Microbes in the wine bottle: Sources, manifestation, and suppression tactics

If a bottled wine is microbially unstable, it most likely will reveal itself to the consumer in a number of unpleasant ways. In a Practical Winery and Vineyard Journal article (Monitoring microbes during cellaring/bottling, practicalwinery.com, January/February 2010), Lisa Van de Water identified the danger signals for microbes in the bottle as follows:

• Change in color, aroma, or flavor
• CO₂ production
• Clouding.

Van de Water says that to assess the microbial stability of a wine before bottling, four questions need to be asked:

• Does the wine have residual fermentable sugar?
• Does the wine have residual malic acid?
• Does the wine have Pediococcus?
• Does the wine have Brett.

If there is residual fermentable sugar in the wine, the problem could either be a re-fermentation of the sugar by yeasts or metabolization of the sugars by lactic acid bacteria (LAB). The author recommends filtering 250 ml of wine for yeast and 100 ml for bacteria and culturing the membranes to assess bottle stability. PCR tests can help monitor bottled wines for microbes that are at high levels. As it relates to the question of residual malic acid, Ms. Van de Water says that LAB can cause a number of problems if they metabolize residual malic acid in the bottle. Wines with residual malic acid should be filtered with a 0.45µ membrane. 

The most common microbes to manifest in the bottle are presented in the table below.

Microbe	“Danger” Metric	Manifestation in Bottle
S. cerevisiae	5 - 10 cells/bottle; > 2 g/l RS;  cannot grow without sugar	Cork-pushing carbonation at 1400 mg/l CO₂; perceivable spritz at 800 mg/l CO₂ (would require 1.6 g/l refermentable fructose or glucose); visible haze above 1000 mg/l RS
Zygosaccharomyces	1 cell/bottle; cannot grow without sugar
Surface film yeasts	Cannot grow after 1 week in bottle
Dekkera/Brettanomyces	1 to 50/ml; 0.2 g/l RS	100 mg/l can lead to visible Brett yeast haze; Small amount of CO₂; horse sweat and Band Aid® odors; bitter, metallic finish
Oenococcus	500/ml if malic acid present; ok if no malic acid; 0.2 g/l RS	Cork-pushing carbonation at 1400 mg/l CO₂; perceivable spritz at 800 mg/l CO₂ (would require MLF of 2.4 g/l malic acid); a rise in pH; cloudiness, a fine sediment
Lactobacillus	If a substrate is present; 0.2 g/l RS	Cork-pushing carbonation at 1400 mg/l CO₂; perceivable spritz at 800 mg/l CO₂ (would require MLF of 2.4 g/l malic acid); a rise in pH; cloudiness; a fine sediment
Pediococcus	Any visible under a microscope; 0.2 g/l RS	Cork-pushing carbonation at 1400 mg/l CO₂; perceivable spritz at 800 mg/l CO₂ (would require MLF of 2.4 g/l malic acid); biogenic amines (sometimes); unpleasant aftertaste (sometimes); cloudiness; a fine sediment
Acetobacter	Present only if the closure is compromised

Sources: Lisa Van de Water, Monitoring microbes during cellaring/bottling, Practical Winery and Vineyard Journal, practicalwinery.com, January/February 2010); Christian Butzke, Preventing Refermentation, extension.purdue.edu.

As it relates to the question of what causes microbial instability to manifest in the bottle, the answer is “a cascade of events.” First, the environment has to be hosting microbes. The sources of the infection could be (i) spoilage yeasts transiting from the vineyard in sour-rot grapes; (ii) Brettanomyces bruxellensis,  which can make "the winery itself a primary habitat surviving in the walls ... interior surfaces of presses and fermentation tanks, or on the wood of barrels"; or (iii) residual sugar or malic acid, which could serve as the substrate for fermentative action by S. cerevisiae or LAB. The second event, which builds on the first, is an insufficiently sanitary regime to allow defeat/suppression of the microbes prior to bottling. And third,  a non-robust testing regime that allows the bottling and sale of wines that are microbially unstable. 

The following steps are recommended in order to minimize the opportunities for bottle-manifested microbial instability :

1. Test the wines in the cellar to determine exactly what types of microbes, if any, are resident in the wine. To detect Brettanomyces, Ms. Van de Water recommends culturing the wine on media containing 50 ppm of the antibiotic cycloheximide (to inhibit growth of other yeasts) and, if present, the culture will manifest “white, hemispherical colonies in three to seven days.” The Brettanomyces culture also produces a strong acetic acid smell. Levels of 100 CFUs/ml is cause for concern. Conduct enzymatic tests to determine whether fructose and glucose are above the level of 0.5 g/l. The ratio of glucose to fructose is important so the sugars should be tested separately (According to Ms. Van de Water, “Saccharomyces is reluctant to use fructose if the glucose/fructose ratio is less than 1:6 to 1:10). Use enzymatic tests to determine whether LAB is present in the wine. Malic acid levels should come in below 30 mg/l. If the wine is shown to be microbially stable upon completion of the tests, then go to Step 3. If not, go to Step 2.
2. Filter the wines with a .45 µ membrane
3. Take preventative steps to combat oxidation. According to Enology Note, “excessive oxygen pickup during filtration and bottling reduces the sulfur dioxide concentration notably.” Preventative steps to combat oxidation include: SO₂ additions during or just prior to wine movement, nitrogen blanketing; CO₂ sparging; and the flushing of lines and receiving tanks.

