Institute of Nutritional Science Chair of Food Science Justus Liebig University Giessen Non-target screening of emerging bioactive and hazardous compounds in complex matrices via multi-hyphenated techniques Cumulative dissertation for the degree of Doctor rerum naturalium (Dr. rer. nat.) Submitted to the Faculty of Agricultural Science, Nutritional Science, and Environmental Management Justus Liebig University Giessen by Tamara Schreiner (M. Sc.) Fulda, Germany Giessen, 2022 I With permission of the faculty Agricultural Science, Nutritional Science, and Environmental Management Justus Liebig University Giessen Examining committee: 1st Reviewer: Prof. Dr. Gertrud Morlock 2nd Reviewer: Prof. Dr. Bernd Lindemann 3rd Examiner: Prof. em. Dr. Wolfgang Schwack 3rd Examiner: Prof. em. Dr. Bernd Honermeier II III Table of Contents Declaration ........................................................................................................................................................ V Acknowledgments ....................................................................................................................................... VII Scientific contributions ................................................................................................................................ IX List of figures ................................................................................................................................................ XIII List of abbreviations ................................................................................................................................... XV 1. Introduction .................................................................................................................................... 1 1.1. Non-target screening for bioactivity in food ......................................................................... 1 1.2. Status-quo high-performance thin-layer chromatography hyphenations ................. 3 1.2.1. High-performance thin-layer chromatography–effect-directed analysis ................... 4 1.2.2. Microplate assays versus planar assays ................................................................................ 7 1.2.3. High-performance thin-layer chromatography–effect-directed analysis– high-resolution mass spectrometry ......................................................................................... 7 1.2.4. On-surface metabolisation .......................................................................................................... 8 1.3. Online desalting and orthogonal separation ........................................................................ 9 1.4. Mass analysers: single quadrupole versus orbitrap......................................................... 10 1.5. Diffusion susceptibility of silica gel layers ........................................................................... 11 1.6. Scope ................................................................................................................................................. 12 1.7. Progress achieved through multi-hyphenated techniques ............................................. 12 1.7.1. Establishment of an eight-dimensional hyphenation (Publication 1) ....................... 13 1.7.2. Application field study using the eight-dimensional hyphenation (Publication 2)15 1.7.3. Ten-dimensional hyphenation for non-target screening (Publication 3) ................. 17 1.7.4. Application field study using the ten-dimensional hyphenation (Publication 4) ... 19 1.7.5. Diffusion reduction permitting multiplexed assay formats (Publication 5) ............ 19 1.7.6. Application field study for multiplexed estrogen screen (Publication 6) .................. 20 1.8. References ....................................................................................................................................... 23 2. Publication 1 ................................................................................................................................. 33 3. Publication 2 ................................................................................................................................. 57 4. Publication 3 .............................................................................................................................. 109 5. Publication 4 .............................................................................................................................. 155 6. Publication 5 .............................................................................................................................. 173 7. Publication 6 .............................................................................................................................. 227 8. Summary ...................................................................................................................................... 253 9. Zusammenfassung ................................................................................................................... 255 IV V Declaration I declare: this dissertation submitted is a work of my own, written without any illegitimate help by any third party and only with materials indicated in the dissertation. I have indicated in the text where I have used texts from already published sources, either word for word or in substance, and where I have made statements based on oral information given to me. At any time during the investigations carried out by me and described in the dissertation, I followed the principles of good scientific practice as defined in the “Justus Liebig University Giessen Statute for Ensuring Good Scientific Practice”. Giessen, August 2022 ____________________________ Tamara Schreiner VI VII Acknowledgments My special thanks go to my first supervisor Prof. Dr. Gertrud Morlock, who gave me the opportunity to do my doctorate in her research group. Thank you very much for this instructive time and your dedicated support in all matters. I would also like to thank Prof. Dr. Bernd Lindemann for proving the second review for my doctorate. Further, a big thank you to my wonderful colleagues Stefanie Kruse, Annabel Mehl, Alisa Ronzheimer und Isabel Müller, who have always been there to help and advise me, and who have also sometimes cushioned my bad mood. I could not think of any better collegial support. I am also grateful for the good souls of the working group Julia Heil and Simone Cutts. Without you, the team would not be the same. I would also like to extend my sincere thanks to the thesis students Dorena Sauter, Maren Friz, and Naila Eggerstorfer for assistance in laboratory work. Last but not least, I want to thank my beloved family, my parents and my sister, for their support during my studies and my PhD. Without you I would never have made it. I love you. VIII IX Scientific contributions Peer-reviewed original research papers 1) Schreiner, T., Morlock, G.E.: Non-target bioanalytical eight-dimensional hyphenation including bioassay, heart-cut trapping, online desalting, orthogonal separations and mass spectrometry, J. Chromatogr. A 1647, 2021, 462154 Author contributions: Tamara Schreiner: Conceptualization, Methodology, Experimental Analysis, Data Analysis, Writing – Original Draft. Gertrud E. Morlock: Conceptualization, Methodology, Supervision, Writing – Review and Editing. 2) Schreiner, T., Sauter, D., Friz, M., Heil J., Morlock, G.E.: Is our Natural Food our Homeostasis? Array of A Thousand Effect-Directed Profiles of 68 Herbs and Spices, Front. Pharmacol. 12, 2021, 755941 Author contributions: TS carried out all bioassays and mass spectrometry experiments, analyzed the data, and wrote the manuscript draft. DS and MF prepared the B. subtilis, α-/β-glucosidase, β-glucuronidase, pYAS, pYES, pYAAS, pYAES (bio)autograms. JH prepared the samples and A. fischeri, tyrosinase, and α-amylase (bio)autograms. GM initiated the project, concept and methodology, obtained research funding, supervised the study, and revised the manuscript. 3) Schreiner, T., Eggerstorfer, N., Morlock, G.E.: Effects of gastrointestinal digestion on the bioactivity of convenience tomato products examined by ten-dimensional hyphenation – submitted to J. Agric. Food Chem., 2022 Author contributions: Tamara Schreiner: Conceptualization, Methodology, Investigation, Data Analysis, Writing – Original Draft. Naila M. Eggerstorfer: Investigation. Gertrud E. Morlock: Conceptualization, Methodology, Supervision, Writing – Review and Editing. 4) Schreiner, T., Eggerstorfer, N., Morlock, G.E.: Effects of gastrointestinal digestion on bioactivity of meal replacement products studied by ten-dimensional hyphenation – submitted to Food Funct., 2022 Author contributions: Tamara Schreiner: Conceptualization, Methodology, Investigation, Data curation, Writing – Original Draft. Naila M. Eggerstorfer: Investigation. Gertrud E. Morlock: Conceptualization, Methodology, Supervision, Funding Acquisition, Writing – Review and Editing. X 5) Schreiner, T., Ronzheimer, A., Friz, M., Morlock, G.E.: Multiplex planar bioassay with reduced diffusion on normal phase, identifying androgens, verified antiandrogens and synergists in botanicals via 12D hyphenation, Food Chem. 395, 2022, 133610 Author contributions: T. Schreiner: Methodology, Investigation, Formal analysis, Writing – original draft. A. Ronzheimer: Methodology, Investigation, Formal analysis, Writing – original draft. M. Friz: Investigation. G.E. Morlock: Conceptualization, Methodology, Supervision, Funding acquisition, Project administration, Writing – review & editing. 6) Ronzheimer, A., Schreiner, T., Morlock G.E.: Multiplex planar bioassay detecting estrogens, antiestrogens, false-positives and synergists as sharp zones on normal phase, Phytomedicine 103, 2022, 154230 Author contributions: A. Ronzheimer: Conceptualization, Methodology, Investigation, Data curation, Writing – original draft. T. Schreiner: Conceptualization, Methodology, Investigation, Data curation, Writing – original draft. G.E: Morlock: Conceptualization, Methodology, Supervision, Writing – review & editing. Peer-reviewed original research paper (co-authorships) 1) Morlock, G.E., Ziltener, A., Geyer, S., Mehl, A., Schreiner, T., Kamel, T., Tersteegen, J., Brümmer, F.: Indo-Pacific bottlenose dolphins self-medicate with invertebrates in coral reefs, iScience, 2022, 104271 Author contributions: Conceptualization: A.Z., F.B., T.K. (sampling), G.E.M. (analysis); Methodology: A.Z., J.T., F.B. (sampling), G.E.M. (analysis); Sampling: A.Z., J.T.; Investigation: S.G. (all HPTLC-assay experiments), A.M. (HRMS), T.S. (hormonal-effective bioassays and respective MS), J.T. (HPTLC-A. fischeri assay); Funding acquisition: A.Z., G.E.M., F.B.; Project administration: A.Z., G.E.M.; Supervision: F.B. (dive protocol), A.Z. (sampling), G.E.M. (analysis); Writing – original draft: G.E.M., A.Z.; Writing – review & editing: G.E.M., A.Z., F.B., J.T. 2) Schreiner, T., Morlock, G.E.: Investigation of the phytoestrogenic potential of rose, red and white wines via effect-directed 9D hyphenation – submitted to J. Chromatogr. A, 2022 Author contributions: Tamara Schreiner: Conceptualization, Methodology, Investigation, Data Analysis, Writing – Original Draft. Gertrud E. Morlock: Conceptualization, Funding acquisition, Methodology, Supervision, Writing – Review and Editing. XI Poster presentations Presenter in italics 1) Curious2022 Future Insight™ conference, Darmstadt, 12-14.07.2022: Morlock, G.E., Ziltener, A., Geyer, S., Mehl, A., Schreiner, T., Kamel, T., Tersteegen, J., Brümmer, F.: Dolphins' beauty secrets, poster presentation, awarded with the poster prize XII XIII List of figures Figure 1. Common hyphenation possibilities for non-target screening ................................. 