Wednesday, November 30, 2016

"The Sweet Wine Lovers' Manifesto"

Friend of Pairteas and wine genius Tim Hanni MW (= Master of Wine) is working on a new book, “The Sweet Wine Lovers' Manifesto,” and I’ve just read through a draft. It’s a bit embarrassing insofar as he mentions me a bit, but then I do prefer sweet wines, so the book does speak to my interests and preferences. 

What sweetness does, as I have mentioned before, is to cut the burn of alcohol to which I am genetically very sensitive. According to my 23andme results, I carry T at rs161364 in both of my copies of the TRPV1 gene, the hot receptor, on chromosome 17.  This means that I am exquisitely sensitive to the burn of alcohol.* This double T state is also relatively rare in people of European descent like me: while about 39% of people of European descent carry one T, only 10% carry two. The percentage of Asians and Native Americans with two T’s is even lower, between 4 and 5%, while the percentage of Sub-Saharan Africans is essentially zero.** In other words, the T is a gain-of-function genetic mutation that occurred after anatomically modern humans walked out of Africa to settle the rest of the world.

Interestingly, alcohol is also bitter. It activates at least two different bitter receptors: TAS2R13 and TAS2R38.* The latter receptor is also the one that responds to PROP (6-n-propylthiouracil) that researchers have used extensively to test for taste sensitivity. I happen to have the genetic markers that lead to high sensitivity to bitter for both of these receptors as well, but the burn I get from alcohol is so intense that bitterness takes a back seat.

Another genetic feature that I have: the most active version of a protein called gustin. This protein helps to govern the number and function of your taste papillae.*** A person can carry one or two copies of the most active (A) form of the gene for this protein, and I happen to carry two (I’m A/A at rs2274333), hence my tongue is a carpet of taste papillae and taste buds. Lest you think this is a great thing, you should know that being so sensitive means that I taste nasty stuff more strongly, too, and alcohol burns all the more.

This is my tongue, stained with blue food coloring. Taste papillae don't stain very much, which is why most of the tongue is pink, while the upper part, towards the back of the tongue, is blue. The fissures appear when there is a carpet of taste papillae, and you can also see some round papillae sticking up separately. Compare with the picture below, of a person with few papillae—most of the tongue is blue; the taste papillae are the pinkish dots that are scattered on the tongue surface.

Sweet activates TRPV5, which turns off TRPV1—that’s why I can tolerate Harvey’s Bristol Cream Sherry at 17.6% alcohol, while I cannot abide a Cabernet Sauvignon at 16% alcohol.

Though another confession is needed here, namely that I hate the taste of green peppers, which is characteristic of Cabs…but it’s the burn that gets to me first.

In any case, while genetics alters your degree of sweet wine liking (or to be more exact, your degree of high alcohol dry wine dislike), in fact more people enjoy sweet wines than really enjoy those high alcohol dry wines. According to Tim, if it weren’t for the snob appeal of the latter, people would be drinking sweet wines much more often, and not just with dessert. Keep an eye out for his book with more details!

* Alissa L. Allen, John E. McGeary, and John E. Hayes. Polymorphisms in TRPV1 and TAS2Rs associate with sensations from sampled ethanol. Alcohol Clin Exp Res. 2014 October ; 38(10): 2550–2560. doi:10.1111/acer.12527. 

** Note I used the HAPmap data for these percentages. While some people quarrel with HAPmap data, those percentages seem to correspond to my experience.

*** Melis M, Atzori E, Cabras S, Zonza A, Calò C, et al. (2013). The Gustin (CA6) Gene Polymorphism, rs2274333 (A/G), as a Mechanistic Link between PROP Tasting and Fungiform Taste Papilla Density and Maintenance. PLoS ONE 8(9): e74151. doi:10.1371/journal.pone.0074151.

Tuesday, November 22, 2016

The flavors of your tea: cold versus hot brewing

Have been asked to give a talk at next year’s World Tea Expo about the differences between hot and cold brewed tea. Friend of Pairteas Marzi Pecen was recently in Japan, where she especially enjoyed a cold brewed oolong—enjoyed it even more than hot brewed! — so I was curious to find out how cold brewing might affect the composition of the resulting tea.