In addition to the steps above, the following actions should be taken to insure that the wine retains its stability during the bottling process (Lansing, Managing Bottling Operations, winebusiness.com, May 2011): 

• Purge the cellar tank and blanket with CO₂ before filling with transferred wine
• When releasing wine from cellar tanks to bottling line filters, verify that the tank wine meets target parameters for alcohol %, DO, CO₂, and other chemical metrics
• Verify that in-line wine filtration upstream of the line filter is in place and meets target parameters for sterile wine filtration
• Perform clean-in-place for wine lines and follow at least every 72 hours to ensure microbial stability
• Purge filter bowls with nitrogen to minimize DO intrusion
• Perform all necessary QC for the rinses, filter, corks, and other line components to maintain the highest quality.

©Wine -- Mise en abyme

Extended-bottle-aging versus easy-drinking wines: Winemaking contrasts

At the broadest level, red wine can be described as either: (i) easy-drinking or (ii) styled for extended bottle aging. While the definition of the second class is fully described, the first is not. Mary Gorman-McAdams MW says that easy-drinking wines are viewed as “having broad appeal, especially to novice and marginal drinkers .. are straightforward, fairly simple, inexpensive, and most likely fruity ...” (Wine Words: Easy Drinking, www.thekitchn.com). To that I would add that these wines are generally consumed within a short time of purchase. I will contrast the winemaking decisions made in a number of different areas during the construction of these two wine styles.

Picking Decision

The winemaker will pick at optimal ripeness but the definition will differ between these two styles. For the easy-drinking wine, where fruitiness is a key characteristic, ºBrix may be the only metric and once a target level is attained, picking may commence. Given the urge to produce these wines as inexpensively as possible, every day on the vine is an additional day of weather risk in a low-margin environment. 

For the extended-bottle-aging wine, the stuffing for the develpment of the wine in bottle has to be sourced from the vine. The decision on optimal ripeness will cover a number of objective (sugar, acid, pH, ratio between sugar and acid) and subjective (color, ease of removal of berries from pedicel, texture, aroma, flavor) criteria. Phenolic ripeness is critical in this environment as its components are key enablers of bottle aging and the associated customer appreciation.

Once the decision has been made to harvest, the question then becomes how to implement. The choices are manual (more selective and thorough, less damaging to the grapes -- Boulton et al., Principles and Practices of Winemaking) and machine (faster, more economical, less prone to error, allows work at night -- Boulton et al.) harvesting. In the case of extended-bottle-aging wines, the majority of these grapes are picked by hand to minimize damage as well as to provide a first-level of selection in the field. These grapes are placed in small containers (to minimize crushing pressures) and then transported to the winery as quickly as possible. In many cases, the grapes for the easy drinking wines are machine harvested. It is obviously much more expensive to manual-harvest with in-field selection than it is to machine-harvest but the mantra in that space is that high-quality grapes are a requirement for high-quality wines.

For premium (used interchangeably with extended-bottle-aging herein) wines, grapes are generally picked by lot, fermented in that manner and then blended either after fermentation or after aging. Easy drinking wines are generally picked en masse and any blends would be of the field or fermentation tank variety.

Harvest Reception

Manual Sorting is another level of quality selection that is employed in the construction of high-quality wines and brings with it an associated cost. Manual sorting is not generally employed in the construct of easy-drinking wines.

Fermentation Management

According to Jackson (Wine Science: Principles and Applications), maceration is one of the primary areas in which winemakers can differentiate their wines. In the case of wines for early consumption, the must is generally pressed after 5 days (Jackson) which allows good extraction from the skin (color and enough tannins to ensure its stability) while avoiding the harsh tannins resident in the seeds. The lack of harsh tannins allow these wines to be approachable earlier. Extended-bottle-aging wines are generally macerated for longer periods (up to 3 weeks +) and thus gain the benefits of the “high molecular weight tannins” which polymerize and precipitate out in the bottle. These tannins soften up over time, while aromas and flavors develop, thus providing the user with a benefit for the investment in time (and money).

The size of the fermentation vessel is also differentiated depending on the style of wine being pursued. According to Jackson, premium wines tend to be fermented in 50 - 100-hl vessels which seem to “provide an appropriate balance between economics and ease of operation and the desire to maintain wine individuality.” For the standard wines the practice is to employ fewer but larger (200 to 2000 hl) tanks (Jackson).

