Wouter J. C. de Bruijn

Separation and identification of (iso)flavonoids, including their isomers, is of great interest due to their differential bioactivity. Yet, identification of substituted (iso)flavonoids and differentiation of their isomers in complex plant extracts by LC-MSⁿ is complicated due to the large diversity of these compounds, along with the subtle differences in their structures that result in highly similar retention times and fragmentation patterns. In this study, a cyclic ion mobility-MS (cIMS-MS) method was developed to strengthen the identification of prenylated and glycosylated (iso)flavonoids using collisional cross section (^cIMSCCS_N2) values. Among the ions investigated, the lithium adducts were most stable and offered high-resolution separation. We created a library containing the ^cIMSCCS_N2 values of lithium adducts of 51 (iso)flavonoids. For 47 of the 50 isomer pairs, the ^cIMSCCS_N2 values of the two isomers within the pair were statistically significantly different, confirming that the ^cIMSCCS_N2 values in the library can be used in combination with retention times, m/z values, and fragmentation patterns to achieve identification of these compounds. Furthermore, pre-cIMS trap fragmentation of (iso)flavonoid glycosides can be used to determine ^cIMSCCS_N2 values of the corresponding aglycones for their identification. In addition, a modeling approach was explored, revealing the potential of integrating molecular dynamics simulations for CCS prediction and identification of unknowns. The combined LC-MSⁿ and cIMS-MS approach was able to strengthen identification of several prenylated and glycosylated (iso)flavonoids in a proof of concept analysis of a licorice root extract, thereby demonstrating the power of this approach for the identification of substituted (iso)flavonoids in complex extracts.

Trends in clean label food and the growing popularity of alcohol-free beers have led to a demand for natural compounds to prevent microbial spoilage. Oxidative coupling products of hydroxycinnamic acids and their amides, i.e. phenolamides, obtained via chemo-enzymatic synthesis and from barley rootlets, a beer brewing byproduct, were tested for their inhibitory activity against the beer spoilage yeast Saccharomyces cerevisiae subsp. diastaticus. Hydroxycinnamic acids, their dimers, and hydroxycinnamoylagmatines (HCAgms) did not inhibit the yeast. However, inhibitory activity was observed with HCAgm dimers. In particular, CouAgm-4-O-7'/3-8'-DCouAgm and FerAgm-4-O-7'/3-8'-DFerAgm were active at 46-255 μg/mL in beer. Notably, the newly synthesized agmatine amide of poacic acid, a stilbenoid dimer of ferulic acid, showed increased antiyeast activity compared to its monomeric precursors. These results demonstrate that amidation and oxidative coupling can increase the antiyeast activity of hydroxycinnamic acid derivatives, and that hydroxycinnamoylagmatine dimers possess potential as preservatives against wild yeast spoilage in beer.

The characterization of isomeric oligosaccharides, such as galactooligosaccharides (GOS), in complex mixtures is a long-standing analytical challenge. Due to the exponential increase in the number of possible structures with an increasing degree of polymerization (DP), it is difficult to obtain a complete set of standards. Cyclic ion mobility mass spectrometry (cIMS-MS) has the potential to overcome this challenge by comparing oligosaccharide fragment ions with their respective disaccharide standards, thus reducing the reliance on higher DP oligosaccharide standards. To achieve this, effective separation and recognition of disaccharide fragments using cIMS is a primary prerequisite. Therefore, we studied the ability of combined cIMS separation and collision-induced dissociation (CID) fragmentation to identify isomeric disaccharides consisting of galactose (Gal) and glucose (Glc) with varying compositions and linkage types. Comparison of the drift time selectivity and fragment yields of lithium, sodium, potassium, rubidium, and cesium adducts revealed that lithium overall performed best, by providing good cIMS resolution and far superior fragment yields compared to the other metal ions. cIMS drift times were influenced primarily by the composition of disaccharides and consistently followed the order Gal-Gal < Gal-Glc < Glc-Glc for disaccharides with the same linkage type. CID cross-ring fragmentation patterns were linkage-type specific and did not strongly depend on composition or anomeric configuration. Overall, unknown disaccharides can straightforwardly be identified using our cIMS-MS/MS approach, as monosaccharide compositions can be deduced from cIMS drift times and linkage types from CID fragmentation patterns. These findings are a stepping stone for the identification of larger oligosaccharides by cIMS-MS-based approaches.

Prenylated phenolics are plant-derived compounds with antimicrobial activity against methicillin-resistant Staphylococcus aureus (MRSA), acting by targeting membranes resulting in fast permeabilization. Studies quantifying their membrane permeabilization capacity are lacking, limiting our understanding of the structural properties driving this effect. This study evaluated antimicrobial activity and permeabilization efficacy of 36 C- and O-prenylated phenolics, including 11 C- and O-prenylated phenolics chemically synthesized for this study. Minimum inhibitory concentrations (MICs) were obtained using the broth microdilution assay. Membrane permeabilization was measured by propidium iodide uptake using fluorescence spectrometry and microscopy. The most potent MRSA permeabilizers were luteone (29) and neobavaisoflavone (22), with EC₁₀ of 27 ± 7 and 28 ± 8 μg mL^-1, respectively. Diprenylated phenolics showed a strong negative correlation between permeabilization and their hydrophobic-to-polar surface area ratio (r_pearson = 0.88). For monoprenylated phenolics, prenyl configuration (chain) and molecular shape (globular) were important for effective permeabilization. Interestingly, potency of antimicrobial prenylated phenolics (MIC ≤ 50 μg mL^-1) was not correlated to permeabilization potency, suggesting other mechanisms of action in addition to membrane permeabilization. These quantitative findings on membrane permeabilization by prenylated phenolics contribute to our mechanistic understanding of how these compounds can inhibit microbial growth.