Oral Bacteria Shown to Produce Aromatic Volatiles from Glycosidic Precursors: Implications for Perception of Aromas and Flavors in Wine

How a wine tastes is dependent upon many factors, including (but not limited to) the variety, the vintage, where the grapes are grown (soil, climate, etc), as well as the viticultural and winemaking techniques employed during processing.  The compounds responsible for how wine tastes are known as free volatile compounds, as well as aromatic precursors, the latter of which are present at much higher concentrations.  Non-volatile sugar-bound conjugates (a.k.a. “glycosidic compounds) have been well studied and have been shown to be released over time during wine aging or by using specific winemaking techniques. Specific glycosidic compounds known to be released over time, thus affecting how a wine develops and tastes, include terpenes, C13 norisoprenoids, benzenic derivatives, volatile phenols, and C6 compounds.  All of these glycosidic compounds have low

Photo By Véronique PAGNIER (Own work) [Public domain], via Wikimedia Commons

Photo By Véronique PAGNIER (Own work) [Public domain], via Wikimedia Commons

odor thresholds, thus requiring very little to elicit a sensory response.

While wine aromatics have been extensively studied, it is not well known exactly how compounds responsible for aromatic character in wines interact with the physiological make-up of the human mouth. In addition to environmental and chemical sources, it is possible that the perception of different wine aromas can be altered by physiological factors like mouth temperature, saliva composition, or the oral microbial community present in each individuals’ mouths. Studies focusing on onions, bell peppers, and grapes found that the microbial community in the human mouth hydrolyzed odorless compounds into their corresponding volatile aromatic compounds, giving reason to believe something similar could potentially happen with wine.  Perhaps the microbiota living in the human mouth can hydrolyze these odorless precursors and convert them into their corresponding aromatic compounds, just like it’s been shown with other foods.

A 2015 study in the journal Food Chemistry aimed to evaluate whether or not human oral microbiota can convert odorless aromatic precursor compounds in wine into their corresponding aromatic glycosidic compounds. The results could potentially have a profound impact on our understanding of how we taste and evaluate wines.

Brief Methods

Experiment 1: In vitro

For the in vitro experiments, three microbes commonly found in the human mouth were cultured on sterile growth media.  Different concentrations of grape extract were added to the microbes, with bacteria growth measured after 24-48 hours, depending upon the specific microbe.  From these growth measurements, inhibition of growth was also calculated.

(Want more details? Just ask…)

Experiment 2: Ex vivo:

For the ex vivo experiments, fresh saliva was collected from three volunteers (ages 28-31).  Prior to collection, the volunteers had not taken any antibiotics or other medications, and were non-smokers.  Volunteers did not consume any food or

Photo courtesy Flickr user Hamed Saber

Photo courtesy Flickr user Hamed Saber

beverages within two hours of the saliva collection time.

The saliva from each volunteer was divided up into four different treatments: 1) fresh saliva in an aerobic culture, 2) fresh saliva in an anaerobic culture, 3) sterilized saliva (pasteurized), and 4) non-enzymatic saliva (heated).  Microbe counts were measured after 24-48 hours at 37oC.

Grape extract (comparable to 40g of grapes) was added to the saliva cultures and bacterial growth was monitored.  Glycoside hydrolysis by the microbes was measured by monitoring and analyzing the volatile compounds released after four time periods (0hr, 2hr, 24hr, and 72hr).

To test the ability of human oral microbiota to hydrolyze glycosidic compounds in general, a standard solution of octyl-β-D-glucopyranoside was cultured with the saliva samples and monitored over time.

Selected Results

  • No human oral microbiota was found in the sterile or non-enzymatic saliva treatments (as expected).
  • Adding octyl-β-D-glucopyranoside to human saliva resulted in hydrolysis and the release of the volatile compound aglycone.
  • None of the oral microbes were inhibited by the glycosidic extract.
  • Every oral microbe tested was able to hydrolyze the glycosidic compounds in the grape extract, resulting in the release of terpenes, benzenic derivatives, and C6 alcohols.
    • Note: Many of these compounds can be associated with various aromatic characteristics in wines (e.g. terpenes can produce flowery or citrus aromas and certain benzenic compounds such as β-phenylethanol can produce rose aromas.)
  • While some aromatic compounds were found as a result of the oral microbes hydrolyzing the glycosidic compounds in the grape extract, other common compounds were not present (i.e. C-13 norisoprenoids, vanillins, and volatile phenols).
  • The ability to hydrolyze and the resulting aromatic compounds produced from this hydrolysis depended upon the type of oral microbe present.
    • A. naeslundii was the producer of the most linalool and its oxides, which is associated with floral notes in grapes and wine.
    • E. faecalis, A. naeslundii, and S. mutans produced the most aroma-causing aglycones.
    • G. adiascens, V. dispar, and F. nucleatum produced the fewest aroma-causing aglycones.
      • NOTE: Some of these bacteria have difficulty growing in culture media, so it isn’t clear whether their lack of glycosidic hydrolysis is real or a function of being unable to grow under the experimental conditions.
    • There were significant differences between the three volunteers in terms of the species make up of their mouths. No person had the same species in the same amounts.
    • While the microbial counts were statistically the same for each volunteer, the aromatic aglycones produced were significantly different, which is likely due to the different species present in each individual’s mouth.


