Growing & Winemaking
Understanding Congeners in Wine
Congeners (Latin for “born together”) live up to their name. They’re the inevitable but often neglected siblings of an alcoholic fermentation. As yeasts ferment the sugars from must or any other source to ethanol and carbon dioxide, they also deliver an array of other compounds, appropriately called congeners. They’re yeast’s supplement to the flavor complexity of wine that complements the contributions of the grapes themselves.
Congeners exert their greatest impact on some spirits because by distillation they become concentrated in them. But beers and wines are also affected, if to a lesser extent. However, one congener, active amyl alcohol (see below) and the esters derived from it, are considered to be important, desirable flavor components of beer.
What are congeners?
The list of congeners, which includes acetaldehyde and a variety of esters (particularly the ethyl esters of C8 to C12 fatty acids), is extensive. But the group collectively called fusel oil is the most abundant and might be the most intriguing. Certainly it’s the congener with the most ambiguous reputation. Fusel oil, long blamed for hangovers, is recognized for adding flavor complexity. Fine traditional brandies, for example, would be drab without fusel oil.
Fusel oils are a mixture of higher (greater molecular weight than ethanol) alcohols. The odd name, fusel, comes from an old German word, which roughly translates as “bad spirits,” chosen, undoubtedly for the somewhat unpleasant smell of a concentrated mixture of them.
Fusel oil is called “oil” because in the process of distilling alcoholic beverages, the mixture separates as an upper oily layer on the plates of a continuous still containing 100° to 135° proof. Fusel oil has four major components: a C3 alcohol (normal propyl alcohol), a C4 alcohol (isobutyl alcohol) and two C5 alcohols (isoamyl and active amyl alcohols).
Fermenting yeast always make these four major fusel-oil components, along with smaller amounts of other higher alcohols. The not-completely answered question is, why? Their formation is not part of the alcoholic fermentation by which yeast derive metabolic energy. Yeast don’t benefit energetically by making fusel oil, nor for the most part in any other way.
How is fusel oil made?
More than 100 years ago (in 1907), the origin and reason for fusel oil seemed abundantly clear. That year, F. Ehrlich published a paper (later confirmed by others) showing that fusel oil components are metabolic detritus. They’re the slightly altered form of the carbon skeletons of certain amino acids, discarded after yeast have stripped off the nitrogen atoms that they need for growth. Small amounts of these amino acid raw materials are present in grape juice.
Except for normal propyl alcohol, all the major components of fusel oil—isobutyl, isoamyl and active amyl alcohols—contain the carbon skeletons of a biosynthetically related group of amino acids (sometimes called the branched-chain amino acids): valine, leucine and isoleucine, respectively. This explanation for why yeast makes fusel oil is still widely held. Curiously, only yeasts—a wide variety of them—are known to make fusel oil.
But Ehrlich’s explanation was incomplete, and fundamental questions remained: Why are these few amino acids preferentially attacked? And the greater puzzle is: How and why does yeast make fusel oil, even when amino acids are not available as a source of nitrogen?
This dilemma was brought into sharp focus by a detailed experiment that John Castor and Jim Guymon did more than 50 years ago at the University of California, Davis, Department of Viticulture and Enology. They followed the disappearance of branched-chain amino acids and the formation of fusel oil (difficult and time-consuming measurements to make in those days) during the fermentation of a French Colombard must by the Montrachet strain of yeast.
Castor and Guymon’s results shattered Ehrlich’s implied one-to-one link between amino acid utilization and fusel oil formation. They found that fusel oil formation continued—in fact it speeded up—after all the branched-chain amino acids in the must had been depleted. And the formation of fusel oil (along with fermentation) continued even after growth of the yeast (and its need for amino acids) stopped. Later it was shown that yeast cells suspended in a solution of glucose alone in the complete absence of amino acids make fusel oil as they ferment.