Jackson also mentions the use of rotary fermentors which, due to their horizontal deployment and spirally shaped paddles, keep the juice in constant contact with the pomade, thus allowing rapid extraction of anthocyanin and flavors. With the needed elements extracted, the must can be pressed before the heavier phenolics are extracted, making the resultant wine much more approachable.

Yeasts also provide another area of differentiation between the two wine styles. Early-drinking wines will always be inoculated and will use fast-fermenting yeasts with fruity characteristics. Efficiency is the driver here. Extended-bottle-aging wines will, in some cases, use natural yeasts for fermentation as they pursue additional complexity. In the cases where inoculation is practiced, the characteristics pursued would be varietal enhancement and increased flavor complexity.

Pressing

Henderson (http://www.santarosa.edu/~jhenderson/Red%20Crush.pdf) identifies four “times to press” depending on stylistic preferences:

• When the wine still has residual sugar -- good for young fruity style wines, light character, usually 2 - 5ºBrix
• As soon as dryness occurs -- good for medium- to full-bodied reds
• Several days after dryness -- additional flavor extraction
• Extended maceration -- 1 to 6 weeks after dryness. Cap sinks and harsh tannins settle out, making for a softer wine.

In the case of the easy drinking wine, the second of the four methods is the one most likely to be employed (based both on stylistic and economic considerations --additional flavor not a requirement and the fermenter needed for additional work). The extended-bottle-aging wine would employ either method 3 or 4 in its hunt for additional complexity.

Malolactic Fermentation (MLF)

In practice, most red wines undergo MLF. The process is encouraged (Bauer and Dicks, Control of Malolactic Fermentation in Wine, S. Afr. J. Enol. Vitic. 25(2), 2004): in cooler areas where grapes have high malic acid content; in cases where the wine is aged in oak barrels; and when the wine style calls for long-term aging in bottle. The practice is sometimes forsworn in warmer, lower-acid areas and in the cases where undesirable organoleptic changes or the production of biogenic amines result.The main effects of MLF on wine are (i) a reduction in titratable acidity (by 0.1 to 0.3%) and an increase in pH (0.15 to 0.30). In addition, dramatic organoleptic changes to the wine are evidenced (Lonvaud-Funel, Microbiology of the Malolactic Fermentation: Molecular Aspects, FEMS Microbiology Letters):

• The specific taste of malic acid disappears
• Sugars are catabolized to produce mainly lactic and acetic acid
• Citric acid is transformed into acetic acid and carbonyl compounds, notably the butter-flavored diacetyl
• Wine taste and color are modified due to the metabolic activity of bacteria on phenolic compounds (tannins, anthocyannins).

Easy-drinking red wines would not materially benefit from the organoleptic changes resulting from MLF. The goal in the vase of these wines are to carry the fruity flavors to the market at the lowest possible cost and in the shortest time possible. However, if appropriate steps are not taken to inhibit the lactic acid bacteria, in-bottle-MLF could result.

Aging

Easy-drinking wines do not benefit from maturation in the cellar and, as such, are not normally subjected to the maturing practices (oak barrels, multiple rackings, etc.) afforded extended-bottle-aging wines. Some of the oak flavor characteristics that accrue as a result of barrel aging can be imparted to easy-drinking wines by introducing oak chips, planks, or extracts into the tanks during alcoholic fermentation.

Premium wines are generally aged in oak (new French, for the most part, with the % new dependent on the winemaker’s style) for extended periods (1 to two years before bottling and for some period post-bottling) resulting in (Dharmadhikari): a replacement of the bright red colors of the young wine by its polymeric form; reduction in grapey aroma and increase in pleasing aromas and flavors; and reduction of astringent and harsh tastes, increasingly replaced by smoother, rounder tastes. According to Jackson, the expense and effort of aging in oak is justified by the additional flavor and complexity gained.

Some premium wines are aged in bottle prior to market release in order to take advantage of reductive changes to tannins (softer), acid (softer), and flavor compounds (increased complexity). Easy drinking wines are not so treated.

Blending

Wine blends are constructs of two or more varietals, and/or micro/macro-climates, and or clones, and/or juice types (press or free-run) that are implemented by the winemaker to, among other reasons, overcome wine deficiencies or defects, improve balance, or enhance complexity. As regards the structural components (structure, texture, flavor), Dr. Zoecklein (Components of Red Wine Mouthfeel)argues that a balanced relationship must exist between the tastes of sweetness, on the one hand, and acid, astringency and bitterness on the other, in order to yield the perception of a quality wine to the taster.  The preferred relationship is captured in his Palate Balance Equation:

         Sweet ⇄ Acid + Phenolics (Astringency and Bitterness),

where

          Sweet = Carbohydrates + Polysaccharides + Ethanol,
          Acid = Population of organic acids, and
          Phenolics = Skin, seed, and stem phenols + barrel phenols + enological tannins + volatile    phenols.

According to classof1855.com, complexity in wine is demonstrated by "multiple layers and nuances of bouquet and flavors that are formed mostly in mature wines because aging contributes to this attribute." Further, "complexity creates interest and often unfolds layer upon layer on the nose and in the mouth if the wine is at its peak. Compared to complex wines, other wines seem shallow or one-dimensional."