To create a tea with cold brewing you rely on the water solubility of compounds at roughly room temperature. By contrast, many compounds that are not soluble at room temperature become more soluble and are released from the tea leaves at hot temperatures. One major group of compounds that are barely soluble at cooler temperatures are the catechins and polyphenols. Their release from tea requires near boiling to boiling temperatures, and a longer steep even then. For a cold brew to become significantly bitter, the brewing time has to be quite prolonged and the proportion of leaf to water high.*

Another compound of interest in oolong is dimethyl sulfide. That is the compound that gives the tea, whether green or oolong, a seashore/oceanic/marine flavor, and is one of the two major compounds in oolong that make people dislike the tea. It is insoluble at room temperature. The other compound is indole, which at very low concentrations is pleasantly floral, but at higher concentrations (or if you are sensitive to it) has an odor that can be politely called animalic. Interestingly, indole is barely water soluble at room temperature, so it will appear in a cold brew at very low concentrations if properly done—in other words at the concentration range where it is pleasant. 

What will appear in a cold brew are the delightfully malty, fruity, and floral compounds in oolong, for example nerolidol, which is considered the hallmark of high quality oolong, and (R)-(-)-linalool, which gives the tea a lavender and sweet basil-like aroma, together with the jasmine-related compounds that make really good oolongs so distinctive.**

Above is nerolidol — the double lines indicate double bonds between carbon molecules, with each carbon at the juncture between lines and at each end. This compound is in the form of a relatively short carbon chain, with an OH (oxygen-hydrogen) group, which help make it more soluble in water—the molecule can let go of the hydrogen, which is then "replaced" by the hydrogen of a water molecule.
Contrast this structure with indole, below, which has two linked circles of carbon atoms—these circle structures are much less soluble in water, because they tend to bond to themselves in a stack, and water can't get in between. Images from Wikipedia.

So if you like your tea bitter and astringent, or if you want catechins and polyphenols for health reasons, hot brewing is the way to go, but if you want to appreciate the deliciously delicate flavors that oolongs offer, it may be best to cold-brew them!

* Sheng-Dun Lin, Joan-Hwa Yang, Yun-Jung Hsieh, En-Hui Liu, Jeng-Leun Mau. Effect of Different Brewing Methods on Quality of Green Tea. Journal of Food Processing and Preservation . 38 (2014) 1234–1243. doi:10.1111/jfpp.12084. 

** Zhu J, Chen F, Wang L, Niu Y, Yu D, Shu C, Chen H, Wang H, Xiao Z. Comparison of Aroma-Active Volatiles in Oolong Tea Infusions Using GC-Olfactometry, GC-FPD, and GC-MS. J Agric Food Chem. 2015 Sep 2;63(34):7499-510. doi: 10.1021/acs.jafc.5b02358. Epub 2015 Aug 19.

Wednesday, November 16, 2016

Remember the previous two posts, where I discussed an article about wine liking and the effects of cheese on liking?* There was one wine, in particular — Madiran — that had high astringency, and was relatively disliked compared with other wines used in the study, and disliked even more with each successive sip. Cheese significantly decreased the progressive dislike, and I attributed this change to the effects of fat in the cheese. Here I reconsider this proposition.

Tannat grapes used in Madiran — wines made from these grapes in France have an extra high tannin content. Image from Wikipedia.

Friend of Pairteas and wine expert Tim Hanni MW pointed out to me the other day that he had tried the experiment of evaluating wine astringency with and without olive oil, to test the notion that astringency could be diminished by fat. He found that olive oil did nothing to diminish astringency, and suggested that the effect of the cheeses on wine astringency that I discussed in last week’s post was due to other factors in the cheese, such as salt and sourness, and not due to the fat content. In fact, as he found, you can diminish wine's astringency by taking a pinch of salt and then sucking on a lemon before drinking your wine. I've seen this process in action and it really works.