2 Figure 2. Schematic representation of the eight-dimensional hyphenation ....................... 13 Figure 3. Calculated salt and nutrient load on a HPTLC silica gel 60 F254 MS-grade plate for respective (bio)assays. ................................................................................................... 15 Figure 4. Irreversible mitochondrial conversion of the blue dye resazurin ....................... 21 XIV XV List of abbreviations 10D ten-dimensional 12D twelve-dimensional 8D eight-dimensional AChE acetylcholinesterase BChE butyrylcholinesterase DAD diode array detector ddMS2 data-dependent MS2 EDA effect-directed analysis F254 fluorescence indicator, green fluorescent at 254 nm fix fixated zones through Degalan coating FLD fluorescence light detection hAR human androgen receptor hER human estrogen receptor HPLC high-performance liquid chromatography HPTLC high-performance thin-layer chromatography RF retention factor HRMS(/MS) high-resolution (tandem) mass spectrometry LC liquid chromatography MS mass spectrometry MU 4-methylumbelliferone MUG 4-methylumbelliferyl β-D-galactopyranoside m/z mass-to-charge ratio NP normal phase NTS non-target screening pYAS planar yeast androgen screen pYAAS planar yeast antagonist androgen screen pYAVAS planar yeast antagonist verified androgen screen pYES planar yeast estrogen screen RP reversed phase RP-18 W wettable reversed phase TAGs triacylglycerols TIC total ion current TLC thin-layer chromatography XVI TOF time-of-flight UV ultra violet V verification Vis visible light, white light illumination Introduction 1 1. Introduction Current non-target screening (NTS) strategies for food focus mainly on illicit adulterations [1, 2], emerging contaminants [1–3], or pesticides [4, 5]. For the screening of new drugs, nature offers an invaluably rich source of plant-based bioactive secondary metabolites [6–9]. Imposing pharmacological effects in humans, nature-derived bioactive constituents could help in preventing diseases or maintaining the health status [7]. Considering thousands of unknown compounds in natural products, the need for comprehensive technologies to determine unknown natural features is increasing [10]. To cover and record this diversity of substances, high-performance liquid chromatography (HPLC) coupled with high-resolution mass spectrometry (HRMS) is usually used [11, 12]. The enormous datasets resulting from HPLC–HRMS analyses need substantial reduction and prioritisation for further evaluation [10, 12]. Data processing can bias the results of NTS strategies by focusing on the most abundant compounds, neglecting those of minor abundance [12]. Since the experimentally generated datasets cannot be handled properly without processing, the amount of data to be evaluated must be reduced otherwise, e.g., by prioritising bioactive compounds [10]. To implement the bioactivity aspect in current NTS strategies, HPLC–HRMS methods have to be expanded by biological and biochemical detection methods. The direct combination of analytical methods and biological activity testing still remains challenging [13]. Enzymatic and microbiological assays require neutral, aqueous, or buffered conditions, while identification methods, such as HPLC–HRMS, need salt-free, pure organic solvents. In recent years, different groups worked on natural product research and the identification of unknown bioactive compounds by in silico combination of data from analytical and biological screenings [6, 14–17]. 1.1. Non-target screening for bioactivity in food Screening approaches for biologically active compounds from natural products are expensive, time-consuming, elaborate, and require a lot of equipment [8, 13]. The two major challenges for the NTS for bioactivity are genericity and multidisciplinarity. Sample preparation is a critical step concerning selectivity and sensitivity. To avoid any loss of substances or discrimination of some analytes, sample preparation requires non-selective and wide-scope methods. However, an utmost generic sample preparation goes along with matrix interferences, especially for complex samples [18]. The second challenge is Introduction 2 the interdisciplinary hyphenation of chemical analysis and biological detection [10] which makes it difficult to assign the bioactivity to individual compounds from complex matrices [11, 15, 19]. Effect-directed analysis (EDA) commonly comprises several analytical methods, such as fractionation, chromatography, bioassay, and mass spectrometry [11] embedded in off-line [17, 20, 21], at-line [9, 16], or on-line [6] workflows (Figure 1). The most commonly employed chromatographic technique for EDA is HPLC [6, 9, 11, 15–17, 21]. The advantage of HPLC is its high separation capacity. Nevertheless, HPLC coupled with (bio)assays could be problematic due to solvent incompatibility and limited combination possibilities with assay variants. Off-line screenings usually consist of iterative microfractionation into microplate format, followed by respective (bio)assays, and subsequent HPLC–HRMS analysis of bioactive fractions [16, 17, 21]. This technique harbours the risk of overlooking bioactivity as only the whole fraction is determined as sum parameter and opposing effects can cancel each other out [22]. A clear assignment of a substance to an effect is biased since there are usually several chemically related analytes in one fraction. Off-line formats on the other hand are compatible with almost any (bio)assay regardless of incubation time or assay material [16]. For at-line and on-line methodologies, a flow-split is embedded after chromatographic separation. In at-line approaches, one part is fractionated in Figure 1. Common hyphenation possibilities for non-target screening by combing separation techniques, such as high-performance liquid chromatography (HPLC) or high-performance thin-layer chromatography (HPTLC), with high-resolution mass spectrometry (HRMS), and (bio)assays. Comparison of off-line, at-line, and on-line hyphenation options proposed in the literature [6, 11, 16, 17, 23]. Introduction 3 microplates, while the other part is analysed via HRMS [16]. In on-line variants, chromatographed effluent is divided proportionally into HRMS and to a completely automated (bio)assay procedure realised through continuous injection of necessary reagents into a reaction coil [6]. Computational correlation in at-line and on-line approaches enables precise assignment of the detected effect to an individual molecule while being limited in (bio)assay variants [16]. On-line variants are additionally restricted to fast-reacting biological systems and influenced by the organic solvent inevitably present after HPLC separation, interfering with biological or biochemical assays [11] Besides HPLC, other separation methods, such as gas chromatography [20] and high-performance thin-layer chromatography (HPTLC) [10, 11, 23–25] are also used in EDA but are less reported. In Figure 1, the currently performed off-line NTS for bioactivity with HPTLC is displayed for the sake of completeness but firstly introduced in section 1.2.3. 1.2. Status-quo high-performance thin-layer chromatography hyphenations HPTLC is an optimised version of thin-layer chromatography (TLC), employing advanced instrumentation and improved layer properties (smaller particle size, more homogenous particle distribution), which allow reliable results through standardisation. The capability in selectivity options, application volumes, and derivatisation possibilities, renders HPTLC an emerging technology for various applications [26]. In particular, the matrix robustness of HPTLC is attractive to handle complex samples without time-consuming sample preparation [10]. Although HPTLC has less separation power compared to HPLC, it is superior in some regards. The whole sample is stored on the plate and can be detected by different exposures (white light illumination, ultra-violet light, or fluorescence light). In HPLC, compounds that do not interact with the stationary phase are rushing through the column, eluting within the dead volume, and thus are not captured by any detector. While in HPTLC multiple samples are chromatographed side-by-side under the same conditions, HPLC has a lower throughput and a greater variation in analysis conditions due to sequential execution. After chromatography, HPTLC plates could be chemically, biochemically, and biologically derivatised [10, 11, 26]. For HPLC, common detectors are limited in versatility or even destructive, and implementations of derivatisations within the analysis are still technically complex [19]. Introduction 4 1.2.1. High-performance thin-layer chromatography–effect-directed analysis EDA as a form of post-chromatographic derivatisation is therefore easier to combine with HPTLC. A great advantage of HPTLC–EDA is the complete evaporation of the organic solvent from the silica gel layer [11]. Although, a disadvantage, especially for normal phase (NP)-HPTLC, is the diffusion susceptibility of zones in aqueous bioassays during long incubation times [11, 27]. HPTLC–EDA techniques can be divided into bioautography and biochemical assays. Both are performed post-chromatographically. Besides the differentiation in bioautography and biochemical detection, two application variants of planar assays can be distinguished. First, the whole-plate assay, where the biological material is distributed all over the plate after chromatography [28]. Second, the start zone assay, where e.g. an enzymatic metabolisation process takes place exclusively on the application zone, and chromatography is performed after the assay [29–31]. To ensure the viability of the organisms and the proper quaternary structure of enzymes on the (HP)TLC plate, appropriate conditions on the silica gel layer must be given, i.e., almost neutral pH, moist environment, and optimal working temperature for the respective organism or enzyme [11]. In case of acidic or alkaline mobile phases, a neutra- lisation step has to be involved before the assay procedure. To prevent the (HP)TLC plate from drying, moist and humid conditions during incubation can be achieved by placing the plate in a closed, humidity-controlled environment, e.g. a tight and moistened box. Both, biochemical assays and bioautography, are commonly used for bioprofiling for new effective substances from natural complex matrices [10, 28, 32]. The focus of the subsequent assay description is on the EDA applied in the present work, which targets physiological effects which play a role in the development of human diseases. In general, biochemical assays are based on the principle of enzyme-substrate interaction. Commonly employed biochemical assays, used in both microplate and HPTLC, include cholinesterases, glucuronidase, tyrosinase, glucosidases, and amylase. Executing planar biochemical assays, enzyme solutions are applied onto the plate and after an incubation period, inhibiting effects are visualised using substrates that are usually enzymatically converted to colourful products. Zones in which the enzymatic function is inhibited, remain colourless [11], except for the α-amylase assay, where enzyme-inhibiting zones are dyed blue-violet [33]. The enzymes acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) are involved in neurotransmission processes. The disease pattern of Alzheimer’s is characterised by a deficiency of the neurotransmitter acetylcholine in the synapses of the cerebral cortex. AChE and BChE catalyse the Introduction 5 degradation of this neurotransmitter, with leads to a further deterioration of neuronal transmission. Inhibition of these