Multivariate chemical mapping of antibiotics and identification of structurally representative substances

Pharmaceutical metabolite identification in lettuce (Lactuca sativa) and earthworms (Eisenia fetida) using liquid chromatography coupled to high-resolution mass spectrometry and in silico spectral library

Pharmaceuticals released into the aquatic and soil environments can be absorbed by plants and soil organisms, potentially leading to the formation of unknown metabolites that may negatively affect these organisms or contaminate the food chain. The aim of this study was to identify pharmaceutical metabolites through a triplet approach for metabolite structure prediction (software-based predictions, literature review, and known common metabolic pathways), followed by generating in silico mass spectral libraries and applying various mass spectrometry modes for untargeted LC-qTOF analysis. Therefore, Eisenia fetida and Lactuca sativa were exposed to a pharmaceutical mixture (atenolol, enrofloxacin, erythromycin, ketoprofen, sulfametoxazole, tetracycline) under hydroponic and soil conditions at environmentally relevant concentrations. Samples collected at different time points were extracted using QuEChERS and analyzed with LC-qTOF in data-dependent (DDA) and data-independent (DIA) acquisition modes, applying both positive and negative electrospray ionization. The triplet approach for metabolite structure prediction yielded a total of 3762 pharmaceutical metabolites, and an in silico mass spectral library was created based on these predicted metabolites. This approach resulted in the identification of 26 statistically significant metabolites (p < 0.05), with DDA + and DDA - outperforming DIA modes by successfully detecting 56/67 sample type:metabolite combinations. Lettuce roots had the highest metabolite count (26), followed by leaves (6) and earthworms (2). Despite the lower metabolite count, earthworms showed the highest peak intensities, closely followed by roots, with leaves displaying the lowest intensities. Common metabolic reactions observed included hydroxylation, decarboxylation, acetylation, and glucosidation, with ketoprofen-related metabolites being the most prevalent, totaling 12 distinct metabolites. In conclusion, we developed a high-throughput workflow combining open-source software with LC-HRMS for identifying unknown metabolites across various sample types.

Application of chemometric tools in the QSAR development of VOC removal in plastic waste recycling

Deodorization and, in a broader sense, the removal of volatile organic compounds (VOCs) from plastic waste have become increasingly important in the field of plastic recycling, and various new decontamination techniques have been developed. Both in research and industrial practice, the selection of VOCs has been random or unsubstantiated, making it difficult to compare studies and assess decontamination processes objectively. Thus, this study proposes the use of Statistical Molecular Design (SMD) and Quantitative Structure - Activity Relationship (QSAR) as chemometric tools for the selection of representative VOCs, based on physicochemical properties. Various algorithms are used for SMD; hence, several frequently used D-Optimal Onion Design (DOOD) and Space-Filling (SF) algorithms were assessed. Hereby, it was validated that DOOD, by dividing the layers based on the equal-distance approach without so-called 'Adjacent Layer Bias', results in the most representative selection of VOCs. QSAR models that describe VOC removal by water-based washing of plastic waste as a function of molecular weight, polarizability, dipole moment and Hansen Solubility Parameters Distance were successfully established. An adjusted-R2 value of 0.77 ± 0.09 and a mean absolute error of 24.5 ± 4 % was obtained. Consequently, by measuring a representative selection of VOCs compiled using SMD, the removal of other unanalyzed VOCs was predicted on the basis of the QSAR. Another advantage of the proposed chemometric selection procedure is its flexibility. SMD allows to extend or modify the considered dataset according to the available analytical techniques, and to adjust the considered physicochemical properties according to the intended process.

Tool for selecting indicator substances to evaluate the impact of wastewater treatment plants on receiving water bodies

The elimination of organic micropollutants (OMPs) from wastewater could in future become mandatory for operators of wastewater treatment plants (WWTPs). Indicator substances are a great help and a cost-efficient way in monitoring the pollution of water bodies with OMPs caused by the discharge of WWTPs. However, with the still increasing number of OMPs in our environment, the selection of suitable indicator substances presents a challenge. A concept was developed to help identify representative indicator substances. The derived indicator substances are not only used to assess water pollution, but can also be used to calculate elimination efficiencies of WWTPs. In the present investigations, the indicator substances were used to evaluate the reduction of OMPs in the water body on the basis of the expansion of a WWTP with an ozonation plant. The transferability of the tool was verified with a second WWTP. Furthermore, the impact of the number of measurements was analysed via statistical combinatorics. With the tool, 36 substances were classified, leading to the identification of 9 suggested indicator substances. Among them ibuprofen and diclofenac attracted attention due to their ecotoxicological relevance. Detailed data analyses were carried out using principal component analysis (PCA) and loads.

Selection of representative emerging micropollutants for drinking water treatment studies: a systematic approach

Micropollutants remain of concern in drinking water, and there is a broad interest in the ability of different treatment processes to remove these compounds. To gain a better understanding of treatment effectiveness for structurally diverse compounds and to be cost effective, it is necessary to select a small set of representative micropollutants for experimental studies. Unlike other approaches to-date, in this research micropollutants were systematically selected based solely on their physico-chemical and structural properties that are important in individual water treatment processes. This was accomplished by linking underlying principles of treatment processes such as coagulation/flocculation, oxidation, activated carbon adsorption, and membrane filtration to compound characteristics and corresponding molecular descriptors. A systematic statistical approach not commonly used in water treatment was then applied to a compound pool of 182 micropollutants (identified from the literature) and their relevant calculated molecular descriptors. Principal component analysis (PCA) was used to summarize the information residing in this large dataset. D-optimal onion design was then applied to the PCA results to select structurally representative compounds that could be used in experimental treatment studies. To demonstrate the applicability and flexibility of this selection approach, two sets of 22 representative micropollutants are presented. Compounds in the first set are representative when studying a range of water treatment processes (coagulation/flocculation, oxidation, activated carbon adsorption, and membrane filtration), whereas the second set shows representative compounds for ozonation and advanced oxidation studies. Overall, selected micropollutants in both lists are structurally diverse, have wide-ranging physico-chemical properties and cover a large spectrum of applications. The systematic compound selection approach presented here can also be adjusted to fit individual research needs with respect to type of micropollutants, treatment processes and number of compounds selected.

Identification and application of physical and chemical parameters to predict indicator bacterial concentration in a small Californian creek

This study of Aliso Creek in California aimed to identify physical and chemical parameters that could be measured instantly to be used in a model to serve as surrogates for indicator bacterial concentrations during dry season flow. In this study, a new data smoothing technique and ranking/categorizing analysis was used to reduce variation to allow better delineation of the relationships between adopted variables and concentrations of indicator bacteria. The ranking/categorizing approach clarified overall trends between physico-chemical data and the indicators and suggested sources of the bacteria. This study also applied a principle component regression model to the data. Although the model was promising for predicting concentrations of total and fecal coliforms, it was somewhat weaker in predicting enteroccocci.