A blog devoted to exploring wines made from unusual grape varieties and/or grown in unfamiliar regions all over the world. All wines are purchased by me from shops in the Boston metro area or directly from wineries that I have visited. If a reviewed bottle is a free sample, that fact is acknowledged prior to the bottle's review. I do not receive any compensation from any of the wineries, wine shops or companies that I mention on the blog.

Wednesday, April 25, 2012

Favorita and DNA Microsatellite Analysis - La Morra, Piemonte, Italy

What is one to do with a grape like Favorita?  According to the Italian wine authorities, Favorita is a distinct grape variety, but according to the scientists, it is genetically identical to both Vermentino and Pigato.  What are those of us who are fanatical about classification to do with Favorita, Pigato and Vermentino?

Those of you who are familiar with this blog are aware that this is an issue that has caused me to spill a considerable amount of ink.  My most complete treatment of the issue can be found in the Roter Traminer article, though I do touch on the issue in my Malvasia Nera post as well as my Pigato post.  The question I'm wrestling with in all of those posts concerns the point at which a clonal variant becomes differentiated enough to be considered a separate grape variety.  Where do we draw that line?  How different is different enough to be considered separate?

Believe it or not, this question was actually much easier to answer before the arrival and widespread use of DNA testing for grapes.  Ampelography was (and technically still is, I suppose, though there are precious few ampelographers in the traditional sense of the word anymore) the science of grapevine identification and differentiation using the physical characteristics of the vine itself.  Leaf shape and size was perhaps the most widely used measure for ampelographers, but other factors such as grape size and color and bunch size and shape were used not only to identify unknown vines in the vineyard but also to differentiate grape varieties from one another.  This was not a perfect system by any means and there have been some fairly famous gaffes made by ampelographers in the past (like the mistaking of Carmenere with Merlot or of Sauvignon Blanc with Friulano in Chile, for example, though to be fair both of those cases were also eventually sorted out by ampelographers), but there's a certain kind of elegance and intuitive sense to classifying plants this way.

The areas where ampelography has been found the most wanting, especially in recent years, concern those cases of vines that are likely clonally descended from a common ancestor, but which have been spread out geographically over a long period of time and have developed mutations along the way. All of the mutations that happen to grapevines which have an effect on the physical appearance of the plant are called somatic mutations.  These mutations can affect everything from leaf shape to bunch size to grape size.  Some are subtle while others are dramatic, though most of the dramatic mutations probably started out subtle and were exaggerated over time through clonal selection.  The most famous somatic mutations involve Pinot Noir and its tendency to mutate into vines that produce white berries, which are known as Pinot Blanc, or pink berries, which are known as Pinot Gris.  These somatic mutations cannot be picked up with current DNA analysis techniques because those techniques do not analyze the entire DNA string of a grape.  I've mentioned DNA analysis enough on this site that I really should take some time to try to explain exactly what it is and how it works so buckle up because there's some serious science ahead.

As I'm sure you already know, DNA is one of the basic building blocks of life.  It is composed of two strands of molecules which run parallel (technically anti-parallel, but let's not quibble) to one another in a twisting design known as the double helix.  These strands are made up of four different nucleobases, each of which has its own unique partner on the opposite strand that it is connected to by sugar or phosphate molecules.  The nucleobases are Guanine (G) which pairs with Cytosine (C) and Adenine (A) which pairs with Thymine (T).   Some of the arrangements of nucleobases are involved in synthesizing proteins, and these areas are known as genes, but some of the sections of DNA are just filler or what I like to call Junk DNA.  It turns out that there are a lot of Junk DNA sites throughout the entire DNA sequence of an organism, but the ones that we're interested in are called microsatellite sequences. Microsatellite sequences are repetitions of between one and six nucelobases that are scattered randomly throughout an organism's DNA.  So a typical bit of DNA might look something like this:


That bit in the middle where AG repeats is a microsatellite site, and it turns out that there are between 1,000 and 10,000 of these sites scattered throughout the genomes of most complex multi-cellular organisms (not all of them are repeats of just two nucleobases...it could just be one [AAAAAAAA] or as many as six [GATCACGATCACGATCACGATCAC]).  It also turns out that some of these sites are highly variable between individuals and are inherited in very specific ways.  What scientists have done is to find where these sites are in humans as well as in grapes and they have pinpointed the areas that are the most variable between individuals.  These sites are given names like VVS1, ZAG64, and VVMD28 among others and they are the regions that scientists look at when they are trying to differentiate between different grape varieties or confirm synonymities between differently named grapes that are genetically identical.  All individuals have microsatellite sites in the same places on their genome, but the size of the site will be different between individuals, so the number of base pairs in the repeat are counted up and compared across subjects to see which are identical and which are different (this is not very important in humans because humans are obviously distinctive from one another and are not propagated clonally as grapes are...see my Ciliegiolo post for a discussion of this issue).  Most studies use between 6 and 10 different microsatellite sites in their analyses in order to differentiate between individual varieties because fewer sites don't give enough discriminatory power while more sites don't give you any added benefit.  Parentage studies use the same technique, but look at different microsatellite sites which have been found to be directly inherited from each parent.  This is the same technology, incidentally, used in paternity tests in humans and made famous by such trash-talk heavyweights as Jerry Springer.

It's important to remember that these microsatellite sites are Junk DNA that are essentially just taking up space.  They don't encode for any proteins and don't have any effect on the physical appearance of the plant.  Remember for a moment what we said about somatic mutations: they are mutations that have an effect on the physical appearance of the plant, which means that they are mutations that occur in areas of the plant's DNA that are actually doing some kind of work.  The kinds of mutations that affect berry skin color are happening in areas of the plant's DNA outside of the areas that microsatellite site analysis is looking at, so these mutations are essentially undetectable with this kind of technology.  This can be a good thing, as it allows us to trace the clonal history of grapes like Pigato, Favorita and Vermentino, which are geographically dispersed, to a single grape ancestor, but it can also possibly lead us to overly reductionistic conclusions.  Pigato, Favorita and Vermentino are genetically identical according to the microsatellite analysis, but does that mean they're the same grape?

The case of Pinot Noir is significant here, I think.  Pinot Noir, Pinot Blanc and Pinot Gris are genetically identical, but are considered to be different grape varieties by virtually every expert on earth, while grapes like Nielluccio, Sangiovese and Prugnolo Gentile are considered to be clonal variants of the same grape, Sangiovese.  The difference in the two cases is obvious: the Pinot mutation has an obvious physical manifestation, namely the color of the grape skin.  But it's more than just that.  The color of the grape skin has a direct effect not only on the wines that are made from that grape, but also in the kinds of wines that can be made from it and the methods that one must employ to make those wines.  The physiological difference entails a different approach to making wines from that grape, while the physiological differences between the Sangiovese clones are much more subtle.  One could mistake wines made from the Sangiovese clones with one another while it would be much more difficult to make that mistake with wines made from the Pinot varieties.  

The situation with Pigato, Favorita and Vermentino is, unfortunately, not so clear cut.  There are obvious physical differences: Pigato has a characteristic splotchiness on the grape skins that Favorita and Vermentino lack while the leaves, clusters and buds of Favorita are different enough from those of Vermentino that the Italian Ministry of Agriculture decided that it was a separate grape in 1964, stating unequivocally that "Favorita and Vermentino must be considered as two distinct cultivars," but given the new information that DNA analysis has provided, can we still say the same thing today with the same amount of confidence?  The splotchiness of the skins of Pigato has no real effect on the wines made from the grape or on the range of wines that could possibly be made from it, and the physiological differences between Favorita and Vermentino seem to be merely cosmetic.  One could argue that wines made from each of the three grapes are distinctive from one another, but one would expect that anyway given the range of places and climates where each is primarily found (Pigato in Liguria, Vermentino in Sardinia and Liguria and Favorita in Piemonte).  Wines made from Chablis are much different from those made in Pugliny-Montrachet, for example, but Chardonnay is the grape in both wines.  