Easy-drinking wines are in pursuit of neither complexity or balance and so do not actively pursue these elements where blending is practiced. Blending in easy drinking wines would tend to be more blends of convenience rather than pursuit of higher quality ideals. Field blends and co-fermentations would be the order of the day, for the most part. In the case of blending extended bottle-aging wines, the lots are generally fermented and aged separately and then blended prior to bottling in a fairly extended process which may utilize multiple individuals on a blending team.

Finishing/Bottling

If the wine is made to be drunk within a year, metatartic acid can be used as an inexpensive means of establishing short-term tartrate stability (Jackson). Metatartaric acid works by restricting potassium bitartrate crystalization and interfering with the growth of calcium tartrate crystals (Jackson). This treatment is only effective for about 1 year, however, as metatartaric slowly hydrolyzes back to tartaric acid.

Given enough time, racking and fining can produce stable, crystal clear wines (Jackson). The early bottling associated with easy-drinking wines does not provide enough time for these two practices to fully clarify the wine so other procedures (such as centrifugation and filtration) have to be employed in order to realize the necessary clarity.

Filtration is avoided by many premium winemakers because of a fear of flavor-stripping.

Screw Caps are very effective as a closure choice but their usage to date has been primarily in the easy-drinking market due to "stigma" issues as well as some concern that the reductive environment they support does not promote favorable conditions for extended-bottle-aging. Synthetic stoppers have issues with oxygen ingress but are acceptable for wines that will be drunk early.

Bag-in-box containers protect wines from oxidation for 9 months or longer and are a viable container for easy-drinking wines. These containers have some stigma issues to overcome, however, and have not been used as premium wine containers.

©Wine -- Mise en abyme

Monitoring the malolactic fermentation of wine

I have recently described the malolactic fermentation process. In general, a winemaker needs to be able to tell if/when MLF has commenced, how it is proceeding, if any problems are in the offing, and when it has concluded. Key to addressing these concerns are the establishment of benchmarks for the beginning and end of MLF and a regime for tracking its progress. 

According to Boulton et al. (Principles and Practices of Winemaking, Chapman and Hall, 1996), there is a difference in perception as to what constitutes the MLF duration depending on whether you are a microbiologist or a winemaker. The microbiologist sees MLF as commencing with the “introduction of viable bacteria into the wine or must” and ending “when the bacteria have gone through the growth phase and have re-entered their final resting or stationary phase.” The winemaker, on the other hand, will equate the start of fermentation with a noticeable drop in malic acid levels (grape juice contains between 1 and 8 g/l malic acid) and as completed when the malic acid has finally disappeared (vintessential.com.au (Malolactic Fermentation Monitoring -- Resources for Winemakers) recommends a figure of < 0.05 g/l as safe for declaring the end of MLF while Katherine Mansfield (Monitoring Malolactic Fermentation 3 Ways, Cornell University College of Agriculture and Life Sciences, Cornell Cooperative Extension) cites a number of 30 mg/l).

Even though acknowledging the microbiologists' view of MLF duration, Boulton et al. accept the winemakers perspective in that they see the “measurement of the disappearance of malic acid as the accepted means for determining whether the malolactic fermentation has occurred.” Lerm et al., place this measurement of the presence of malic acid within a broader context: “The continuous monitoring of MLF is essential and often neglected by winemakers.” Continuous monitoring “allows the winemaker to follow the progress of malic acid degradation as well as the bacteria responsible for the fermentation. This is also a way for the winemaker to identify any difficulties before they can affect the quality of the wine.”

There are a number of tools that are available for the monitoring of MLF and they are summarized in the table below. 

Selected MLF Monitoring Techniques and their Attributes

Monitoring Technique	Advantages	Disadvantages
Paper Chromatography (PC) separate compounds based on their polarity visually follow disappearance of malic acid commonly used in winery	easy to use simple, affordable and indicative of MLF progress	strictly qualitative so still need quantitative values to verify MLF completion not precise not specific for L-malic acid
Thin Layer Chromatography (TLC) - similar to PC but uses TLC plates instead of paper	easy to use simple and affordable results in one hour,; much faster than PC	not precise not specific for L-malic acid strictly qualitative so still need quantitative values to verify MLF completion
Reflectance Reflectoquant® based on reflectance photometry use reactive test strips to analyze for various wine components	a fraction of the cost of a spectrophotometer half the cost of an enzymatic kit measure multiple wine parameters fastest method currently available (5 min/sample) relative accuracy of 10%	measure relative malic acid levels so still need to qualify absolute levels operating range 1 to 60 mg/L, so some samples need to be diluted or decolorised need to be calibrated with reference method
Enzymatic Analysis uses enzyme that specifically react with L-malic acid then use UV-visible spectrometer to monitor enzymatic reaction most commonly used method MLF complete if malic acid is less than 200 to 300 mg/l	quantitative excellent precision kits readily available quantify very low levels of malic acid results in 30 minutes	more complex more expensive short shelf life of reagents after activation require use of accurate micro-pipettes turbid samples need to be centrifuged
Capillary Electrophoresis (CE)	highly accurate short analysis time; fast results	extremely expensive not recommended for everyday use in winery
Fourier-transform Infrared (FT-IR) Spectroscopy - use infrared spectra to quantify wine parameters	accurate small sample volume short analysis time, fast results	expensive equipment accuracy dependent on reference values and calibration curve
High Performance Liquid Chromatography (HPLC) - separation of compounds based on polarity and interaction with stationary or solid phase	- highly accurate	extremely expensive not recommended for everyday use in winery