enzymes is the main target in the deceleration of Alzheimer’s disease [34, 35]. For detoxification, substance elimination from the human body is often regulated via glucuronidation pathways. Compounds are marked by adding a glucuronide moiety to the molecule. In the intestine, these processes can be reverted by bacterial β-glucuronidase, resulting in gastrointestinal malfunction and disease. Enzyme inhibitors address explicitly the extra loop of the bacterial enzyme, which is missing in the human ortholog, to prevent bacterial deglucuronidation of the molecule to be eliminated [36]. Tyrosinase is a key regulator of melanogenesis. Also, it is responsible for the browning of foods. By using molecular oxygen, tyrosinase converts monophenols via the intermediate diphenols to the final quinones [37]. Polymerisation of built quinones results in the respective pigments, such as melanin. Excessive production of melanin results in hyperpigmentation, acne, and freckles [38]. The inhibition of tyrosinase activity is of high interest to the cosmetics industry, as it can prevent skin abnormalities and be used as skin whitening agent [37, 38]. The three enzymes α-glucosidase, β-glucosidase, and α-amylase are involved in the breakdown of oligosaccharides into monomers. The small Greek letter indicates which kind of bond, they are able to cleave, either α- or β-linked oligosaccharides. The released glucose molecules are absorbed from the intestine and lead to an elevation of the blood glucose level. Since diabetes patients are poor in lowering hyperglycaemia levels, the inhibition of these enzymes is desirable for the treatment of diabetes. By the prevention of oligosaccharide hydrolysis, less free glucose is available for absorption and thus a balanced blood glucose level could be maintained [33, 39]. In bioautography, living organisms are applied onto the (HP)TLC surfaces to detect biological effects [11]. As a result, a bioautogram is obtained. Recently, several microorganisms were employed for HPTLC–EDA, e.g., Aliivibrio fischeri (A. fischeri) [40], Bacillus subtilis (B. subtilis) [41], Salmonella typhimurium (S. typhimurium) [42], or Saccharomyces cerevisiae (S. cerevisiae) [27, 43, 44]. In direct bioautographic assays, the bioactivity can be instantly measured, for example as enhanced or attenuated bioluminescence of the Gram-negative A. fischeri bacteria. The natively bioluminescent A. fischeri immediately shows a change in photon emission when exposed to harmful or toxic substances for the marine bacteria. Reduced photon emission is an indicator for antimicrobial and toxic substances [11, 40, 45], whereas an increased photon emission Introduction 6 indicates a stimulated energetic cell metabolism. Weins and Jork (1996) introduced this in situ bioassay to TLC plates [45]. Indirect approaches firstly expose reporter organisms equipped with specific operons to their trigger molecules. This exposure induces the production of enzymes decoded in the operon. Bioactivity is then measured indirectly through enzymatic substrate conversion. As examples of indirect bioautography, the yeast estrogen and androgen screens (YES and YAS), the SOS-Umu-C assay, and the B. subtilis bioassay are described more precisely. The endocrine system regulates the hormonal homeostasis of almost all vertebrates. An imbalance has severe impacts on juvenile development, menstrual cycle, fertility, and other hormonal regulated processes [44]. Endocrine-disrupting chemicals are detected in the YAS and YES as well as in their antagonistic versions. Employed yeast strains are genetically modified to contain the human androgen/estrogen receptor (hAR/hER) [46, 47]. In presence of hormone-like substances, the ligand-receptor interaction induces the expression of β-galactosidase through the cleavage of the repressor protein, which formerly prevented gene transcription of the enzyme [43]. Hormonal effects are indirectly measured as enzymatic activity mediated through hydrolysis of chromogenic or fluorescing substrates. For antagonistic assays, an agonist is artificially introduced to the bioassay system. Agonists and antagonists compete for free binding sites at the receptors. Antagonists block the receptor so that less enzyme expression is induced. Antagonistic effects are determined as reduction of β-galactosidase activity [48]. For planar antagonist screening, the right side of each track is oversprayed with an agonist-stripe along with the chromatographed track (see Publication 5, Figure 1). β-Galactosidase activity is then detectable over the whole track, except for those zones containing antagonistic substances [44]. To screen for genotoxic substances, the SOS-Umu-C assay can be used. The principle is based on the SOS DNA repair mechanism which is regulated via several genes [49]. In response to DNA damage induced by mutagens and genotoxins, the genetically modified S. typhimurium TA1535/pSK1002 activates the SOS signal cascade, resulting in the expression of the enzyme β-galactosidase. Comparable to hormonal effects, genotoxicity can be determined indirectly as β-galactosidase activity [42, 49, 50]. Antibiotics are commonly used to treat bacterial infections. EDA for potential drug candidates can be realised with the Gram-positive B. subtilis bioassay. Present substances with antibiotic-like properties lead to cell death or bacteriostatic conditions. The cell Introduction 7 viability is detected indirectly via a tetrazolium salt that is converted into a violet formazan by the dehydrogenases of intact Bacilli [41, 51]. 1.2.2. Microplate assays versus planar assays Microfractionation for microplate assays require an upstream separation technique, such as chromatography or electrophoresis (Figure 1). Fractions are usually collected time-controlled or peak-wise [19]. In case of chromatography-based fractionation, vessel capacity and flow should match, otherwise a split must be installed, resulting in a considerable loss of sample. Time-controlled fractionation often collects several analytes with similar chemical properties in one vessel which leads to the determination of the sum of all biological responses comprised in this well. Evaporating organic solvent from the microwells is mandatory for (bio)assay procedures, but some compounds only show slight solubility in aqueous media and therefore are not re-dissolved after drying [19]. Planar (bio)assays are a good alternative to microplate assays where HPTLC serves as an upstream separation technique. After chromatography, the complete solvent can be evaporated and aqueous assay media, buffers, and substrates can be applied on the same surface without analyte loss. Biological effects can be directly assigned to specific zones on the plate, which can be further evaluated to individual substances causing the bioactivity. Planar (bio)assays showed to be more sensitive compared to microplate assays, probably due to lower matrix interferences [30, 42, 51]. 1.2.3. High-performance thin-layer chromatography–effect-directed analysis– high-resolution mass spectrometry State-of-the-art HPTLC–EDA–HRMS hyphenations are comparable to off-line NTS methods (Figure 1), whereby HPTLC–EDA and HPTLC–HRMS are performed in a time- shifted manner, which only allows indirect coupling. On one plate (20 cm × 10 cm) samples were chromatographed in duplicate as two sets, and cut into two identical halves (10 cm × 10 cm), of which only one was subjected to the (bio)assay [23]. According to the effect profile of the (bio)autogram, zones of interest were selected on the remaining clean chromatogram, either manually [23] or automated [52]. Plates were imprinted by an elution head, eluting the selected zones to an HRMS instrument. After the MS recording, accurate positioning of the elution head was verified by performing the same (bio)assay on the imprinted chromatogram [23, 52]. Introduction 8 This procedure has disadvantages in several respects. Duplicate application on one HPTLC plate reduces the number of samples being analysed in a single chromatographic run. Zone marking on the identical half without (bio)assay must be as precise as possible. Incorrectly placed imprints cannot be re-analysed so easily. Correction attempts, which are often only a few millimetres away, would result in a leakage of the elution head, since the silica gel layer is already damaged near this zone. Moreover, the (bio)assay procedure has to be performed twice, requiring twice the time, effort, and material [53]. Currently, the high matrix and salt load of the (bio)assay hampers direct hyphenation of HPTLC– EDA–HRMS. The aim to directly elute the bioactive zone from the (bio)autrogram into HRMS can only be realised by including a desalting step, reducing the interfering matrix and salts. 1.2.4. On-surface metabolisation NTS for bioactive compounds in food is reasonable as it is more straightforward compared to random, non-prioritised NTS strategies. Considering effects on living organisms, compound distribution, half-life, resorption, and metabolisation in vivo have to be taken into account. Effective constituents from the diet may not reach the organism in the same form as they were applied [13]. Digestion plays a major role in the breakdown of macronutrients such as saccharides, lipids, and proteins [31, 54]. The three main digestive enzyme classes are proteases, lipases, and amylases, which are distributed differently throughout the gastrointestinal tract (GIT) [55]. Enzymes involved in digestion processes are delivered by the salivary gland, liver, and pancreas [54]. Most accurate results in the prediction of food breakdown, nutrient release, and absorption into the bloodstream are provided by in vivo experiments [54, 55]. To avoid ethical issues, static digestion models were standardised in vitro to mimic oral, gastric, and intestinal metabolisation [31, 56]. Current in vitro models are a good alternative to animal or human studies providing comparable results for the digestion of macronutrients. Prediction models for the in vitro digestion of micronutrients, such as bioactive compounds, carotenoids, or polyphenols are deviating from in vivo results [57]. The realistic simulation of the digestion requires the right pH and composition of enzymes, electrolytes, and co-factors [56]. Morlock et al. (2021) recently introduced the harmonised digestion procedure [56] from solution-based assays to planar surfaces. In brief, samples are applied to the silica gel layer. Digestive enzymes and respective co-factors are oversprayed. To set humid conditions, the plate is wetted and incubated. Introduction 9 After simulated digestion, samples are chromatographed and derivatised either chemically, biochemically, or biologically. On-surface digestion with subsequent HPTLC showed comparable results to in vitro digestion followed by planar chromatography. The on-surface nanoGIT+active system is advantageous because enzyme and sample consumption is kept minimal (nano). Metabolisation and separation are performed on the same layer and could further be combined with multiple detection options [31]. Especially drugs or toxins are metabolised in the liver. Liver metabolism plays a major role in activation or inactivation of compounds introduced to the human body. To transfer the complex liver biotransformation processes to in vitro models, a homogenised liver extract from rats, the so-called S9 mixture, is commonly used. S9 metabolisation affects bioactivity, e.g., metabolic activation of estrogenic properties of bisphenol A [58], stimulation of mutagenic potential of aflatoxin B1 [59], or conferment of neurotoxic attributes to chlorpyrifos [30]. Planar on-surface assays exploiting the S9 activation system could either be performed as start-zone assay [30], or whole-plate assay [60] by adding the liver homogenate directly to the (bio)assay material applied. 