High-Throughput Quantitation of Key Cocoa Tastants by Means of Ultra-High-Performance Liquid Chromatography Tandem Mass Spectrometry and Application to a Global Sample Set.

Abstract

Historically often described as the food of gods, cocoa-based products exhibit a pleasant aroma as well as a desirable astringent, bitter, and sour taste, which results in a high consumer preference. The key taste components of cocoa were identified and characterized by combining sensory analysis, fractionation, and structure elucidation. Cocoa astringency is driven by N-phenylpropenoyl-l-amino acids, polyphenol glycosides, and flavan-3-ols, while the latter compound class also contributes to bitterness. The key principle for cocoa bitterness was shown to be the combination of alkaloids and 2,5-diketopiperazines. To understand the influence of plant genetics, breeding, and processing on the sensory profile of cocoa products, high-throughput sensometabolite quantitation must be performed throughout all of these steps. In this work, we present a rapid, sensitive, and robust quantitation method on a single ultra-high-performance liquid chromatography tandem mass spectrometry (UHPLC-MS/MS) platform, requiring minimal workup for any sample type from farm to fork. This method was applied to a global set of 75 cocoa bean samples from all over the world before and after using a uniform roasting protocol. Within this world map, geographical origin did not predetermine cocoa taste profiles, whereas simulated processing by roasting was confirmed to be crucial in profile development. This method will open avenues for further studies to ultimately enable chocolate producers to control and optimize the taste properties of products as well as to monitor raw material selection and processing.