Source: Lerm et al., Table 7.

Of the mechanisms listed above, the one that is most commonly used in wineries today is paper chromatography. Paper chromatography detects the presence of malic acid but does not tell its concentration. As such, it should never be used to make a decision regarding the end of MLF. According to Mansfield, the lower level of malic acid detection for paper chromatography is 100 mg/L, way above the 30 mg/L considered by her lab to be the “safe number” for the end of MLF. The risk associated with prematurely calling the end of MLF is residual malic acid which can be metabolized by latent LAB in the bottle with a host of resultant problems (change in color, aroma, flavor; CO₂ production; clouding (Lisa Van de Water, Monitoring microbes during cellaring/bottling, practicalwinery.com, January/February 2010)). In a table accompanying the article, the author shows that Oenococcus would be found at levels of 500+/ml if residual malic acid were present. Ms. Van de Water suggests checking wines during MLF for the presence of spoilage bacteria and doing so by microscope as well as by PCR.

A robust MLF monitoring protocol which incorporates some of the tools listed in the table above is as follows:

1. Take representative samples of cellar barrels for testing. Care should be taken to ensure that ease of access does not determine the barrels utilized for sampling.
2. Test consistently.  I propose that the sampling be conducted every two weeks or at topping.
3. Position paper chromatography as a tool in the MLF monitoring toolbox for measuring progress. Paper chromatography detects the presence of malic acid but does not tell the concentration. The level of malic acid in the wine needs to be below 30 mg/L in order for the microbial stability benefits of the MLF process to accrue. 
4. Utilize enzymatic analysis in the MLF monitoring protocol as a means of certifying the end of MLF. As the cost of the equipment for enzymatic tests are very high, I recommend the services of an outside lab for this purpose. As recommended by vintessential.com.au, the costs of these outside tests can be minimized by submitting barrel composites. If no malic acid is detected, then all of the barrels will be assumed to have completed MLF. If the test comes back positive, then the barrels would be tested individually in order to identify the offending barrel(s).
5. Include the human factors (nose, palate, and skills; Dr Wann, Power Point presentation) in the MLF monitoring process. Dr. Wann recommends: (i) checking for CO₂ evolution; (ii) smelling and tasting the barrel at every topping (heads up on potential spoilage activity); (iii) maintaining an appropriate temperature (between 18 and 22℃); and (iv) to be aware of increasing pH levels (potential foothold for spoilage organisms). In that CO₂ is one of the outputs of the MLF process, its presence is indicative of ongoing malate degradation.
6. Institute a number of post-MLF processes to ensure the microbial stability of the wines through to the blending and bottling processes. According to Wibowo et al., additions of SO₂ and storing the wine at higher temperatures leads to the progressive loss of viability of any bacteria surviving the MLF. The wines themselves may be sterilized by filtering with membranes having pore sizes of 0.22 to 0.45µm. Lafon-Lafourcade et al., sees the decline in LAB accelerated by increasing temperatures, lowering the wine pH, and increasing the alcohol concentration, actions which, it seems, combine to provide a toxic environment for the bacteria. The addition of SO₂ does result in a rapid loss in cell viability, they agree, but growth recommences at a later date. The specific post-MLF operations that should be undertaken, according to Boulton et al. are as follows:

• Transfer the wines off the lees and a rough filtration
• Adjustments of temperature and pH and addition of SO₂. The recommendation here is for 0.8 mg/L
• Fining operations could also be performed at this time

Following this set of procedures will minimize the potential for microbes indicating their presence in the bottle.


©Wine -- Mise en abyme

The malolactic fermentation of wine

In recent posts I have discussed malic acid and lactic acid bacteria (LAB), two of the major actors in the malolactic fermentation (MLF) drama. In this post I will elaborate on the arena within which this drama unfolds and how the interaction of malic acid and other minor players with LAB results in lactic acid and other metabolites. Let us begin by defining MLF.