So what are we to do?  It would be ideal if the genomes of Pigato, Favorita and Vermentino were completely sequenced so that an analysis could be done to see where the differences are in their DNA and to see just how widespread they are.  We know that they're ultimately descended from the same plant and have mutated independently over time to the extent that they are visually distinguishable from one another, but we just don't know how significant the mutations have been on a molecular level and we can't know unless we have the entire DNA sequence of each grape.  Sequencing an individual grape variety is no small task, though, and I doubt seriously that there's much clamor for that kind of work on any of these three grapes. Given the standards that are currently used to differentiate grape varieties, I don't see any way to consider Pigato, Favorita and Vermentino as separate cultivars.  The differences just don't seem significant enough to me to merit differentiation.  If any readers have any thoughts, please share in the comments section as I'm very interested to see how others might feel about this (if anyone other than me ever thinks about this kind of stuff).

The grape we call Favorita is grown exclusively in Piemonte around the Roero region and is thought to have arrived in this area over 300 years ago via Ligurian oil merchants.  The grape is known for having large berries and has been a popular table grape in Piemonte for many years.  It is thought that moniker "Favorita" was bestowed as a result of its prowess and esteem as a table grape and not as a wine grape.  When it was used for wine, historically it was blended with Nebbiolo in an effort to smooth over some of Nebbiolo's rough edges.  Varietal wines made from Favorita have enjoyed some success recently, but plantings lag well behind those of Arneis and Chardonnay in the Langhe.  Many examples that we find here in the US are slightly spritzy, though I'm not sure how prevalent that style is elsewhere.

I was able to find a bottle of the 2009 Gianni Gagliardo "fallegro" Favorita at Bauer Wine and Spirits for about $17.  In the glass the wine was a medium lemon gold color.  The salesperson at Bauer told me this wine was supposed to be a little fizzy, but mine was completely flat.  The nose was a total blank with maybe a suggestion of lemon or apple, but it was really a total cipher.  This is apparently a characteristic of the grape itself, as Nicolas Belfrage in his Barolo to Valpolicella refers to the grape as "non-aromatic [with] any scented character it may have coming by illicit blending."  On the palate the wine was medium bodied with medium acidity.  It was somewhat cidery with ripe apple and apple peel fruit along with a touch of pear and melon as well.  The wine was bottled with a synthetic cork so I wonder if perhaps there was a faulty seal which may have caused my bottle to go flat and take on a bit of oxygen.  I haven't had a chance to try another Favorita-based wine, so I'm really not sure how representative my bottle is. This bottling is supposedly the benchmark for Favorita wines, but it didn't do much for me.  


Mark Middlebrook said...

Fascinating information on ampelography and DNA analysis - thank you! I've always wondered what it meant to say that two grape varieties are "the same".

2009 Galgiardo Fallegro is an old vintage for that wine. Current vintage probably is 2011. I suspect that's why it tasted so flat. It's a fun, spritzy, uncomplicated wine. One could quibble over the price - it's arguably the kind of wine that should be glugged down at an osteria near where it's produced - but I've enjoyed it and sold it at Paul Marcus Wines in Oakland. I don't recall seeing other exported versions of Favorita, although I suspect that there may be one or two.

Anonymous said...

Hi, I really enjoyed this account and the explanations of the DNA stuff. We had our first taste of Favorita yesterday and today and have found it so refreshingly different to anything else. I have been puzzling over the bouquet - which I thought was citrusy-ginger but thnk might be more realistically citrusy-pear. Will definitely try again.

Terroir definitely affects the vine and hence the wine so it is quite easy to see how the same gentic organism might produce wines of completely different character. If clonal variations are superimposed upon that then it is easy to see how a clonal variety might take on a more distinct identity when it grows in a different terroir.

Fascinating stuff!