This observation got me thinking about the very complex system of trigeminal and taste activation. 

First, how do tannins interact with the trigeminal receptors to create the sensation of  astringency? 

Tanninns have to bind to both bitter receptors and TRPV1, the trigeminal hot receptor, simultaneously. If they are displaced from either of these receptors by another compound your won’t sense astringency. 

So what do salt and lemon juice do in this regard?

With respect to salt, it activates a specific part of the TRPV1 receptor, and changes its conformation. The following is speculation on my part, but it is likely that the salt, by activating the TRPV1 receptor in its way, bumps the tannins off their binding site, so the can no longer cause astringency. Further, salt inhibits the ability of bitter sensitive taste bud cells to send on their message, providing another way for the sensation of astringency to be blocked.

Second, lemon juice contains compounds that interact with bitterness receptors, acid receptors, and cold receptors (TRPA1), so it can act through each of these pathways to influence astringency. By interacting with bitterness receptors, lemon juice can bump tannins off the bitter attachments they need to create the astringency sensation. Next, the pathway for bitterness sensation acts through cells that respond to citric acid. If these cells are activated by sufficient amounts of acid, the message they send will be dominated by sourness rather than bitterness, so astringency will be inhibited. Finally, when you activate TRPA1 you inhibit TRPV1, the trigeminal receptor needed for astringency sensation, so here you have yet another way in which lemon juice could decrease astringency.

Why didn’t olive oil work to decrease wine’s astringency? As we noted before, fats do inhibit TRPV1 activity, so one would expect an effect—after all, you can turn down the heat of capsaicin with some fatty food. Two possibilities come to mind. 

One possibility is that the part of TRPV1 to which tannins bind is not affected by the presence of fat. It is worth noting that there is good evidence that capsaicin binds to a different part of TRPV1 from salt—changes in the DNA of TRPV1 that increase capsaicin heat are in a different place in the resulting TRPV1 molecule from changes in the DNA that affect salt sensitivity. So the notion that tannins might bind a different site in the molecule from fat is not unreasonable.

The second possibility has to do with olive oil itself. Good quality olive oil (and Tim would be sure to use the best!!) is astringent all by itself, plus it contains bitter compounds, as well as oleocanthol, which activates another trigeminal receptor, the cold receptor TRPA1—activation of this receptor gives you the catch in your throat when you take in really good olive oil. That catch feeling may augment the aversive sensations rather than decrease them. 

That said, one thing is quite clear: if you don’t like bitterness and astringency in your wine or your tea, add a tiny bit of salt and some lemon juice, and violà — deliciousness prevails!

Mara V. Galmarini, Anne-Laure Loiseau, Michel Visalli, and Pascal Schlich. Use of Multi-Intake Temporal Dominance of Sensations (TDS) to Evaluate the Influence of Cheese on Wine Perception. Journal of Food Science Vol. 81, Nr. 10, 2016. doi: 10.1111/1750-3841.13500. 

Monday, November 7, 2016

Wine and Cheese Pairing: 2 - wines with cheeses

In my previous post, I discussed a fascinating paper by Galmarinin and her colleagues, where French study participants tried four different wines with and without cheese.* In that post, I noted that the participants preferred the sweet white wine (Pacherenc) over the drier white wine (Sancerre); the two white wines over the Pinot Noir (Bourgogne); and all of these wines over the wine with the highest level of alcohol and of tannins and therefore astringency, Madiran (a blend of Cabernet Sauvignon and Tannat).

In this post, we will look at the effects of the cheeses consumed before sips of the wine. The researchers used four cheeses, as shown in the following table:

Note: all the cheese contained between 73 and 75% fat.