1.3. Online desalting and orthogonal separation By eluting active zones directly from the (bio)autogram, assay salts and media components are co-transferred automatically. In case of alkaline or acidic mobile phases, additional salt load from neutralisation buffers is co-eluted to HRMS. Therefore, employing a desalting step is crucial to prevent ion suppression, clogging of the ion source through crystallisation, and signal suppression [61]. In the fields of proteomics [62], genomics [63], and metabolomics [61, 64], on-line desalting is already embedded in HPLC–HRMS analyses due to very saline sample preparation. Technically they all follow more or less the same principle. A second stationary phase is introduced in form of a guard column [63, 64], second separation column [62], or a solid-phase-extraction column [61] before [61, 63, 64] or after [62] the analytical column. Analytes are retained on the stationary phase, while salts and buffer materials are discarded [63]. After the desalting step, samples are transferred to HPLC–MS. With embedding a desalting step into the HPTLC–EDA–HRMS workflow, it might be unnecessary to perform HPLC, since the chromatographic separation has already occurred on the HPTLC plate. Nevertheless, an orthogonal chromatography is reasonable for several aspects. As already mentioned, HPTLC has a lower separation capacity compared to HPLC [11]. Chemically similar compounds could retard at the same Introduction 10 migration distance on the plate, which makes an orthogonal separation mandatory to assign the bioactivity to an individual substance. As it is technically less complex to combine EDA with HPTLC, rather than HPLC, the latter can be used to separate potentially co-eluting analytes from the planar (bio)autogram. Moreover, additional information about the unknowns is gained as two retention factors (RF) are determined within a single workflow: First, the migration distance on NP-HPTLC plates and second, the retention time on the reversed phase (RP) HPLC column. 1.4. Mass analysers: single quadrupole versus orbitrap Usually, mass analysers are selected according to the research purpose. For targeted approaches, MS instruments with low resolution power are adequate, e.g., for the quantification of known analytes. For example, single quadrupole mass spectrometers are applied in routine analysis. The operating principle of single quadrupole instruments is based on four metal rods arranged in parallel. A radio frequency voltage is applied to the rods. Opposite rods have opposite potential. The resulting voltage field forces the ions on amplitudes through the quadrupole. The mass-to-charge ratio (m/z) can be calculated from the ion trajectories [65]. Single quadrupole instruments do not provide enough resolution to calculate a molecular formula from the obtained m/z ratio. Only with a list of suspected compounds, this MS is sufficient as identifying detector, since the assumed candidates can be confirmed against available standards. For NTS purposes, higher resolution instruments are required, e.g., time-of-flight (TOF) or orbitrap mass analysers. Since its launch in 2005, orbitrap technology has replaced several TOF instruments that were most commonly used until then [66, 67]. The basic operating principle is relying on outer electrodes and a central electrode on which voltages are applied. Ions are introduced in the gas phase between the outer and inner electrodes. The electric field resulting from the outer electrodes forces the ions to oscillate along the axis, while the inner electrode induces rotation. The trajectory of the ions can be described as circular spirals which can be mathematically converted into a mass spectrum via Fourier transformation [67]. Several technological advances made the orbitrap even more attractive for NTS. The possibility of pulsed injection via an external storage device, the so-called curved linear trap, decoupled the mass analyser from the ion source and was a prerequisite to introduce a fragmentation option. A higher-energy collision induced dissociation cell in form of a multipole was added to obtain structural information by fragmentation [67, 68]. To implement this advantageous mass analyser in Introduction 11 routine analysis and for research purposes, the orbitrap technology was produced in benchtop format as hybrid quadrupole-orbitrap instrument [68]. Since the orbitrap is superior compared to the single quadrupole concerning resolution, exact mass analysis, and fragmentation option, it is preferred in suspect screening and NTS. 1.5. Diffusion susceptibility of silica gel layers NP-silica gel layers are most commonly used in HPTLC. Unfortunately, they are susceptible to diffusion due to long incubation times with aqueous bioassay media [27, 69, 70]. State-of-the-art HPTLC–bioautography methods are based on immersion of the HPTLC plate into the cell suspension for a few seconds. Incubation of an immersed NP plate at nearly 100% humidity causes band broadening, zone distribution over the plate, and blurred signals [27, 70, 71]. Salts from bioassay and neutralisation buffers additionally promote these effects. Klingelhöfer and Morlock (2014) tried to counteract the problem by introducing a hybrid layer, the so-called wettable reversed phase (RP-18 W) HPTLC plate. As a result, they obtained sharp-bounded bands at the expense of sensitivity [71, 72]. To reduce the diffusion on NP layers, the immersion step was replaced by a spray-on technology, considerably improving the assay quality [69, 72]. By spraying, the cell suspension is applied in a more controlled fashion and a defined cell layer thickness on the planar chromatogram [72]. Nevertheless, signals are still blurred after the bioassay. For planar immunoassays, several washing and incubation steps are required in which the silica gel layer could flake from the support [73]. By fixation of the layer with a plastic, usually polyisobutyl methacrylate [74], the silica is kept stable throughout the procedure [73, 75]. Besides the layer stability through long incubation times, the impregnation with the plastic showed another advantage, in particular sharp-bounded zones on NP. Other options to modify the silica gel layer are polyethylene glycol or poly-D-lysine. Through chemical reactions of the silanol groups with the coating agent, a change in interaction, surface topology, and chemistry of the silica conferred biocompatible properties to the layer [76, 77]. Realising low-diffusion bioassays on NP plates is the prerequisite for multiplex assays. Introduction 12 1.6. Scope The lack of easy, information-rich, and highly-streamlined NTS strategies for bioactive ingredients in food initiated the development of new, innovative, fast, and multi-hyphenated techniques. To obtain the most comprehensive information possible, chromatographic, biological, biochemical, spectroscopic, and spectrometric methods were drawn from the analytical toolbox. A few of the main advantages of HPTLC were decisive for the use of this methodology as a basis for further hyphenations. First, HPTLC offers the possibility for high-throughput screening, second, the matrix robustness, and third, the variety in biological and (bio)chemical hyphenation options. As most NTS strategies exploit HRMS as an identification method, a direct hyphenation of HPTLC, (bio)assay, and HRMS was sought to be investigated. State-of-the-art HPTLC–EDA–HRMS experiments require doubled time, material, and effort. The aim is to simplify the commonly executed workflow. Bioactive zones should be directly eluted out of the planar (bio)autogram to an MS instrument. Saline (bio)assay media are co-transferred with the analytes from the plate causing interferences with ionisation and noisy backgrounds in MS recording. This fact makes the reduction in salt load an inevitable step to realise a direct coupling. The variety in possible HPTLC–EDA hyphenations demands a generic desalting before further analyses. To overcome the problematical occurrence of co-elutions, a second chromatographic step is desired. Up to now, HPTLC–EDA cannot be completely automated, which is why the subsequent identification analysis steps should be as automated and standardised as possible. New NTS multi-hyphenated workflows based on HPTLC–EDA–HRMS require therefore the simplification of the current workflow, online-desalting, automation, and a second separation technique to reach the ultimate goal of structure elucidation of unknown bioactive compounds. 1.7. Progress achieved through multi-hyphenated techniques For elucidation of unknown bioactive compounds in food, it is essential to obtain as much information as possible. As already stated by Wilson and Brinkman (2007), “it is often necessary to have data from more than one spectroscopic technique” to identify a compound from complex mixtures [78]. Each information gained in the established multi-hyphenated workflows was denoted as a dimension. In this study, a comprehensive highly streamlined NTS strategy with up to twelve dimensions was developed which is presented in the following sections. The method is versatile and can be used both in routine and research, e.g., for monitoring of food quality and authenticity, discovery of Introduction 13 new drugs, or generally for the identification of unknown bioactive substances from any matrix. 1.7.1. Establishment of an eight-dimensional hyphenation (Publication 1) The first multi-hyphenation resulted in an eight-dimensional (8D) workflow (Figure 2) arranged in the following order, i.e., separation on NP-HPTLC plate (1D), detection under white light illumination (Vis, 2D), UV light (UV at 254 nm, 3D), and fluorescence light (FLD at 366 nm, 4D), bioprofiling (5D), and subsequent heart cut elution, on-line desalting, and RP-HPLC separation (6D) with diode array detection (DAD, 7D) and single quadrupole mass spectrometry (MS, 8D) [53]. The dimensions 1D–4D are routinely performed in HPTLC analysis. By focusing on bioactivity (5D), antibacterial effects, or those stimulating the energetic cell metabolism of the marine A. fischeri bacteria, were screened in Cinnamomum verum and C. cassia. Once a biological effect is detected, it is of highest interest to identify the compound(s) responsible. As identification strategy, the dimensions 6D–8D were provided. The direct hyphenation of the first five dimensions with the following three turned out problematic. The transfer of the bioactive zone from the planar medium to RP-HPLC–DAD–MS was performed using an elution-based TLC–MS interface. Thereby, not only the analyte(s), but the saline and nutrient-rich bioassay media were eluted from the plate. Those media or buffers from biochemical and microbiological Figure 2. Schematic representation of the eight-dimensional hyphenation exploiting NP-HPTLC separation (1D), multi-imaging detection (UV/Vis/FLD, 2D–4D), bioassay (5D), heart cut elution, analyte trapping/online desalting, RP-HPLC separation (6D) with diode array detection (DAD, 7D) and mass spectrometric detection (MS, 8D), resulting in separation of the co-eluting cinnamaldehyde (CA) and 2-methoxy cinnamaldehyde (2-MCA), and interference-free mass spectra [53]. Introduction 14 assays (5D) would interfere with ionisation and mass spectrometric detection (8D). Thus, salts and media components had to be eliminated before MS measurement. The very saline (approx. 