How yeast does this was answered by Jim Guymon, Ed Crowell and me, also at UC Davis, in the early 1960s. In a series of papers, we showed that the major fusel oil components are synthesized along the same metabolic route as their corresponding amino acids. But rather than proceeding all the way to amino acids, the route branches off to make fusel oil. We showed this using strains of yeast that as a result of mutation had lost the ability to synthesize a particular amino acid. Such strains do not make the corre
sponding fusel oil component. For example, mutant strains that are unable to synthesize leucine don’t make isoamyl alcohol, and strains that can’t synthesize isoleucine don’t make active amyl alcohol.
(Jim Guymon, professor of enology, brandy specialist and connoisseur, had a remarkable nose for fusel oil. I saw him—by tasting—correctly rank three Zinfandel wines according to their content of fusel oil.)
In these same studies we discovered the metabolic route of formation of normal propyl alcohol, the fusel oil component that does not correspond directly to any amino acid. Its route of synthesis, too, is integral to that of the branched chain amino acids, but in a curious and unexpected way. An intermediate (alpha-oxoglutarate) in the series of metabolic reactions leading to the synthesis of isoleucine is made into normal propyl alcohol by the same route by which the skeletons of branched-chain are discarded. Why? What possible benefit does its formation or presence offer the yeast? That’s a mystery, but normal propyl alcohol seems always to be a component of fusel oil.
Controlling formation of fusel oil
These results suggested that it might be possible to construct a strain of yeast that would produce little or, perhaps, no fusel oil at all: Simply introduce into the strain sufficient genetic blocks to render it incapable of making any branched-chain amino acid. We did this, and the results were quite surprising. The strain, as expected, did not make any of the usual components of fusel oil, but, unexpectedly, it made a new higher alcohol not normally present in fusel oil: normal butyl alcohol.
We found that the strain had cobbled together remaining bits and pieces of the reaction series that normally make branched-chain amino acids and their corresponding fusel oil components to make this new fusel oil—like higher alcohol, normal butyl alcohol. It made alpha oxobutyrate by a remaining fragment of the isoleucine pathway, and converted this by fragments of leucine and Ehrlich pathways into normal butyl alcohol. It seems as though yeast have a compulsion to make some fusel oil. Even mutant strains make it one way or another by their remaining metabolic tools.
Later, Richard Snow and Ralph Kunkee adapted this approach using the Montrachet variety of Saccharomyces cereviseae to obtain a commercially useful strain that makes only minimal amounts of isoamyl alcohol, the fusel oil component that some consider the most undesirable for wine flavor. And we had found another way to limit fusel oil. Aeration stimulates the formation of all fusel oil components; strict exclusion of air from a fermentation reduces the amount of them.
Beyond wine
It’s an intriguing fact, as first so dramatically shown by Louis Pasteur, the founder of modern microbiology, in Études Sur le Vin that the consequences of research on wine sometimes extend beyond wine itself. Research on the formation of fusel oil in wine is a meaningful example.
The biotechnology industry, which uses microorganisms to make therapeutically useful proteins such as insulin and human growth hormone, was distressed to discover that these proteins sometimes contained an abnormal amino acid, norvaline (it’s not in our genetic code nor normally present in protein), and they wanted to eliminate it. They found that norvaline was synthesized via the route by which we found that mutant strains of yeast make the abnormal fusel oil component, normal butyl alcohol. This information led to means of eliminating norvaline from their products.
Knowledge about how fusel oil is made is now being applied to the manufacture of synthetic fuels. Fusel oil genes from yeast have been transferred by Shota Atsumi to other microbes (notably Escherichia coli) to produce normal butyl alcohol, which is an excellent fuel for internal combustion engines.
What to do?
We now know how fusel oil, the major congener, is made. We still have no idea why yeasts, nearly all of them, make it. We know it affects the taste of wine, and we have some knowledge about how to manage its formation. But we are not sure that we want to.
Upon retirement from the University of California, Davis, as professor of microbiology, John L. Ingraham took up consultation in biotechnology and writing. His latest book, March of the Microbes, Sighting the Unseen, which takes a birdwatcher’s approach to microbiology, was released by Harvard University Press in February 2010. To comment on this article, e-mail edit@winesandvines.com.