According to Sauvageot and Vivier (Effects of Malolactic Fermentation on Sensory Properties of Four Burgundy Wines, AJEV 48(2), 1997), MLF is a bacterial conversion -- most commonly performed by Leuconostoc strains due to their tolerance of the high acid and alcohol content associated with wine -- of L-malic acid to L-lactic acid and CO₂. The MLF process can be represented thusly (Lerm et al., Malolactic Fermentation: The ABCs of MLF, S. Afr. J. Enol. Vitic. 31(2), 2010):

L-malic acid  + LAB  →    L-lactic acid          + CO₂
(dicarboxylic)                  (monocarboxylic)

wherein a carboxyl group (C(O)OH) is removed from the dicarboxylic L-malic acid. The reaction is catalyzed by the LAB along one of three pathways (Lerm et al., Bauer and Dicks, Control of Malolactic Fermentation in Wine: A Review, S. Afr. J. Enol. Vitic. 25(2), 2004):

1. Direct conversion of malic acid to lactic acid via malate decarboxylase (the preferred pathway for wine LAB)
2. L.casei and Enterococcus faecales possess a malic enzyme that converts L-malic to pyruvic acid which is in turn reduced to lactic acid by L-lactate dehydrogenase
3. Via L. fermentum,  malate is reduced by malate dehydrogenase to oxaloacetate, followed by decarboxylation to pyruvate which is further reduced to lactic acid.

The main effects of MLF on wine are (i) a reduction in titratable acidity (by 0.1 to 0.3%) and an increase in pH (0.15 to 0.30). In addition, dramatic organoleptic changes to the wine are evidenced (Lonvaud-Funel, Microbiology of the Malolactic Fermentation: Molecular Aspects, FEMS Microbiology Letters):

• The specific taste of malic acid disappears
• Sugars are catabolized to produce mainly lactic and acetic acid
• Citric acid is transformed into acetic acid and carbonyl compounds, notably the butter-flavored diacetyl
• Wine taste and color are modified due to the metabolic activity of bacteria on phenolic compounds (tannins, anthocyannins).

By synthesizing anti-bacterial compounds and depriving the wine of nutrients, MLF also contributes to its microbial stability (Lonvaud-Funel).

In practice, most red wines, and selected white wines, undergo MLF. The process is encouraged (Bauer and Dicks, Control of Malolactic Fermentation in Wine, S. Afr. J. Enol. Vitic. 25(2), 2004): in cooler areas where grapes have high malic acid content; in cases where the wine is aged in oak barrels; and when the wine style calls for long-term aging in bottle. The practice is sometimes forsworn in warmer, lower-acid areas and in the cases where undesirable organoleptic changes or the production of biogenic amines result.

MLF is initiated either naturally or through inoculation of the wine with an LAB strain. In the case of indigenous initiation, upon the completion of alcoholic fermentation, and following a lag phase, the surviving LAB begin to multiply rapidly. MLF begins when their numbers approach 10⁶ cells/ml (Savageot and Vivier; Wibowo et al.; Lonvaud-Funel; Lerm et al.). Lafon-Lafourcade et al., posit that this growth originates from winery equipment which serve as incubators for the LAB. Oenococcus. oeni is the main species that develops here but Lactobacillus and Pediococcus spp. may proliferate and conduct the MLF if the wine pH approaches 4.0. If MLF is not desired, clarification of must or newly fermented wine will remove the majority of the LAB and, if excessive, its potential nutrient sources, and reduce the possibility of indigenous inoculation (Wibowo et al.). In addition, wines that have undergone thermovinification (rapid heating of the must to near boiling point in order to extract anthocyanins and tannins) are less susceptible to MLF (Wibowo et al.).

During the time between the end of AF and the initiation of MLF, no SO₂ can be applied to the wine because of its toxic effect on the LAB. During this time, then, the wine is exposed to the potential of oxidation and attack by spoilage organisms. The use of starter cultures reduces this risk by shortening the time between the end of AF and the initiation of MLF and by ensuring a rapid onset of MLF with a very high population of viable bacteria (on the order of 10¹¹ cells/g, according to Lerm et al.). Given the environment within which the LAB has to operate, a starter culture should have the following characteristics (Lerm et al.):

• Tolerance to low pH, high ethanol, and SO₂
• Good growth characteristics under winemaking conditions
• Compatibility with the S. cerevisiae strain(s) being used for alcoholic fermentation
• Ability to survive the production environment
• Does not produce biogenic amines
• Does not produce off-flavors or off-odors
• Production of aroma compounds that will favorably impact the wine’s aroma profile.

There are a number of factors that affect the development of LAB and, as a result, the activation and effectiveness of MLF. For example, wine pH affects (Wibowo et al.):

• LAB growth rate
• The LAB species that proliferate
• The metabolic behavior of the species that grow
• The survival of LAB.

Temperature is synergistic with ethanol levels as it relates to inhibiting LAB (Lerm et al.):

• The optimal growth temperature of LAB decreases at high ethanol concentrations
• Elevated temperatures lower the ability of LAB to withstand increased ethanol concentration
• Temperatures of 25℃ and above, combined with ethanol levels of 10 -14%, almost completely inhibits LAB growth.

A listing of the factors beyond pH and temperature that affect LAB growth and development are provided in the tables below.

