O-18
Comparison of methods for assessing polyphenols in beer
Presenter: Karl Siebert, Dept. of Food Science & Technology, Cornell University, Geneva, NY, USA
Co-Author: P.Y. Lynn, Dept. of Food Science & Technology, Cornell University, Geneva, NY, USA

There are numerous methods for determining polyphenols in beer. Many of these have biases, suffer interferences, or give differing responses depending on the nature of the protein or polyphenol molecules in a sample. Several prominent classes of beer polyphenols have been shown to have quite different affinities for proteins. These differences determine both the extent to which the polyphenols participate in forming hazes and sediments and their responses in a number of analytical methods. The beer haze-active (HA) polyphenols are mainly the proanthocyanidin "dimers" procyanidin B3 and prodelphinidin B3. HA polyphenols have at least two sites that can attach to HA proteins and thus bridge them together. Some of these bind more strongly to proteins than other polyphenols and have been demonstrated to be more haze-active. Some polyphenols have only a single binding site; they can attach to HA protein but do not lead to haze. Much larger amounts of less-HA and non-HA polyphenols are normally present in beer than the HA polyphenols. The former include catechin and epicatechin. Most polyphenol determination methods (e.g., the International Method for Total Polyphenols or the Folin-Ciocalteu procedure) measure total polyphenols, with no differentiation between those that are HA and those that show little or no haze-forming activity. At least one method, induction of haze with polyproline (the HA polyphenol method), focuses on polyphenols able to form haze, but it suffers from poor reproducibility and sensitivity and also suffers interference from endogenous HA protein. HPLC with electrochemical detection can individually determine HA, non-HA, and weakly HA polyphenols but requires equipment of considerable sophistication and expensive standards. Different beers contain differing amounts of HA and non-HA protein as well as HA and non-HA polyphenols. It was of interest to systematically compare the responses of a number of polyphenol-determination approaches on samples containing combinations of HA and non-HA proteins and HA and non-HA polyphenols. Mixtures containing various combinations of HA and non-HA proteins (water-soluble hordein and lysozyme, respectively) and HA and non-HA polyphenols (tannic acid and epicatechin) were prepared in buffer model systems. A battery of polyphenol methods (the International Total Polyphenol method, the Folin-Ciocalteu method, polyproline haze induction, and HPLC) was applied to each sample, and the results were compared. The different methods gave quite different responses to the different test compounds. Results obtained when these methods are applied to beer must be interpreted with caution.

Karl Siebert received a Ph.D. in biochemistry from Penn State in 1970. He joined the Stroh Brewery Company in Detroit, where he spent 18 years and held positions from research associate to director of research. In 1990, Dr. Siebert joined Cornell University as professor of biochemistry in the Department of Food Science and Technology. He served five years as department chairman and now has a predominantly research appointment. Dr. Siebert served on ASBC technical subcommittees and was a member and chairman of the Technical Committee. He is serving his second stint on the JASBC editorial board (1980–1992; 1996–). He is active as a consultant in the beverage industry.