the concentration of pigments, the concentration of cofactors, the oxidative state of pigments, the pH and free SO₂ of the wine. These wines did not have identical pH and free SO2 values, so caution should be taken when interpreting color values. Co-inoculated wines had slightly higher color intensity in three of the four lots. For each, free SO2 values were very similar and pH values were slightly higher in the co-inoculated wine, which would be expected to decrease color intensity (Figure 4).

 

When wines were tasted by the winemaker and WRE Staff, there were no large or consistent differences in sensory characteristics between sequential and co-inoculated wines. These wines were not tasted at a sensory session.

 

 

In 2023, Zach Pierce from Ingleside Vineyards explored how the timing of malolactic inoculation impacted the chemical and sensory characteristics of Sangiovese, Merlot, and Petit Verdot as well as cellar efficiancy. While this experiment did not result in any notable differences in wine chemistry or sensory characteristics, co-inoculation may speed up the process in years with slower rates of malolactic fermentation. In this video, Joy and Zach discuss the ins and outs of this project, with helpful tips for anyone considering malolactic co-inoculation. 

Click here to watch the video