TABLE 1. The influence of different winemaking practices on LAB growth

Practice	Influence
Degree of must clarification	Significant impact on bacterial growth. Yeast produce more medium chain fatty acids in highly clarified must
Skin contact prior to AF	Direct effect on extraction of nitrogenous and other macromolecules stimulate LAB growth and malolactic activity
Choice of yeast strain	Inhibitory and stimulatory effects differ between strains
Aging of wine on yeast lees	Yeast autolysis release nutrients that stimulate LAB growth and malolactic activity

Source: Lerm et al., TABLE 2

TABLE 2. Yeast activity inhibiting LAB via the production of yeast metabolites

Yeast Metabolite	Effect on LAB and/or MLF
Ethanol	Affects growth ability
SO2	AF with SO2 producing yeast strain results in wine inhibitory to MLF
Medium chain fatty acids	Affect LAB growth and reduce ability to metabolise malic acid. Combination of fatty acids (hexanoic, octanoic and decanoic acid) cause greater inhibition than individual compounds.
Metabolites of protein nature	Peptide produced by S. cerevisiae during AF: inhibit O. oeni by disruption of cell membrane; inhibition dependent on SO2

Source: Lerm et al., TABLE 3

According to Boulton et al. (Principles and Practices of Winemaking, Chapman and Hall, 1996), there is a difference in perception as to what constitutes the MLF period, depending on whether you are a microbiologist or a winemaker. The microbiologist measures the MLF from the “introduction of viable bacteria into the wine or must” and it ends “when the bacteria have gone through the growth phase and have re-entered their final resting or stationary phase.” The winemaker, on the other hand, will equate the start of fermentation with a noticeable drop in malic acid levels (grape juice contains between 1 and 8 g/l malic acid) and as completed when the malic acid has finally disappeared (vintessential.com.au (Malolactic Fermentation Monitoring -- Resources for Winemakers) recommends a figure of < 0.05 g/l as safe for declaring the end of MLF while Katherine Mansfield (Monitoring Malolactic Fermentation 3 Ways, Cornell Cooperative Extension) cites the number 30 g/ml. That discrepancy in the metric notwithstanding, Boulton et al., see the “measurement of the disappearance of malic acid as the accepted means for determining whether the malolactic fermentation has occurred.”

Measurement of the degradation of malic acid is one of the key aspects of monitoring MLF, the topic of my next post

©Wine -- Mise en abyme

Lactic acid bacteria: Pre-, intra-, and post-malolactic-fermentation development

Either concurrent with, or post-alcoholic-fermentation, a number of wines are subjected to a process --malolactic fermentation (MLF) -- wherein the harder malic acid is converted to lactic acid by lactic acid bacteria (LAB). The perceived benefits of this process to the final wine are:

• Deacidification, with a resultant increase in pH
• Increased microbial stability through removal of malic acid as a possible carbon substrate; and
• Modification of the wine's aroma profile.

I described the origin and evolution of malic acid in my most recent post and provide a similar treatment of LAB in the current.

The genera from which the LABs are drawn are shown below. The species associated with wine are (Wibowo et al., AJEV, 36 (4) 1988):

• Oenococcus.oeni -- mainly during MLF
• Pediococcus cerevisiae -- mostly after MLF; predominantly in wines with high pH
• Pediococcus pentosaceous -- mostly after MLF; predominantly in wines with high pH
• Lactobacillus spp. -- mainly after MLF.

Phylogenetic trees of Lactobacillales constructed on the
basis of concatenated alignments of ribosomal proteins
Source: pnas.org

Doctor Murli Dahrmadikari (Lactic Acid Bacteria and Wine Spoilage, extension.iastate.edu) describes LAB thusly:

These organisms are gram positive, catalase negative, nonsporing cocci, coccobacilli or rods. They are microaerophilic (sic) that means that they grow well under conditions of low oxygen content. Since they can grow under low oxygen conditions, they can grow throughout the wine (as opposed to on the surface of the wine) even though the container is kept full. The bacteria can metabolize sugars, acids and other constituents in wine and produce several compounds. Some of these are undesirable and constitute spoilage.

LAB utilize two pathways for the metabolism of glucose and a third for the metabolism of pentose (Lerm et al., Malolactic Fermentation: The ABCs of MLF, S. Afr. J. Enol. Vitic. 31 (2), 2010). One of the glucose pathways (EMP) converts glucose into pyruvate over a number of steps and then into lactic acid. In this process, 1 mole of glucose yields 2 moles of lactic acid plus 2 ATPs. This process is called homolactic fermentation and all Pediococcus species utilize this mechanism. The second glucose pathway (6-PG/PK) yields lactic acid, carbon dioxide, ethanol, acetate, and 1 ATP. The bacteria utilizing this pathway are called heterolactic fermenters and this includes all strains of Leuconstoc, some Lactobacillus strains and Oenococcus.oeni. The pentose pathway combines pentose with a phosphate derivative before coalescing with the later portions of the 6-PG/Pk pathway for completion. The outputs of the pentose pathway are lactic acid, acetic acid, and carbon dioxide.