45 g/L [79]) A. fischeri bioassay was selected to establish a desalting step between HPTLC and MS. For the reduction of MS interferences, four different approaches were tested: no desalting, desalting with an RP/ion exchange hybrid column, and with two different short guard columns (stationary material either capped with phenyl moieties or RP alkyl chains). All were mounted on a 2-position, 6-port switching valve as suggested by Fountain et al. (2004) [63]. Salt reduction capacities of the devices were determined by the elution of A. fischeri medium from a sample-free plate and comparison of MS total ion current (TIC) intensities in both polarity modes [53]. The RP guard column showed the best TIC reductions with −44.2% in ESI-positive ion mode, and −32.2% in negative ion mode, compared to non-desalted TICs. After the implementation of the RP guard column as a desalting device, the optimal elution solvent and time were evaluated. The selected guard column fulfilled two tasks simultaneously: trapping the analyte(s) on RP material while discarding the salts solved in the aqueous elution solvent. Since water has the greatest elution strength on silica gel plates, and a higher organic portion would cause analyte breakthrough, the best choice was 10% aqueous methanol as elution solvent. The duration of elution was set to 45 s, as enough analytes were transferred to give a strong signal response and breakthrough from desalting cartridge was kept minimal. As the example of cinnamon extracts showed, the orthogonal RP-HPLC separation proved to be reasonable. In the A. fischeri bioassay of the NP-HPTLC-separated samples, a dominant antibacterial zone near the solvent front was detected. To prove the functionality of the just established 8D hyphenation, it was aimed to identify this zone via RP-HPLC–DAD–MS. Results revealed two distinct signals of 2-methoxy cinnamaldehyde and cinnamaldehyde hidden behind this antibacterial zone. Both substances were retained at the same migration distance on the plate so it was impossible to assign the effect to one or the other. Standards confirmed that both contributed to the antibacterial effect detected in bioautography [53]. In conclusion, all information gained with the 8D workflow was essential to identify unknown substances exhibiting bioactivity. In this first multi-hyphenated pilot study the compounds identified were not unknown for cinnamon or their biological activity. Nevertheless, even more dimensions and better MS instrumentation would facilitate compound elucidation and identification. Introduction 15 1.7.2. Application field study using the eight-dimensional hyphenation (Publication 2) The functionality of the 8D hyphenation was shown for the A. fischeri assay and cinnamon as a sample. To prove the versatility and robustness, the workflow was transferred to 13 other biochemical and microbiological assays and 68 botanical samples [28]. As the previously investigated mobile phase for the botanical screening contained 12% acid [80], a neutralisation step was integrated into the workflow directly before (bio)assay. Interpretation of the effect-directed analyses (EDAs) is summarised in Table 1. The variety of (bio)assays required different neutralisation agents, buffers, and substrates. Salt and nutrient load of the individual assays are plotted in Figure 3. A detailed description of (bio)assay materials, preparation, and volume applied onto 20 cm × 10 cm HPTLC silica gel 60 F254 MS-grade plates is given in Publication 2 [28]. Calculations were based on the total (bio)assay load applied onto the 20,000 mm2 area, and downsized to the given format of the elution head (2 mm × 4 mm), resulting in an overall area of 8 mm2 of co-transferred analyte(s), neutralisation and (bio)assay salts, and nutrients. Figure 3. Calculated salt and nutrient load on a HPTLC silica gel 60 F254 MS-grade plate for respective (bio)assays. Salt load is given in µg per 8-mm2 zones, for neutralisation buffers and assay material (buffers, enzymes, substrates, dyes) respectively. Calculations were based on the experimental design described in the material and method section of Publication 2 [28]. For hormonal assays the amount of agonist stripe is negligible and calculations were made for 4-methyl umbelliferyl β-D-galactopyranoside as a substrate (*). SOS-Umu-C assay load was calculated with resorufin-β-D-galactopyranoside substrate and as it was performed on the thicker HPTLC silica gel 60 plates, requiring twice the volume, the amounts were halved for comparability (**). Introduction 16 Table 1. Interpretation of applied planar biochemical and biological assays (modified [25]). Chromogenic or fluorescent response depended on the used substrates (see Publication 2). e v a lu a ti o n im p ro v em en t o f ch o li n er gi c n eu ro tr an sm is si o n , p re ve n ti o n o f ac et yl ch o li n e h y d ro ly si s im p ro v em en t o f ch o li n er gi c n eu ro tr an sm is si o n , p re ve n ti o n o f ac et yl ch o li n e h y d ro ly si s n o rm al is ed d et o xi fi ca ti o n v ia g lu cu ro n id at io n p at h w ay s, o th er w is e h am p er ed b y b ac te ri al e n zy m e o rt h o lo gs b al an ce d m el an o ge n es is , p re v en ti o n o f h y p er p ig m en ta ti o n a n d sk in a b n o rm al it ie s b al an ce d b lo o d g lu co se le ve ls b y in h ib it in g th e ly si s o f o li go sa cc h ar id es in to it s gl u co se m o n o m er s b al an ce d b lo o d g lu co se le ve ls b y in h ib it in g th e ly si s o f o li go sa cc h ar id es in to it s gl u co se m o n o m er s b al an ce d b lo o d g lu co se le ve ls b y in h ib it in g th e ly si s o f o li go sa cc h ar id es in to it s gl u co se m o n o m er s p ro b io ti c- li k e ef fe ct , i n cr ea se d b io lu m in es ce n ce is r el at ed t o en h an ce d e n er ge ti c ce ll m et ab o li sm ( p h o to n e m is si o n ) an ti b ac te ri al s, d ec re as ed b io lu m in es ce n ce is r el at ed t o r ed u ce d ce ll v ia b il it y an d d ec re as ed e n er ge ti c ce ll m et ab o li sm ge n o to xi n s, m u ta ge n s an ti b ac te ri al s en d o cr in e d is ru p ti n g ch em ic al s, a go n is ts o f th e h u m an an d ro ge n r ec ep to r (h A R ) en d o cr in e d is ru p ti n g ch em ic al s, a n ta go n is ts o f th e h A R en d o cr in e d is ru p ti n g ch em ic al s, a go n is ts o f th e h u m an es tr o ge n r ec ep to r (h E R ) en d o cr in e d is ru p ti n g ch em ic al s, a n ta go n is ts o f th e h E R re sp o n se co lo u rl es s o n r o se b ac k gr o u n d co lo u rl es s o n r o se b ac k gr o u n d co lo u rl es s o n b lu e b ac k gr o u n d co lo u rl es s o n g re yi sh -b ro w n b ac k gr o u n d co lo u rl es s o n r o se b ac k gr o u n d co lo u rl es s o n r o se b ac k gr o u n d vi o le t o n c o lo u rl es s b ac k gr o u n d b ri gh t d ar k p in k f lu o re sc en ce o n b ro w n - re d d is h b ac k gr o u n d co lo u rl es s o n r o se b ac k gr o u n d b lu e fl u o re sc en ce re d u ct io n o f b lu e fl u o re sc en ce b lu e fl u o re sc en ce re d u ct io n o f b lu e fl u o re sc en ce b io ch e m ic a l a ss a y ac et yl ch o li n es te ra se [ 3 4 ] b u ty ry lc h o li n es te ra se [ 3 4 ] β -g lu cu ro n id as e [3 6 ] ty ro si n as e [3 7 , 3 8 ] α -g lu co si d as e [3 9 ] β -g lu co si d as e [3 9 ] α -a m yl as e [3 3 ] b io a u to g ra p h y A li iv ib ri o f is ch er i [4 0 , 4 5 , 7 9 ] Sa lm o n el la t yp h im u ri u m T A 1 5 3 5 /p SK 1 0 0 2 [ 4 2 ] B a ci ll u s su b ti li s [4 1 ] Sa cc h a ro m yc es c er ev is ia e B J1 9 9 1 [ 4 4 ] Sa cc h a ro m yc es c er ev is ia e B J3 5 0 5 [ 4 3 ] en zy m e in h ib it io n as sa y s G ra m - n eg at iv e b ac te ri a G ra m -p o si ti ve b ac te ri a ye as t ce ll s Introduction 17 The 8D hyphenation showed robust and reliable results through all assays and media. Since the used MS instrument had only a single quadrupole mass analyser, the resolution was too low to calculate the molecular formula. Evaluation of obtained data was based on literature and database research. One of the following criteria had to be true for a tentative assignment of the signal to a respective compound: Either the spectral data (at least absorbance maximum and mass) matched any database, the compound was already described for the botanical, or the substance already proved to have the respective biological properties. Some of the tentative assignments were verified or rebutted against available standards.For further optimisation of a non-target screening strategy for bioactive unknowns in complex mixtures, the 8D workflow was improved by the application of an MS instrument with higher resolving power and fragmentation option. Moreover, the question arose, if bioactive compounds can unfold their effective potential after passage through the digestive tract, or if they were metabolised to a related, but inactive molecule. 1.7.3. Ten-dimensional hyphenation for non-target screening (Publication 3) A combination of NTS, bioactivity studies, and identification of effect-compounds is highly desirable but still neglects human metabolisation pathways after oral administration [7, 13]. Harsh conditions in gastrointestinal food processing, residence time, absorption capability, and solubility of bioactive compounds influence their effectiveness [7]. The incorporation of the recently established nanoGIT system [31] and the upgrade to HRMS/MS, expanded the workflow to ten dimensions (NP-HPTLC-nanoGIT+active– UV/Vis/FLD–EDA–heart cut–RP-HPLC–DAD–HRMS/MS, see Publication 3, Figure 1) [81]. The change to another mass spectrometer required the development of a new HRMS/MS method. The first HRMS dimension was a full scan in polarity switching mode as usual for NTS strategies [82–84]. For fragmentation scans, four options were evaluated: all ion fragmentation, data-dependent MS2 (ddMS2) in Top5 mode, and variable or multiplexed data-independent acquisition. The advantages and drawbacks of each MS2 acquisition method were considered, and finally, the best agreement was rated with the ddMS2 mode. The newly developed ten-dimensional (10D) hyphenation provided the following information: 1) Gastrointestinal on-surface metabolisation gave evidence about lipids, sugars, and proteins and whether these macronutrients were hydrolysed by the respective enzymes (lipases, amylases, and proteases). Introduction 18 2) Metabolic changes were tracked by NP-HPTLC separation. The observed RF values also provided relative details about substances’ polarity. 3–5) The visualisation of the planar chromatogram revealed information about compounds with UV activity (UV 254 nm), native fluorescence (FLD 366 nm), or chromophores (white light illumination, Vis). 6) Screening for bioactivity was realised with one of the proposed (bio)assays as in Table 1. In the presented cases concerning the 10D hyphenation, EDA particularly targeted AChE, BChE, α-, and β-glucosidase inhibitors, as well as substances affecting the energetic cell metabolism of the A. fischeri bacteria. 7) After heart cut elution and online-desalting, an orthogonal information on substances’ polarity was obtained by means of the retention time on RP-HPLC column. 