Tannin quantification in red grapes and wine: comparison of polysaccharide- and protein-based tannin precipitation techniques and their ability to model wine astringency.

Quantification of red grape tannin and red wine tannin using the methyl cellulose precipitable (MCP) tannin assay and the Adams-Harbertson (A-H) tannin assay were investigated. The study allowed for direct comparison between the repeatability of the assays and for the assessment of other practical considerations such as time efficiency, ease of practice, and throughput, and assessed the relationships between tannin quantification by both analytical techniques. A strong correlation between the two analytical techniques was observed when quantifying grape tannin (r(2) = 0.96), and a good correlation was observed for wine tannins (r(2) = 0.80). However, significant differences in the reported tannin values for the analytical techniques were observed (approximately 3-fold). To explore potential reasons for the difference, investigations were undertaken to determine how several variables influenced the final tannin quantification for both assays. These variables included differences in the amount of tannin precipitated (monitored by HPLC), assay matrix variables, and the monomers used to report the final values. The relationship between tannin quantification and wine astringency was assessed for the MCP and A-H tannin assays, and both showed strong correlations with perceived wine astringency (r(2) = 0.83 and r(2) = 0.90, respectively). The work described here gives guidance to those wanting to understand how the values between the two assays relate; however, a conclusive explanation for the differences in values between the MCP and A-H tannin assays remains unclear, and further work in this area is required.