This diagram shows the effect of the four cheeses on wine liking:

As you can see, none of the cheeses had any effect on liking of the Pacherenc, still the preferred wine overall. These cheeses had only a slight effect on Sancerre liking. Where we see a difference is with the Bourgogne and the Madiran. With respect to the Bourgogne, the cheeses eliminate the slight decrease in liking from sip 1 to sip 3. With the Madiran, the decrease in liking with successive sips in the absence of cheese is notable, but when the cheeses are consumed between sips there is actually an increase in liking. Overall, Madiran still wasn’t liked as much as the other wines, but the difference among them was minimized.

What is the cheese doing? 

The first graph below shows the results on astringency when the cheeses are consumed. I didn’t include Pacherenc in this graph because it is not astringent, even when sipped alone. 

As you can see, cheese decreases the astringency of each of the other wines. Sancerre has relatively little astringency, so the decreases are minor and the effect of the cheeses on Sancerre liking is minor as well. As astringency in the absence of cheese rises, the effect of the cheeses is more dramatic. In the case of Bourgogne wine, cheese brings the perceived astringency down to levels comparable to those of Sancerre without cheese; in the case of the Madiran it brings the levels to slightly below those of Bourgogne without cheese. These results parallel the results for those of liking, suggesting that that 1) astringency is aversive for this group of tasters; and 2) that these cheeses have an effect on astringency. 

How do cheese affect astringency? 

In two linked ways: first, the fat in cheese turns off TRPV1, the hot receptor and one of the two receptors necessary for astringency to be perceived; second, by turning off TRPV1, TRPA1 (the cold receptor) can be activated. TRPA1 is also activated by the pungency in the cheese, further turning off TRPV1. The net effect is a decrease in astringency, so dislike of the wines is decreased.

What about the relative increase liking that we see with the Madiran when coupled with cheese?

The graph below shows what happens with three significant characteristics of the Madiran:

As you can see, with the cheeses' sourness—another aversive characteristic—decreases, while the perception of the wine’s red fruits increases. In other words the cheeses allowed the pleasant flavors hidden in the wine to come forward. Note that Roquefort is least effective at decreasing astringency and sourness and also least effective in increasing red fruit perception.

Incidentally, the red fruit flavors are perceived with the help of another receptor, TRPV3, a warm receptor. TRPV3 is also inhibited by activation of TRPV1. If TRPV1 is inactivated, then the red fruit chemicals that activate TRPV3 have a chance to be perceived. 

My take-homes from this paper are: 
  • First, it important to consider the effects of wines and foods over time, such as in successive sips—these authors are among the first to take a serious look at this all important feature of our normal consumption, and to develop a method for recording and identifying the changes people perceive as they work their way through a meal.
  • Second, as my friend Tim Hanni** points out, “Big Reds,” such as the Madiran used in this study, are generally not as much liked as the less astringent wines—this observation makes me think that liking a “Big Red” is as much a macho thing of reveling in the aversive, as it is about diminished sensitivity to the effects of TRPV1 activation—something like eating super hot chili peppers to show off. We must remember, however, that the pain induced by activation of TRPV1 by chili peppers declines the more you eat them. It may then be true that the more Big Reds you drink, the more inured and eventually insensitive you are to their astringency.
  • Finally, these data strongly suggest that the biology of TRPV1, TRPA1, and TRPV3 as I have outlined in previous posts holds true.
So if you are stuck with a wine that you don't like because of its astringency, by all means ask for some cheese to go with it!

Mara V. Galmarini, Anne-Laure Loiseau, Michel Visalli, and Pascal Schlich. Use of Multi-Intake Temporal Dominance of Sensations (TDS) to Evaluate the Influence of Cheese on Wine Perception. Journal of Food Science Vol. 81, Nr. 10, 2016. doi: 10.1111/1750-3841.13500.


Tuesday, November 1, 2016

Wine and cheese pairing: I - the wines

My attention was called to a fascinating paper about the effects of pairing cheeses with wines, by a French group at the Centre des Sciences du Goût et de l’Alimentation, CNRS, INRA, Univ. Bourgogne, Franche-Comté, in Dijon, France.* In the next set of posts I will review this paper in detail, because it illustrates very clearly several points that I have been making in this blog.