8) The wavelength scan of the DAD gave evidence about absorbance maxima, and absorbance spectra in the wavelength range of 200–400 nm. 9–10) Exact mass and fragmentation data were recorded by HRMS/MS from which a molecular formula was calculated and structural properties were derived. This workflow was applied to ten convenience tomato products in the matter of sauces or soups. The HPTLC layout allowed a side-by-side comparison of the bioactivity profile of metabolised and non-metabolised samples. Changes in bioactivity caused by intestinal metabolisation were determined with the five (bio)assays AChE, BChE, A. fischeri, α-glucosidase and β-glucosidase. The most emerging metabolic bioactivity change identified through all (bio)assays was caused by fatty acids as products of the digestion of triacylglycerols (TAGs). Lipases from the used pancreatic enzyme mixture hydrolysed the TAGs into glycerol and their respective fatty acids. Plant-based products revealed saturated and unsaturated long-chained fatty acids (≥C14), while products blended with cream or skim milk powder of animal origin, also showed short-chained fatty acids. The latter were only active in the A. fischeri bioassay. Other bioactive ingredients were found not to be affected by digestion, e.g. piperine, the main alkaloid from black pepper, which is active against the A. fischeri bacteria [81]. Beside the possibility to track metabolic changes, the most outstanding improvement of the 10D hyphenation, compared to the previous 8D version, is the opportunity to get structural information of the unknowns through tandem HRMS and thus enable an unambiguous substance assignment. Introduction 19 1.7.4. Application field study using the ten-dimensional hyphenation (Publication 4) The recently developed streamlined and information-rich 10D workflow was applied to six highly-processed differently-flavoured meal replacement products. The totality of the data obtained led to the identification of 13 substances and proved the functionality and robustness of the new NTS strategy for bioactive compounds in foods [85]. Despite the miniaturised simulation of the intestinal digestion processes, resorption is still neglected. As an example, physostigmine, a known alkaloid from plants, is cited. The compound has pronounced AChE inhibitory potential but is only poorly absorbed by humans, while its synthetic analogue rivastigmine possesses better transepithelial permeability and is commonly used as a therapeutic agent against Alzheimer’s disease [34]. Various protocols describe analytic approaches to determine pharmacokinetics and resorption [86]. Embedding an additional dimension concerning the substance uptake through membranes would complete the comprehensive workflow. Other options to gain more dimensions and associated information are multiplexed assay formats, which allow the detection of several biological effects in a single run. 1.7.5. Diffusion reduction permitting multiplexed assay formats (Publication 5) As stated already for the 8D hyphenation, the problem of analyte diffusion on NP plates still remained. Currently, planar yeast androgen/estrogen screens (pYAS/pYES), as well as their antagonistic (A) versions are performed on both plate types silica gel 60 and RP-18 W [27, 44, 69–72]. The latter is not prone to diffusion but is less sensitive compared to NP plates [22]. To overcome the diffusion susceptibility of NP plates and to further improve the workflow, several optimisations were investigated. For fixation of zones after chromatography, different coatings, i.e., poly-D-lysine, PEG 2000, PEG 8000, and a polyisobutyl methacrylate resin (Degalan), were investigated. Out of the tested agents, only Degalan showed a reduction in the observed zone diffusion. As vertical immersion of the plate in Degalan solution caused streaks when taken out and the solution was too low in viscosity to be piezoelectrically sprayable, the final coating procedure took place horizontally in a glass dish. The fixation of zones on NP plates was a precondition and crucial for the further development of the HPTLC–pYAAS bioassay procedure to a multiplex version. The bioassay pYAAS was already performed in a multiplex format to detect androgens and antiandrogens in parallel on one RP-18 W plate. Androgenic properties were determined as β-galactosidase activity, cleaving the non-fluorescent substrate Introduction 20 4-methylumbelliferyl β-D-galactopyranoside (MUG) in its fluorescent product 4- methylumbelliferone (MU) and galactose [44]. In zones containing hAR antagonists, enzyme production was not triggered and a substrate conversion was missing. Antagonistic effects are therefore detectable as fluorescence-diminishing zones in an oversprayed agonist track. The reduction in fluorescence originated either from an antagonistic effect or from absorption properties of a substance. Since those two observations cannot be distinguished, a verification (V) for true antiandrogenicity was required. Verification was achieved by applying a second stripe of the fluorescing product MU parallel to the agonist (see Publication 5, Figure 1). If the fluorescence was reduced in both stripes, an absorbing substance was assumed (false-positive). A true antagonistic effect was stated if fluorescence reduction in the agonist stripe appeared, while MU stripe was non-fluorescent at this position. This design offered two more information that were about false-positives and synergists, which locally enhanced the fluorescence of the testosterone stripe but not of the MU-stripe. As proof-of-principle, 68 botanicals were screened for androgens and antiandrogens on RP-18 W, NP, and Degalan coated NP plates (fix). The twelve samples suspected to comprise antagonists were subjected to the verification procedure. Ten zones were verified to exhibit true antiandrogenic properties, while seven synergistic zones were determined. Those zones were further evaluated with heart cut–RP-HPLC–DAD–HRMS, resulting in 29 potential candidates responsible for the observed effects. The overall NP-HPTLCfix-UV/Vis/FLD–pYAVAS–FLD–heart cut–RP-HPLC–DAD–HRMS/MS hyphena- tion resulted in twelve dimensions (12D), of which four are comprised in the multiplexed pYAVAS bioassay, namely androgens, antagonists, false-positives, and synergists [22]. 1.7.6. Application field study for multiplexed estrogen screen (Publication 6) The previously established 12D hyphenation was transferred to the closely related pYES assay for the detection of estrogens, antiestrogens, false-positives, and synergists [87]. With this multiplex assay, even additive effects were distinguishable from synergistic effects, which is hardly achievable in common approaches [88]. With a modified mathematical analogy [89, 90], those two effects are described as follows. Additive effects are observed as two effective compounds (1) summing up their signal when applied in combination (1 + 1 = 2), while synergy yielded intensified signals by the combination of a non-effective compound (0) with an effective substance (1) to the equation 0 + 1 > 1 [87]. Further optimisations were realised with the alternative green fluorescent substrate Introduction 21 fluorescein-di-(β-D-galactopyranoside). It was preferred over the blue fluorescent MUG, because botanical samples showed less green native fluorescence than blue. For the first time, it was also considered that antagonistic responses could also be caused by cytotoxic properties which could not be proven or excluded by the verification stripe. Apoptosis of the yeast cells as a result of contact with cytotoxic substances automatically prevents β-galactosidase expression and ultimately leads to a lack of substrate conversion. Cytotoxicity was determined using resazurin as a dye. Viable cells irreversibly convert the blue dye via mitochondrial processes into the pink fluorescing product resorufin (Figure 4, detectable at FLD 366 nm) [91]. Since the botanical samples contain fluorescence-diminishing pigments, cytotoxicity was evaluated in white light illumination (Vis) mode, where the positive response appeared as colourless on purple background (as depicted for the positive control menadion, Figure 4) [87]. With the 12D NP-HPTLCfix-UV/Vis/FLD–pYAVAS–FLD–heart cut–RP-HPLC–DAD– HRMS/MS hyphenation, 17 hormonal active substances could be assigned to specific molecules of which only seven were known for their biological properties towards the estrogen receptor. Of the remaining ten, four were structurally related to known phytoestrogens and the other six had not previously appeared in an endocrine context. In conclusion, the 12D hyphenation showed the potential to assign hormonal effects to individual compounds, to distinguish between different hormonal effects, and to identify unknown hormonally active substances from complex mixtures. Figure 4. Irreversible mitochondrial conversion of the blue dye resazurin into its pink fluorescent product resorufin (modified [87, 91]), and corresponding NP-HPTLCfix–planar cytotoxicity (yeast)–FLD/Vis bioautograms. Introduction 22 References 23 1.8. References [1] Y. Fu, C. Zhao, X. Lu, G. Xu. Nontargeted screening of chemical contaminants and illegal additives in food based on liquid chromatography–high resolution mass spectrometry, TrAC, Trends Anal. Chem. 96 (2017) 89–98. doi:10.1016/j.trac.2017.07.014. [2] R. Díaz, M. Ibáñez, J.V. Sancho, F. Hernández. Target and non-target screening strategies for organic contaminants, residues and illicit substances in food, environmental and human biological samples by UHPLC-QTOF-MS, Anal. Methods 4 (2012) 196–209. doi:10.1039/C1AY05385J. [3] A.M. Knolhoff, T.R. Croley. Non-targeted screening approaches for contaminants and adulterants in food using liquid chromatography hyphenated to high resolution mass spectrometry, J. Chromatogr. A 1428 (2016) 86–96. doi:10.1016/j.chroma.2015.08.059. [4] J.W. Wong, J. Wang, W. Chow, R. Carlson, Z. Jia, K. Zhang, D.G. Hayward, J.S. Chang. Perspectives on Liquid Chromatography-High-Resolution Mass Spectrometry for Pesticide Screening in Foods, J. Agric. Food Chem. 66 (2018) 9573–9581. doi:10.1021/acs.jafc.8b03468. [5] J.F. García-Reyes, M.D. Hernando, A. Molina-Díaz, A.R. Fernández-Alba. Comprehensive screening of target, non-target and unknown pesticides in food by LC-TOF-MS, TrAC, Trends Anal. Chem. 26 (2007) 828–841. doi:10.1016/j.trac.2007.06.006. [6] D.Q. Li, J. Zhao, J. Xie, S.P. Li. A novel sample preparation and on-line HPLC-DAD- MS/MS-BCD analysis for rapid screening and characterization of specific enzyme inhibitors in herbal extracts: case study of α-glucosidase, J. Pharm. Biomed. Anal. 88 (2014) 130–135. doi:10.1016/j.jpba.2013.08.029. [7] B. Da Vieira Silva, J.C. Barreira, M.B.P. Oliveira. Natural phytochemicals and probiotics as bioactive ingredients for functional foods: Extraction, biochemistry and protected- delivery technologies, Trends Food Sci. Technol. 50 (2016) 144–158. doi:10.1016/j.tifs.2015.12.007. [8] Ł. Cieśla, R. Moaddel. Comparison of analytical techniques for the identification of bioactive compounds from natural products, Nat. Prod. Rep. 33 (2016) 1131–1145. doi:10.1039/C6NP00016A. References 24 [9] M. Mladic, D.J. Scholten, W.M.A. Niessen, G.W. Somsen, M.J. Smit, J. Kool. At-line coupling of LC-MS to bioaffinity and selectivity assessment for metabolic profiling of ligands towards chemokine receptors CXCR1 and CXCR2, J. Chromatogr. B 1002 (2015) 42–53. doi:10.1016/j.jchromb.2015.08.004. [10] G.E. Morlock. High-performance thin-layer chromatography combined with effect- directed assays and high-resolution mass spectrometry as an emerging hyphenated technology: A tutorial review, Anal. Chim. Acta 1180 (2021) 338644. doi:10.1016/j.aca.2021.338644. [11] S.C. Weiss, N. Egetenmeyer, W. Schulz. Coupling of In Vitro Bioassays with Planar Chromatography in Effect-Directed Analysis, Adv. Biochem. Eng./Biotechnol. 157 (2017) 187–224. doi:10.1007/10_2016_16. [12] B. González-Gaya, N. Lopez-Herguedas, D. Bilbao, L. Mijangos, A.M. Iker, N. Etxebarria, M. Irazola, A. Prieto, M. Olivares, O. Zuloaga. Suspect and non-target screening: the last frontier in environmental analysis, Anal. Methods 13 (2021) 1876–1904. doi:10.1039/d1ay00111f. [13] J. Schwager, M.H. Mohajeri, A. Fowler, P. Weber. Challenges in discovering bioactives for the food industry, Curr. Opin. Biotechnol. 