Integration of network-based approaches for assessing variations in metabolic profiles of alkalized and non-alkalized commercial cocoa powders

Abstract

Cocoa can undergo an alkalization process to enhance its color and solubility. It reduces astringency and alters its composition, particularly in the phenolic compound content, which is related to cocoa health benefits. This study aimed to investigate the impact of alkalization on the composition of seven commercial cocoa powders. A liquid chromatography-based metabolomic approach was employed to assess the metabolic differences between alkalized and non-alkalized cocoa powders. Supervised orthogonal partial least squares discriminant analysis (OPLS-DA) was used to identify the most discriminating variables between groups. A feature-based molecular network (FBMN) was used to explore the chemical space. Three hundred forty-seven metabolites were obtained as the most discriminant, among which 60 were tentatively annotated. Phenolic compounds, lysophosphatidylethanolamines, amino acids, and their derivatives were significantly reduced in alkalized cocoas. In contrast, fatty acids and their derivatives significantly increased with alkalization. Despite the variability among commercial cocoas, chemometrics allowed the elucidation of alterations induced specifically by alkalization in their composition.

Keywords

• Cocoa
• Alkalization
• Metabolomics
• Molecular network
• Chemometrics