The LAB in wine originate from grape and grape leaves and are brought into the winery at harvest (Dharmadhikari; Lerm et al.; Wibowo et al.). The LAB diversity and population density are limited by  grape maturity and sanitary conditions and levels are generally on the order of 100 cells/gm (Wibowo et al.) The species that are present at this time include Pediococcus and Leuconstoc (Lerm et al.).

Once the grapes are brought into the winery, the LAB population increases dramatically, implicating the winery environment in this proliferation. According to Lerm et al., the population rises to 10³ - 10⁴ cells per ml shortly after crushing and prior to alcoholic fermentation (AF). In a study of the evolution of LAB, Lafon-Lafourcade et al. (Occurrence of Lactic Acid Bacteria During the Different Stages of Vinification and Conservation of Wines, Appl. Environ. Microbiol. 1983, 46 (4)) show populations of 10⁴ cells/ml at 14℃ and 19℃ if (i) unsulfited and (ii) sulfited at 50 mg/l. At 100 mg/l of SO2, however, the LAB population declines 10-fold. The LAB species present at this stage are L. plantarium, L. casei, Leuconstoc mesenteries, P. damnosus, and, to a lesser extent, O. oeni (Lerm et al.).

During the course of AF, the LAB population levels fall precipitously with only 200 cells/ml surviving the process (Wibowo et al.; Lafon-Lafourcade et al.). Wibowo et al., attribute this decline to ethanol sensitivity. The only LAB strain that survives AF is O. Oeni.

AF is followed by a lag phase which is, in turn, followed by rapid proliferation of LAB with levels rising to between 10⁶ - 10¹⁰ prior to MLF. Lafon-Lafourcade et al., posit that this growth originates from winery equipment which serve as incubators for the LAB that would perform a natural inoculation of MLF. O. oeni is the main species that develops here but Lactobacillus and Pediococcus spp. may proliferate and conduct the MLF if the wine pH approaches 4.0.

There are a number of factors that affect the development of LAB and, as a result, the activation and effectiveness of MLF. For example, wine pH affects (Wibowo et al.):

• LAB growth rate
• The LAB species that proliferate
• The metabolic behavior of the species that grow
• The survival of LAB.

A listing of the factors beyond pH that affect LAB growth and development are provided in Tables 1 and 2 below.

TABLE 1. The influence of different winemaking practices on LAB growth

Practice	Influence
Degree of must clarification	Significant impact on bacterial growth. Yeast produce more medium chain fatty acids in highly clarified must
Skin contact prior to AF	Direct effect on extraction of nitrogenous and other macromolecules stimulate LAB growth and malolactic activity
Choice of yeast strain	Inhibitory and stimulatory effects differ between strains
Aging of wine on yeast lees	Yeast autolysis release nutrients that stimulate LAB growth and malolactic activity

Source: Lerm et al., TABLE 2

TABLE 2. Yeast activity inhibiting LAB via the production of yeast metabolites

Yeast Metabolite	Effect on LAB and/or MLF
Ethanol	Affects growth ability
SO2	AF with SO2 producing yeast strain results in wine inhibitory to MLF
Medium chain fatty acids	Affect LAB growth and reduce ability to metabolise malic acid. Combination of fatty acids (hexanoic, octanoic and decanoic acid) cause greater inhibition than individual compounds.
Metabolites of protein nature	Peptide produced by S. cerevisiae during AF: inhibit O. oeni by disruption of cell membrane; inhibition dependent on SO2

Source: Lerm et al., TABLE 3

According to Wibowo et al., additions of sulfur dioxide, and storing the wine at higher temperatures, leads to the progressive loss of viability of any LAB surviving MLF. According to Lafon-Lafourcade et al., under standard conditions, the surviving LAB remained viable post-MLF, exhibiting, initially, a tendency for further growth and then showing a slow, progressive decline over a 200-day storage period. At temperatures over 20℃, a rapid decline in viability was recorded and at 26℃, no LAB activity was recorded after 80 days. Lafon-Lafourcade et al. found that the decline in LAB viability was accelerated by increasing temperature, lowering the wine pH, and increasing the alcohol concentration. These actions, it seems, combine to provide a toxic environment for the bacteria. Addition of sulfur dioxide, they agree, does result in a rapid loss in cell viability but growth recommences at a later date.

At the end of barrel aging, the wine microbial population is stabilized but its population prior to bottling is 103 - 104 cells/ml, if unfiltered, with microbes such as Acetobacter, S. cerevisiae, and O. oeni predominant (Renouf et al., Survival of Wine Microorganisms in the Bottle during Storage, AJEV 58(3), 2007). Filtering with a 0.4 micron filter sheet will eliminate all bacteria from the wine.

Now that the two most important players in the MLF drama have been described, I will cover the MLF process itself when I revisit this topic.

©Wine -- Mise en abyme