19 (2008) 66–72. doi:10.1016/j.copbio.2008.02.016. [14] J.J. Kellogg, D.A. Todd, J.M. Egan, H.A. Raja, N.H. Oberlies, O.M. Kvalheim, N.B. Cech. Biochemometrics for Natural Products Research: Comparison of Data Analysis Approaches and Application to Identification of Bioactive Compounds, J. Nat. Prod. 79 (2016) 376–386. doi:10.1021/acs.jnatprod.5b01014. [15] L.K. Caesar, J.J. Kellogg, O.M. Kvalheim, N.B. Cech. Opportunities and Limitations for Untargeted Mass Spectrometry Metabolomics to Identify Biologically Active Constituents in Complex Natural Product Mixtures, J. Nat. Prod. 82 (2019) 469–484. doi:10.1021/acs.jnatprod.9b00176. [16] R.A. Otvos, M. Mladic, G. Arias-Alpizar, W.M.A. Niessen, G.W. Somsen, A.B. Smit, J. Kool. At-Line Cellular Screening Methodology for Bioactives in Mixtures Targeting the α7- Nicotinic Acetylcholine Receptor, J. Biomol. Screening 21 (2016) 459–467. doi:10.1177/1087057115625307. [17] K.T. Kongstad, C. Özdemir, A. Barzak, S.G. Wubshet, D. Staerk. Combined use of high- resolution α-glucosidase inhibition profiling and high-performance liquid chromatography-high-resolution mass spectrometry-solid-phase extraction-nuclear References 25 magnetic resonance spectroscopy for investigation of antidiabetic principles in crude plant extracts, J. Agric. Food Chem. 63 (2015) 2257–2263. doi:10.1021/jf506297k. [18] P. Hajeb, L. Zhu, R. Bossi, K. Vorkamp. Sample preparation techniques for suspect and non-target screening of emerging contaminants, Chemosphere 287 (2022) 132306. doi:10.1016/j.chemosphere.2021.132306. [19] M.G. Weller. A unifying review of bioassay-guided fractionation, effect-directed analysis and related techniques, Sensors (Basel) 12 (2012) 9181–9209. doi:10.3390/s120709181. [20] L. Torras-Claveria, S. Berkov, O. Jáuregui, J. Caujapé, F. Viladomat, C. Codina, J. Bastida. Metabolic profiling of bioactive Pancratium canariense extracts by GC-MS, Phytochem. Anal. 21 (2010) 80–88. doi:10.1002/pca.1158. [21] S.G. Wubshet, J.S. Schmidt, S. Wiese, D. Staerk. High-resolution screening combined with HPLC-HRMS-SPE-NMR for identification of potential health-promoting constituents in sea aster and searocket--new Nordic food ingredients, J. Agric. Food. Chem. 61 (2013) 8616–8623. doi:10.1021/jf402949y. [22] T. Schreiner, A. Ronzheimer, M. Friz, G.E. Morlock. Multiplex planar bioassay with reduced diffusion on normal phase, identifying androgens, verified antiandrogens and synergists in botanicals via 12D hyphenation, Food Chem. 395 (2022) 133610. doi:10.1016/j.foodchem.2022.133610. [23] N.G.A.S.S. Chandana, G.E. Morlock. Eight different bioactivity profiles of 40 cinnamons by multi-imaging planar chromatography hyphenated with effect-directed assays and high-resolution mass spectrometry, Food Chem. 357 (2021) 129135. doi:10.1016/j.foodchem.2021.129135. [24] Á.M. Móricz, P.G. Ott, I. Yüce, A. Darcsi, S. Béni, G.E. Morlock. Effect-directed analysis via hyphenated high-performance thin-layer chromatography for bioanalytical profiling of sunflower leaves, J. Chromatogr. A 1533 (2018) 213–220. doi:10.1016/j.chroma.2017.12.034. [25] G.E. Morlock, A. Ziltener, S. Geyer, J. Tersteegen, A. Mehl, T. Schreiner, T. Kamel, F. Brümmer. Evidence that Indo-Pacific bottlenose dolphins self-medicate with invertebrates in coral reefs, iScience 25 (2022) 104271. doi:10.1016/j.isci.2022.104271. [26] G. Morlock, W. Schwack. Hyphenations in planar chromatography, J. Chromatogr. A 1217 (2010) 6600–6609. doi:10.1016/j.chroma.2010.04.058. References 26 [27] S. Buchinger, D. Spira, K. Bröder, M. Schlüsener, T. Ternes, G. Reifferscheid. Direct coupling of thin-layer chromatography with a bioassay for the detection of estrogenic compounds: applications for effect-directed analysis, Anal. Chem. 85 (2013) 7248– 7256. doi:10.1021/ac4010925. [28] T. Schreiner, D. Sauter, M. Friz, J. Heil, G.E. Morlock. Is Our Natural Food Our Homeostasis? Array of A Thousand Effect-Directed Profiles of 68 Herbs and Spices, Front. Pharmacol. 12 (2021. doi:10.3389/fphar.2021.755941. [29] I. Müller, G.E. Morlock. Validation and quantification of the saccharide release of hydrothermally treated flours after salivary and pancreatic amylolysis by the HPTLC nanoGIT+active method, in submission. [30] E. Azadniya, J. Mollergues, T. Stroheker, K. Billerbeck, G.E. Morlock. New incorporation of the S9 metabolizing system into methods for detecting acetylcholinesterase inhibition, Anal. Chim. Acta 1129 (2020) 76–84. doi:10.1016/j.aca.2020.06.033. [31] G.E. Morlock, L. Drotleff, S. Brinkmann. Miniaturized all-in-one nanoGIT+active system for on-surface metabolization, separation and effect imaging, Anal. Chim. Acta 1154 (2021) 338307. doi:10.1016/j.aca.2021.338307. [32] G.E. Morlock, J. Heil. HI-HPTLC-UV/Vis/FLD-HESI-HRMS and bioprofiling of steviol glycosides, steviol, and isosteviol in Stevia leaves and foods, Anal. Bioanal. Chem. (2020) 1–18. doi:10.1007/s00216-020-02618-4. [33] S. Agatonovic-Kustrin, D.W. Morton. High-performance thin-layer chromatography HPTLC-direct bioautography as a method of choice for alpha-amylase and antioxidant activity evaluation in marine algae, J Chromatogr A 1530 (2017) 197– 203. doi:10.1016/j.chroma.2017.11.024. [34] A. Marston, J. Kissling, K. Hostettmann. A rapid TLC bioautographic method for the detection of acetylcholinesterase and butyrylcholinesterase inhibitors in plants, Phytochem. Anal 13 (2002) 51–54. doi:10.1002/pca.623. [35] H.-R. Adhami, U. Scherer, H. Kaehlig, T. Hettich, G. Schlotterbeck, E. Reich, L. Krenn. Combination of bioautography with HPTLC-MS/NMR: a fast identification of acetylcholinesterase inhibitors from galbanum, Phytochem. Anal. 24 (2013) 395– 400. doi:10.1002/pca.2422. [36] E. Mahran, M. Keusgen, G.E. Morlock. New planar assay for streamlined detection and quantification of β-glucuronidase inhibitors applied to botanical extracts, Anal. Chim. Acta: X 4 (2020) 100039. doi:10.1016/j.acax.2020.100039. References 27 [37] S. Wangthong, I. Tonsiripakdee, T. Monhaphol, R. Nonthabenjawan, S.P. Wanichwecharungruang. Post TLC developing technique for tyrosinase inhibitor detection, Biomed. Chromatogr. 21 (2007) 94–100. doi:10.1002/bmc.727. [38] J. Taibon, A. Ankli, S. Schwaiger, C. Magnenat, V.-I. Boka, C. Simões-Pires, N. Aligiannis, M. Cuendet, A.-L. Skaltsounis, E. Reich, H. Stuppner. Prevention of False-Positive Results: Development of an HPTLC Autographic Assay for the Detection of Natural Tyrosinase Inhibitors, Planta Med. 81 (2015) 1198–1204. doi:10.1055/s-0035- 1546250. [39] C.A. Simões-Pires, B. Hmicha, A. Marston, K. Hostettmann. A TLC bioautographic method for the detection of alpha- and beta-glucosidase inhibitors in plant extracts, Phytochem. Anal. 20 (2009) 511–515. doi:10.1002/pca.1154. [40] G. Eberz, H.-G. Rast, K. Burger, W. Kreiss, C. Weisemann. Bioactivity screening by chromatography-bioluminescence coupling, Chromatographia 43 (1996) 5–9. doi:10.1007/BF02272814. [41] M. Jamshidi-Aidji, G.E. Morlock. Bioprofiling of unknown antibiotics in herbal extracts: Development of a streamlined direct bioautography using Bacillus subtilis linked to mass spectrometry, J. Chromatogr. A 1420 (2015) 110–118. [42] D. Meyer, M. Marin-Kuan, E. Debon, P. Serrant, C. Cottet-Fontannaz, B. Schilter, G.E. Morlock. Detection of low levels of genotoxic compounds in food contact materials using an alternative HPTLC-SOS-Umu-C assay, Altex (2020. doi:10.14573/altex.2006201. [43] M.B. Müller, C. Dausend, C. Weins, F.H. Frimmel. A New Bioautographic Screening Method for the Detection of Estrogenic Compounds, Chromatographia 60 (2004) 207–211. doi:10.1365/s10337-004-0315-8. [44] I. Klingelhöfer, N. Hockamp, G.E. Morlock. Non-targeted detection and differentiation of agonists versus antagonists, directly in bioprofiles of everyday products, Anal. Chim. Acta 1125 (2020) 288–298. doi:10.1016/j.aca.2020.05.057. [45] C. Weins, H. Jork. Toxicological evaluation of harmful substances by in situ enzymatic and biological detection in high-performance thin-layer chromatography, J. Chromatogr. A 750 (1996) 403–407. doi:10.1016/0021-9673(96)00601-2. [46] I.J. Purvis, D. Chotai, C.W. Dykes, D.B. Lubahn, F.S. French, E.M. Wilson, A.N. Hobden. An androgen-inducible expression system for Saccharomyces cerevisiae, Gene 106 (1991) 35–42. doi:10.1016/0378-1119(91)90563-Q. References 28 [47] D.P. McDonnell, Z. Nawaz, C. Densmore, N.L. Weigel, T.A. Pham, J.H. Clark, B.W. O'Malley. High level expression of biologically active estrogen receptor in Saccharomyces cerevisiae, J. Steroid Biochem. Mol. Biol. 39 (1991) 291–297. doi:10.1016/0960-0760(91)90038-7. [48] J.A. Buckley. Quantifying the antiestrogen activity of wastewater treatment plant effluent using the yeast estrogen screen, Environ. Toxicol. Chem. 29 (2010) 73–78. doi:10.1002/etc.11. [49] Y. Oda. Development and progress for three decades in umu test systems, Genes Environ. 38 (2016) 24. doi:10.1186/s41021-016-0054-8. [50] G. Reifferscheid, J. Heil, Y. Oda, R.K. Zahn. A microplate version of the SOS/umu-test for rapid detection of genotoxins and genotoxic potentials of environmental samples, Mutat. Res. Environ. Mutagen. Relat. Subj. 253 (1991) 215–222. doi:10.1016/0165- 1161(91)90134-T. [51] A. Marston. Thin-layer chromatography with biological detection in phytochemistry, J. Chromatogr. A 1218 (2011) 2676–2683. doi:10.1016/j.chroma.2010.12.068. [52] A. Mehl, W. Schwack, G.E. Morlock. On-surface autosampling for liquid chromatography-mass spectrometry, J. Chromatogr. A 1651 (2021) 462334. doi:10.1016/j.chroma.2021.462334. [53] T. Schreiner, G.E. Morlock. Non-target bioanalytical eight-dimensional hyphenation including bioassay, heart-cut trapping, online desalting, orthogonal separations and mass spectrometry, J Chromatogr A 1647 (2021) 462154. doi:10.1016/j.chroma.2021.462154. [54] R. Lucas-González, M. Viuda-Martos, J.A. Pérez-Alvarez, J. Fernández-López. In vitro digestion models suitable for foods: Opportunities for new fields of application and challenges, Food Res. Int. 107 (2018) 423–436. doi:10.1016/j.foodres.2018.02.055. [55] S.J. Hur, B.O. Lim, E.A. Decker, D.J. McClements. In vitro human digestion models for food applications, Food Chem. 125 (2011) 1–12. doi:10.1016/j.foodchem.2010.08.036. [56] M. Minekus, M. Alminger, P. Alvito, S. Ballance, T. Bohn, C. Bourlieu, F. Carrière, R. Boutrou, M. Corredig, D. Dupont, C. Dufour, L. Egger, M. Golding, S. Karakaya, B. Kirkhus, S. Le Feunteun, U. Lesmes, A. Macierzanka, A. Mackie, S. Marze, D.J. McClements, O. Ménard, I. Recio, C.N. Santos, R.P. Singh, G.E. Vegarud, M.S.J. Wickham, W. Weitschies, A. Brodkorb. A standardised static in vitro digestion method suitable References 29 for food - an international consensus, Food Funct. 5 (2014) 1113–1124. doi:10.1039/c3fo60702j. [57] T. Bohn, F. Carriere, L. Day, A. Deglaire, L. Egger, D. Freitas, M. Golding, S. Le Feunteun, A. Macierzanka, O. Menard, B. Miralles, A. Moscovici, R. Portmann, I. Recio, D. Rémond, V. Santé-Lhoutelier, T.J. Wooster, U. Lesmes, A.R. Mackie, D. Dupont. Correlation between in vitro and in vivo data on food digestion. What can we predict with static in vitro digestion models?, Crit. Rev. Food Sci. Nutr. 58 (2018) 2239–2261. doi:10.1080/10408398.2017.1315362. [58] S. Yoshihara, M. Makishima, N. Suzuki, S. Ohta. Metabolic activation of bisphenol A by rat liver S9 fraction, Toxicol. Sci. 62 (2001) 221–227. doi:10.1093/toxsci/62.2.221. [59] J.L. Suit, A.E. Rodgers, M. Jetten, S.E. Luria. Effects of diet on conversion of aflatoxin B1 to bacterial mutagen(s) by rats in vivo and by rat hepatic microsomes in vitro, Mutat. Res. Environ. Mutagen. Relat. Subj. 46 (1977) 313–323. doi:10.1016/0165- 1161(77)90009-7. [60] E. Debon, P. Rogeboz, H. Latado, G.E. Morlock, D. Meyer, C. Cottet-Fontannaz, G. Scholz, B. Schilter, M. Marin-Kuan. Incorporation of metabolic activation in the HPTLC-SOS- Umu-C assay to detect low amounts of genotoxic chemicals, in submission (2022). [61] C.