Uncovering the fragmentation and separation characteristics of sophorolipid biosurfactants with LC-MS-ESI.

The application of liquid chromatography and mass spectrometry (MS) is a challenging area of research for structural identification of sophorolipids, owing to the large number of possible variations in structure and limited knowledge on the separation and fragmentation characteristics of the variants. The aims of this work was to provide a comprehensive analysis of the expected characteristics and fragmentation patterns of a wide range of sophorolipid biosurfactant congeners, providing a methodology and process alongside freely available data to inform and enable future research of commercial or novel sophorolipids. Samples of acidic and lactonic sophorolipid standards were tested using reverse-phase ultra-high performance liquid chromatography and identified using electrospray ionization MS. 37 sophorolipid variants were identified and compared for their elution order and fragmentation pattern under MS/MS. The retention time of sophorolipids was increased by the presence of lactonization, unsaturation, chain length, and acetylation as hydrophobic interactions with the C18 stationary phase increased. A key finding that acidic forms can elute later than lactonic variants was obtained when the fatty acid length and unsaturation and acetylation are altered, in contradiction to previous literature statements. Fragmentation pathways were determined for lactonic and acidic variants under negative [M-H]- and positive [M+NH4]+ ionization, and unique patterns/pathways were identified to help determine the structural components present. The first publicly available database of chromatograms and MS2 spectra has been made available to aid in the identification of sophorolipid components and provide a reliable dataset to accelerate future research into novel sophorolipids and shorten the time to innovation. ONE-SENTENCE SUMMARY: This article describes the process and challenges in identifying different structures of eco-friendly biosurfactants, providing a novel database to compare results.

Determining the accuracy and suitability of common analytical techniques for sophorolipid biosurfactants.

To determine the performance of a sophorolipid biosurfactant production process, it is important to have accurate and specific analytical techniques in place. Among the most popular are the anthrone assay, gravimetric quantification (hexane:ethyl acetate extraction), and high-performance liquid chromatography (HPLC). The choice of analytical tool varies depending on cost, availability, and ease of use; however, these techniques have never been compared directly against one another. In this work, 75 fermentation broths with varying product/substrate concentrations were comprehensively tested with the 3 techniques and compared. HPLC-ultraviolet detection (198 nm) was capable of quantifying C18:1 subterminal hydroxyl diacetylated lactonic sophorolipid down to a lower limit of 0.3 g/L with low variability (<3.21%). Gravimetric quantification of the broths following liquid:liquid extraction with hexane and ethyl acetate showed some linearity (R2 = .658) when compared to HPLC but could not quantify lower than 11.06 g/L, even when no sophorolipids were detected in the sample, highlighting the non-specificity of the method to co-extract non-sophorolipid components in the final gravimetric measure. The anthrone assay showed no linearity (R2 = .129) and was found to cross-react with media components (rapeseed oil, corn steep liquor, glucose), leading to consistent overestimation of sophorolipid concentration. The appearance of poor biomass separation during sample preparation with centrifugation was noted and resolved with a novel sample preparation method with pure ethanol. Extensive analysis and comparisons of the most common sophorolipid quantification techniques are explored and the limitations/advantages are highlighted. The findings provide a guide for scientists to make an informed decision on the suitable quantification tool that meets their needs, exploring all aspects of the analysis process from harvest, sample preparation, and analysis.

Screening of alternative nitrogen sources for sophorolipid production through submerged fermentation using Starmerella bombicola.

To explore a sustainable sophorolipid production, several hydrolysates from agricultural byproducts, such as wheat feed, rapeseed meal, coconut waste and palm waste were used as nitrogen sources. The four hydrolysates overperformed the controls after 168 h of fermentation using Starmerella bombicola ATCC 22214. Wheat feed and coconut waste hydrolysates were the most promising feedstocks presenting a linear relationship between yeast growth and diacetylated lactonic C18:1 production at total nitrogen concentrations below 1.5 g/L (R2 = 0.90 and 0.83, respectively). At 0.31 g/L total nitrogen, wheat feed hydrolysate achieved the highest production, yielding 72.20 ± 1.53 g/L of sophorolipid crude extract and 60.05 ± 0.56 g/L of diacetylated lactonic C18:1 at shake flask scale with productivities of 0.43 and 0.36 g/L/h, respectively. Results were confirmed in a 2-L bioreactor increasing 15 % diacetylated lactonic C18:1 production. Moreover, wheat feed hydrolysate supplemented only with a hydrophobic carbon source was able to produce mainly diacetylated lactonic C18:1 congener (88.5 % wt.), suggesting that the composition of the hydrolysate significantly influences the congeners profile. Overall, this study provides valuable insights into agricultural byproduct hydrolysates as potential nitrogen feedstocks for sophorolipid production and their further application on industrial biotechnology.

Food wastes for bioproduct production and potential strategies for high feedstock variability.

Unavoidable food wastes could be an important feedstock for industrial biotechnology, while their valorization could provide added value for the food processor. However, despite their abundance and low costs, the heterogeneous/mixed nature of these food wastes produced by food processors and consumers leads to a high degree of variability in carbon and nitrogen content, as well as specific substrates, in food waste hydrolysate. This has limited their use for bioproduct synthesis. These wastes are often instead used in anaerobic digestion and mixed microbial culture, creating a significant knowledge gap in their use for higher value biochemical production via pure and single microbial culture. To directly investigate this knowledge gap, various waste streams produced by a single food processor were enzymatically hydrolyzed and characterized, and the degree of variability with regard to substrates, carbon, and nitrogen was quantified. The impact of hydrolysate variability on the viability and performance of polyhydroxyalkanoates biopolymers production using bacteria (Cupriavidus necator) and archaea (Haloferax mediterranei) as well as sophorolipids biosurfactants production with the yeast (Starmerella bombicola) was then elucidated at laboratory-scale. After which, strategies implemented during this experimental proof-of-concept study, and beyond, for improved industrial-scale valorization which addresses the high variability of food waste hydrolysate were discussed in-depth, including media standardization and high non-selective microbial organisms growth-associated product synthesis. The insights provided would be beneficial for future endeavors aiming to utilize food wastes as feedstocks for industrial biotechnology.

Developing an understanding of sophorolipid synthesis through application of a central composite design model.

A key barrier to market penetration for sophorolipid biosurfactants is the ability to improve productivity and utilize alternative feedstocks to reduce the cost of production. To do this, a suitable screening tool is required that is able to model the interactions between media components and alter conditions to maximize productivity. In the following work, a central composite design is applied to analyse the effects of altering glucose, rapeseed oil, corn steep liquor and ammonium sulphate concentrations on sophorolipid production with Starmerella bombicola ATCC 222144 after 168 h. Sophorolipid production was analysed using standard least squares regression and the findings related to the growth (OD600 ) and broth conditions (glucose, glycerol and oil concentration). An optimum media composition was found that was capable of producing 39.5 g l-1 sophorolipid. Nitrogen and rapeseed oil sources were found to be significant, linked to their role in growth and substrate supply respectively. Glucose did not demonstrate a significant effect on production despite its importance to biosynthesis and its depletion in the broth within 96 h, instead being replaced by glycerol (via triglyceride breakdown) as the hydrophilic carbon source at the point of glucose depletion. A large dataset was obtained, and a regression model with applications towards substrate screening and process optimisation developed.