Overall, I found this to be a very fascinating study.  Results indicated that oral microbiota (a.k.a the bacteria living in the human mouth) is able to hydrolyze glycosidic compounds found in grapes and wine, resulting in the production of aromatic volatile compounds. Interestingly, it was also found that of the three volunteers, none of them had the same oral microbial make-up, which led to each of them hydrolyzing different compounds.  This result could have profound implications on how we understand wine tasting and the perception of aromas and flavors for any given wine.  For example, if someone has a mouth chock full of A. naeslundii, they may be more likely to pick up floral notes in the wine that they are tasting, since A. naeslundii was found to hydrolyze glycosidic

Photo By Multimotyl (Own work) [CC BY-SA 3.0 (http://creativecommons.org/licenses/by-sa/3.0) or GFDL (http://www.gnu.org/copyleft/fdl.html)], via Wikimedia Commons

Photo By Multimotyl (Own work) [CC BY-SA 3.0 (http://creativecommons.org/licenses/by-sa/3.0) or GFDL (http://www.gnu.org/copyleft/fdl.html)], via Wikimedia Commons

compounds into linalools, which are associated with floral aromas.  Alternatively, another person with a different percentage of A. naeslundii present in their mouths may not pick up on these floral notes as well, leading to a marked difference in how each of those two individuals perceived the aromas and flavors of the given wine.

Unfortunately, this study did not include a sensory analysis, which would have been a potentially fascinating addition.  Laboratory tests concluded that certain oral bacteria interact with the grape/wine compounds to produce aromatic volatiles, but does this really translate into perceived aroma/flavors? On paper, someone with larger A. naeslundii populations should perceive more floral aromas, but does this actually happen?  I’d like to see the study repeated with the corresponding sensory analysis.

I’d also like to see this study repeated on a much larger scale.  Three saliva donors is not very many at all, so having many more volunteers providing their saliva could make the results that much stronger.

Overall, I thought this was a really cool study that provides support for the idea that human oral microbial communities are at least partially responsible for the perception of aromas and flavors in wine.  Depending upon what bacteria in your mouth, you might pick up certain aromas and flavors while a person next to you with a different oral bacteria community may pick up different aromas.

Perhaps it’s not simply a matter of tasting a lot of wines to improve one’s ability to perceive different aromas and flavors in wine—it is at least in part dependent upon the bacteria populations in your mouth.  If you don’t have the right bacteria, you may never pick up certain aromas or flavors, regardless of how much you practice with standard solutions.

What do you all think of this study?  Do you have any questions regarding the methods or perhaps results that I did not cover? Note: due to space considerations, I did not include everything that was found in the study, so if there is a specific question you have regarding the methods or results that you did not see in my summary, feel free to ask and I’ll go back through the original paper to see if it was covered.

Source: Muñoz-González, C., Cueva, C., Pozo-Bayón, M.A., Moreno-Arribas, M.V. 2015. Ability of human oral microbiota to produce wine odorant aglycones from odourless grape glycosidic aroma precursors. Food Chemistry 187: 112-119.

4 comments for “Oral Bacteria Shown to Produce Aromatic Volatiles from Glycosidic Precursors: Implications for Perception of Aromas and Flavors in Wine

  1. Andrew Holod
    March 31, 2016 at 1:22 pm

    It would seem that retro-nasal olfaction is key to sensing these newly hydrolzed aromatic molecules. The dilemma with including a sensory evaluation is that there is no way to standardize an individual’s tasting mechanism (mouth, olfactory epithelium, manner of in-mouth aeration etc). I agree a larger data set would prove to be more useful. Further an experiment with actual wine as opposed to grape extracts could be of value. Thanks for the excellent post!

  2. Curious
    March 31, 2016 at 6:13 pm

    With the first measurement being taken at the 2 hour mark, was the change at that point so significant that you can infer there may be a perceptible change within seconds?

  3. April 1, 2016 at 1:58 pm

    Fascinating study, Becca. I’m in the process of being tested to identify the microbiome of my small intestine and address a condition called Small Intestine Bacterial Overgrowth (SIBO). The Organic Acid Test (OAT) is a relatively rare test used to diagnose chronic illness. It takes a snapshot of your metabolism and measures 74 markers including populations of yeast and bacteria so any imbalance can be adjusted through medication and diet. I’m sure that there are tests that can map oral microbiota as well. I’d like to see a study that ‘maps’ the microbiomes of a group of top-notch professional tasters to see how they compare. According to my physician, we know almost nothing about the human microbiome so I’m sure that there’s a lot we could learn.

  4. Michael Sheldrick
    April 1, 2016 at 11:56 pm

    Wow! What a great site. A terrific distillation of the paper. When will be able to hack the human microbiome to maximize our sensory experience or health experience, or both?

    A massive problem with more than 600 taxa so far identified: http://www.homd.org

Comments are closed.