First, about the wines themselves (am saving the discussion of cheese effects for the next post):

The four wines chosen for the study were Pacherenc, Sancerre, Bourgogne, and Madiran. Pacherenc is a sweet white wine, Sancerre is a dry one, Bourgogne is a Pinot Noir, and finally Madiran is a combination of Cabernet Sauvignon and Tannat grapes, with a high alcohol (15.2%) and tannin content. 

The consumers in the study—31 local Dijonnais who drank wine and ate cheese regularly, and who had participated in at least one previous sensory study—evaluated the wines initially and then after a second and a third sip. Evaluation consisted of a measure of liking and measures of sensory characteristics such as sweetness, astringency and sourness, as well as fruitiness.

These consumers definitely preferred the sweet Pacherenc wine to the others, and particularly to the Madiran—initial liking scores were double for the Pacherenc compared to the Madiran!  Furthermore, with each succeeding sip, liking for the Pacherenc remained the same or even increased slightly, while liking scores for the Madiran decreased significantly as astringency came to dominate the sensory impression. 

These results fall perfectly in line with the biology of sweetness and astringency: sweetness is perceived at the beginning of a sip and tends to fade a little bit as the sip progresses, but can return in full force with subsequent sips. By contrast, astringency takes a few “beats” to kick in before it starts to dominate a flavor profile.

The difference lies in the function of the receptors and their second messengers. In the case of sweetness, the process of perception involves a series of reactions in the taste cell that occur quite rapidly. The receptors for sweet compounds grab their respective molecules, and send the message through a series of coupled reactions to TRPM5, the second messenger, which in turn enables the cells to send the “sweet” message to the brain. TRPM5 turns on quickly, and then turns off quickly, so you sense a decrease in sweetness over time after the sip. Here are the results for the Pacherenc:

This figure was extracted from Figure 3 of the article, and shows the results for there successive sips of the Pacherenc wine. The thickness of the bars is proportional to the intensity of the sensation, and the length refers to the duration. "The x-axis of each graph represents standardized time between 0 and 1. Different letters on liking scores represent significant differences among sips for each wine according to LSD test. "

By contrast, astringency activates TRPV1, the hot receptor, which is activated by alcohol as well. TRPV1 is a “slow-on slow-off” receptor, so the effect builds. Think of your first bite of a food liberally sprinkled with hot peppers. You may say, “Oh it isn’t that hot!” only to experience a burst of pain milliseconds later. And as you keep eating the food, the effect gets stronger and stronger, and lingers long after you have stopped eating. This is what happens with astringency as well—by the third sip, the astringency is there all the time, and pretty much dominates the picture, as you can see in the diagram below.

This figure was extracted from Figure 3 of the article, and shows the results for three successive sips of the Madiran wine. Note the significant decrease in liking, the slower onset of astringency with the first sip, its quicker onset by the second sip, and its greater overall intensity with the third sip.

It’s worth noting that the alcohol perception for the Madiran wine was slight to non-existent, despite the wine's high alcohol content. In the presence of high tannins, astringency is sensed in preference to the alcohol burn, because astringency involves coupled signals with bitter receptors; the result is that our brains tend to choose astringency as the overall sensation. In addition, all the other possible flavors are virtually drowned out by astringency, and red fruits only have a chance to be perceived when the person actively switches from sensing astringency to sensing the fruit flavors.

Wine Wizard and Friend of Pairteas Tim Hanni MW (= Master of Wine) has been trying to promote the notion that many sophisticated consumers actually prefer the sweeter wines, and that this preference is especially pronounced for people whose palates are more sensitive—in other words whose palates may be more affected by strong sensations such as astringency.** 

Here is a group of French people who agree with him!

* Mara V. Galmarini, Anne-Laure Loiseau, Michel Visalli, and Pascal Schlich. Use of Multi-Intake Temporal Dominance of Sensations (TDS) to Evaluate the Influence of Cheese on Wine Perception. Journal of Food Science Vol. 81, Nr. 10, 2016. doi: 10.1111/1750-3841.13500.