Green coffee processing has been hindered by low oil extraction yields from mechanical pressing and the need of using flammable and hazardous solvents for defatting the protein-rich cake before subsequent protein extraction. To replace the use of flammable solvents and enable the simultaneous extraction of lipids and proteins from green coffee beans at reduced water usage, a multistage countercurrent extraction process was scaled up from 0.05 to 1.14 kg and evaluated regarding protein and oil extractability, physicochemical and functional properties of the extracted protein, and oil recovery. Enzymatic extraction increased protein extractability by ~13% while achieving similar oil extractability when not using enzymes (55%). Proteolysis resulted in the release of smaller proteins with reduced surface hydrophobicity and higher solubility at acidic pH (3.0–5.0). The physicochemical changes observed due to proteolysis resulted in the formation of emulsions with reduced resistance against enzymatic and chemical demulsification strategies, enhancing the recovery of the extracted oil (48.6–51.0%). Proteolysis did not alter the high in vitro digestibility of green coffee proteins (up to 99%) or their emulsifying properties at most pH values evaluated. However, proteolysis did reduce the foaming properties of the hydrolysates compared with larger molecular weight proteins. These findings revealed the impact of extraction conditions on the extractability and structural modifications altering the functionality of green coffee proteins and the synergistic impact of extraction and demulsification strategies on the recovery of the extracted oil, paving the way for the development of structure–function processes to effectively produce green coffee proteins with desired functionality.

Reverse-phase solid-phase extraction (SPE) is regularly used for separating and purifying food-derived oligosaccharides and peptides prior to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. However, the diversity in physicochemical properties of peptides may prevent the complete separation of the two types of analytes. Peptides present in the oligosaccharide fraction not only interfere with glycomics analysis but also escape peptidomics analysis. This work evaluated different SPE approaches for improving LC-MS/MS analysis of both oligosaccharides and peptides through testing on peptide standards and a food sample of commercial interest (proteolyzed almond extract). Compared with conventional reverse-phase SPE, mixed-mode SPE (reverse-phase/strong cation exchange) was more effective in retaining small/hydrophilic peptides and capturing them in the high-organic fraction and thus allowed the identification of more oligosaccharides and dipeptides in the proteolyzed almond extract, with satisfactory MS/MS confirmation. Overall, mixed-mode SPE emerged as the ideal method for simultaneously improving the identification of food-derived oligosaccharides and small peptides using LC-MS/MS analysis.

The production of green coffee oil by mechanical pressing of green coffee beans has been precluded by low extraction yields, which generates a protein-rich byproduct (cake) containing variable amounts of lipids. Subsequent utilization of the cake requires the removal of the residual cake oil by solvent extraction. An eco-friendly extraction strategy, using water, enzymes, and mechanical treatments, was evaluated to concurrently extract lipids and proteins from green coffee flour, without the use of harsh solvents. Among the enzymatic treatments evaluated, the use of 0.5% alkaline protease led to higher protein (62.2%) and oil (47.7%) extractability in a shorter time (30 min). This enzymatic treatment was optimized with respect to solids-to-liquid ratio (SLR) (1:17.5-1:7) and concentration of enzyme (0.1-0.9% w/w). Although optimum extraction conditions (1:17.5 SLR and 0.1% enzyme) achieved high protein (70%) and oil (48%) extractability and reduced enzyme use by 80%, a higher water usage was required. Therefore, a two-stage countercurrent extraction was developed to reduce water usage in the process. The countercurrent extraction strategy not only reduced the amount of water used in the process by 60% but promoted higher protein (72%) and oil (58%) extractability, compared with the single-stage process (62.2 and 47.7%, respectively).

Infant milk formula was used as a model food to compare the sensitivity of thiobarbituric acid reactive substances (TBARS) and hydroperoxide methods to UPLC-MS/MS oxylipin analysis for detecting early lipid oxidation. Two different infant milk formulas were tested during 21 days of storage at 4°C. Formulas 1 and 2 contained canola oil and canola oil + 1% docosahexaenoic acid ethyl ester, respectively. Formulas were sampled up to 21 days of storage. Formula 2 had higher peroxide values than Formula 1 across all time points. However, no significant differences over time in TBARS and peroxide values in either formula were observed. Several oxylipins increased in both formulas starting on day 7 (linoleic acid and alpha-linolenic acid-derived oxylipins in Formula 1 and DHA-derived oxylipins in Formula 2). These results indicate that free oxylipins are effective in detecting early lipid oxidation and distinguishing between formulations containing different fatty acids. PRACTICAL APPLICATION: We have recently shown that primary oxidation products known as oxylipins can be measured in their free form by ultra-high pressure liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) to detect early lipid oxidation. However, a head-to-head comparison of the sensitivity of this approach to conventional spectrophotometric methods has not been evaluated. Our results indicate that free oxylipin measurements are better than conventional methods in detecting early lipid oxidation in milk infant formula  distinguishing between different formulations.