-H. Tung, C.-C. Lin, C.-C. Tung, S.-F. Chen, F. Sheu, T.-J. Lu. Combination of on-line desalting and HPLC-UV-ESI-MS for simultaneous detection and identification of FIP- fve and flammutoxin in Flammulina velutipes, J. Food Drug Anal. 26 (2018) 1045– 1053. doi:10.1016/j.jfda.2017.12.004. [62] H. Luo, W. Zhong, J. Yang, P. Zhuang, F. Meng, J. Caldwell, B. Mao, C.J. Welch. 2D-LC as an on-line desalting tool allowing peptide identification directly from MS unfriendly HPLC methods, J. Pharm. Biomed. Anal. 137 (2017) 139–145. doi:10.1016/j.jpba.2016.11.012. [63] K.J. Fountain, M. Gilar, J.C. Gebler. Electrospray ionization mass spectrometric analysis of nucleic acids using high‐throughput on‐line desalting, Rapid Commun. Mass Spectrom. 18 (2004) 1295–1302. [64] J. Bengtsson, B. Jansson, M. Hammarlund-Udenaes. On-line desalting and determination of morphine, morphine-3-glucuronide and morphine-6-glucuronide in microdialysis and plasma samples using column switching and liquid chromatography/tandem mass spectrometry, Rapid Commun. Mass Spectrom. 19 (2005) 2116–2122. References 30 [65] P.E. Miller, M.B. Denton. The quadrupole mass filter: Basic operating concepts, J. Chem. Educ. 63 (1986) 617. doi:10.1021/ed063p617. [66] P. Eichhorn, S. Pérez, D. Barceló. Time-of-Flight Mass Spectrometry Versus Orbitrap- Based Mass Spectrometry for the Screening and Identification of Drugs and Metabolites, in: A.R. Fernández-Alba (Ed.), TOF-MS within food and environmental analysis: Comprehensive analytical chemistry, Elsevier, Amsterdam, 2012, pp. 217– 272. [67] R.A. Zubarev, A. Makarov. Orbitrap mass spectrometry, Anal. Chem. 85 (2013) 5288– 5296. doi:10.1021/ac4001223. [68] S. Eliuk, A. Makarov. Evolution of Orbitrap Mass Spectrometry Instrumentation, Annu. Rev. Anal. Chem. 8 (2015) 61–80. doi:10.1146/annurev-anchem-071114-040325. [69] C. Riegraf, G. Reifferscheid, S. Belkin, L. Moscovici, D. Shakibai, H. Hollert, S. Buchinger. Combination of yeast-based in vitro screens with high-performance thin-layer chromatography as a novel tool for the detection of hormonal and dioxin-like compounds, Anal. Chim. Acta 1081 (2019) 218–230. doi:10.1016/j.aca.2019.07.018. [70] A. Schönborn, A. Grimmer. Coupling sample preparation with effect-directed analysis of estrogenic activity - Proposal for a new rapid screening concept for water samples, JPC - J. Planar Chromat. 26 (2013) 402–408. doi:10.1556/JPC.26.2013.5.3. [71] I. Klingelhöfer, G.E. Morlock. Sharp-bounded zones link to the effect in planar chromatography-bioassay-mass spectrometry, J. Chromatogr. A 1360 (2014) 288– 295. doi:10.1016/j.chroma.2014.07.083. [72] A. Schönborn, P. Schmid, S. Bräm, G. Reifferscheid, M. Ohlig, S. Buchinger. Unprecedented sensitivity of the planar yeast estrogen screen by using a spray-on technology, J. Chromatogr. A 1530 (2017) 185–191. doi:10.1016/j.chroma.2017.11.009. [73] J. Müthing, U. Distler. Advances on the compositional analysis of glycosphingolipids combining thin-layer chromatography with mass spectrometry, Mass Spectrom. Rev. 29 (2010) 425–479. doi:10.1002/mas.20253. [74] M. Brockhaus, J.L. Magnani, M. Blaszczyk, Z. Steplewski, H. Koprowski, K.A. Karlsson, G. Larson, V. Ginsburg. Monoclonal antibodies directed against the human Leb blood group antigen, J. Biol. Chem. 256 (1981) 13223–13225. doi:10.1016/S0021- 9258(18)43031-1. [75] D. Steil, C.-L. Schepers, G. Pohlentz, N. Legros, J. Runde, H.-U. Humpf, H. Karch, J. Müthing. Shiga toxin glycosphingolipid receptors of Vero-B4 kidney epithelial cells References 31 and their membrane microdomain lipid environment, J. Lipid Res. 56 (2015) 2322– 2336. doi:10.1194/jlr.M063040. [76] N.A. Alcantar, E.S. Aydil, J.N. Israelachvili. Polyethylene glycol-coated biocompatible surfaces, J. Biomed. Mater. Res. 51 (2000) 343–351. doi:10.1002/1097- 4636(20000905)51:3<343:AID-JBM7>3.0.CO;2-D. [77] Y.H. Kim, N.S. Baek, Y.H. Han, M.-A. Chung, S.-D. Jung. Enhancement of neuronal cell adhesion by covalent binding of poly-D-lysine, J. Neurosci. Methods 202 (2011) 38– 44. doi:10.1016/j.jneumeth.2011.08.036. [78] I.D. Wilson, U.A. Brinkman. Hype and hypernation: multiple hyphenation of column liquid chromatography and spectroscopy, TrAC, Trends Anal. Chem. 26 (2007) 847– 854. doi:10.1016/j.trac.2007.07.007. [79] European Committee for Standardization. Water quality - Determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (Luminescent bacteria test): Part 1: Method using freshly prepared bacteria (2009). [80] S. Krüger, L. Hüsken, R. Fornasari, I. Scainelli, G.E. Morlock. Effect-directed fingerprints of 77 botanical extracts via a generic high-performance thin-layer chromatography method combined with assays and mass spectrometry, J. Chromatogr. A 1529 (2017) 93–106. doi:10.1016/j.chroma.2017.10.068. [81] T. Schreiner, N.M. Eggerstorfer, G.E. Morlock. Effects of gastrointestinal digestion on bioactivity of convenience tomato products studied by ten dimensional hyphenation, in submission (2022). [82] I.-L. Wu, S.B. Turnipseed, J.M. Storey, W.C. Andersen, M.R. Madson. Comparison of data acquisition modes with Orbitrap high-resolution mass spectrometry for targeted and non-targeted residue screening in aquacultured eel, Rapid Commun. Mass Spectrom. 34 (2020) e8642. doi:10.1002/rcm.8642. [83] W. Jia, L. Shi, X. Chu, J. Chang, Y. Chen, F. Zhang. A strategy for untargeted screening of macrolides and metabolites in bass by liquid chromatography coupled to quadrupole orbitrap mass spectrometry, Food Chem. 262 (2018) 110–117. doi:10.1016/j.foodchem.2018.04.090. [84] J. Wang, W. Chow, J.W. Wong, D. Leung, J. Chang, M. Li. Non-target data acquisition for target analysis (nDATA) of 845 pesticide residues in fruits and vegetables using UHPLC/ESI Q-Orbitrap, Anal. Bioanal.Chem. 411 (2019) 1421–1431. doi:10.1007/s00216-019-01581-z. References 32 [85] T. Schreiner, N.M. Eggerstorfer, G.E. Morlock. Effects of gastrointestinal digestion on bioactivity of meal replacement products studied by ten dimensional hyphenation, in submission (2022). [86] E.J. Carrasco-Correa, J. Ruiz-Allica, J.F. Rodríguez-Fernández, M. Miró. Human artificial membranes in (bio)analytical science: Potential for in vitro prediction of intestinal absorption-A review, TrAC, Trends Anal. Chem. 145 (2021) 116446. doi:10.1016/j.trac.2021.116446. [87] A. Ronzheimer, T. Schreiner, G.E. Morlock. Multiplex planar bioassay detecting estrogens, antiestrogens, false-positives and synergists as sharp zones on normal phase, Phytomedicine 103 (2022) 154230. doi:10.1016/j.phymed.2022.154230. [88] X. Zhou, S.W. Seto, D. Chang, H. Kiat, V. Razmovski-Naumovski, K. Chan, A. Bensoussan. Synergistic Effects of Chinese Herbal Medicine: A Comprehensive Review of Methodology and Current Research, Front. Pharmacol. 7 (2016) 201. doi:10.3389/fphar.2016.00201. [89] N. Rajapakse, E. Silva, Kortenkamp Andreas. Combining Xenoestrogens at Levels below Individual No-Observed-Effect Concentrations Dramatically Enhances Steroid Hormone Action, Environ. Health Perspect. 110 (2002) 917–921. [90] K.R. Roell, D.M. Reif, A.A. Motsinger-Reif. An Introduction to Terminology and Methodology of Chemical Synergy-Perspectives from Across Disciplines, Front. Pharmacol. 8 (2017) 158. doi:10.3389/fphar.2017.00158. [91] J.L. Chen, T.W.J. Steele, D.C. Stuckey. Metabolic reduction of resazurin; location within the cell for cytotoxicity assays, Biotechnol. Bioeng. 115 (2018) 351–358. doi:10.1002/bit.26475. Publication 1 33 2. Publication 1 Non-target bioanalytical eight-dimensional hyphenation including bioassay, heart-cut trapping, online desalting, orthogonal separations and mass spectrometry Tamara Schreiner, Gertrud E. Morlock* Chair of Food Science, Institute of Nutritional Science, and Interdisciplinary Research Center (iFZ), Justus Liebig University Giessen, Heinrich-Buff-Ring 26-32, 35392 Giessen, Germany Published in Journal of Chromatography A, 1647 (2021) 462154 Received 5 November 2020; Revised 22 March 2021; Accepted 8 April 2021; Available online 20 April 2021 Publication 1 34 Publication 1 35 Publication 1 36 Publication 1 37 Publication 1 38 Publication 1 39 Publication 1 40 Publication 1 41 Publication 1 42 Publication 1 43 Publication 1 44 Publication 1 45 Publication 1 46 Publication 1 47 Publication 1 48 Publication 1 49 Publication 1 50 Publication 1 51 Publication 1 52 Publication 1 53 Publication 1 54 Publication 1 55 Publication 1 56 Publication 2 57 3. Publication 2 Is Our Natural Food Our Homeostasis? Array of a Thousand Effect-Directed Profiles of 68 Herbs and Spices Tamara Schreiner, Dorena Sauter, Maren Friz, Julia Heil, Gertrud E. Morlock* Institute of Nutritional Science, Chair of Food Science, and TransMIT Center for Effect-Directed Analysis, Justus Liebig University Giessen, Giessen, Germany Dedicated to the 75th birthday of Prof. Dr. Teresa Kowalska, University of Silesia, Poland Published in Frontiers in Pharmacology, 12 (2021) 755941 Received 9 August 2021; Accepted 3 November 2021; Published 9 December 2021 Publication 2 58 Publication 2 59 Publication 2 60 Publication 2 61 Publication 2 62 Publication 2 63 Publication 2 64 Publication 2 65 Publication 2 66 Publication 2 67 Publication 2 68 Publication 2 69 Publication 2 70 Publication 2 71 Publication 2 72 Publication 2 73 Publication 2 74 Publication 2 75 Publication 2 76 Publication 2 77 Publication 2 78 Publication 2 79 Publication 2 80 Publication 2 81 Publication 2 82 Publication 2 83 Publication 2 84 Publication 2 85 Publication 2 86 Publication 2 87 Publication 2 88 Publication 2 89 Publication 2 90 Publication 2 91 Publication 2 92 Publication 2 93 Publication 2 94 Publication 2 95 Publication 2 96 Publication 2 97 Publication 2 98 Publication 2 99 Publication 2 100 Publication 2 101 Publication 2 102 Publication 2 103 Publication 2 104 Publication 2 105 Publication 2 106 Publication 2 107 Publication 2 108 Publication 3 109 4. Publication 3 Effects of Gastrointestinal Digestion on Bioactivity of Convenience Tomato Products Studied by Ten-Dimensional Hyphenation Tamara Schreiner, Naila M. Eggerstorfer, Gertrud E. Morlock* Justus Liebig University Giessen, Institute of Nutritional Science, Chair of Food Science, Heinrich-Buff-Ring 26-32, 35392 Giessen, Germany Submitted to Journal of Agricultural and Food Chemistry Received August 2022 Publication 3 110 Publication 3 111 Publication 3 112 Publication 3 113 Publication 3 114 Publication 3 115 Publication 3 116 Publication 3 117 Publication 3 118 Publication 3 119 Publication 3 120 Publication 3 121 Publication 3 122 Publication 3 123 Publication 3 124 Publication 3 125 Publication 3 126 Publication 3 127 Publication 3 128 Publication 3 129 Publication 3 130 Publication 3 131 Publication 3 132 Publication 3 133 Publication 3 134 Publication 3 135 Publication 3 136 Publication 3 137 Publication 3 138 Publication 3 139 Publication 3 140 Publication 3 141 Publication 3 142 Publication 3 143 Publication 3 144 Publication 3 145 Publication 3 146 Publication 3 147 Publication 3 148 Publication 3 149 Publication 3 150 Publication 3 151 Publication 3 152 Publication 3 153 Publication 3 154 Publication 4 155 5. Publication 4 Effects of gastrointestinal digestion on bioactivity of meal replacement products studied by ten-dimensional hyphenation Tamara Schreiner, Naila M. Eggerstorfer, Gertrud E. Morlock* Justus Liebig University Giessen, Institute of Nutritional Science, Chair of Food Science, Heinrich-Buff-Ring 26-32, 35392 Giessen, Germany Submitted to Food and Function Received August 2022 Publication 4 156 Publication 4 157 Publication 4 158 Publication 4 159 Publication 4 160 Publication 4 161 Publication 4 162 Publication 4 163 Publication 4 164 Publication 4 165 Publication 4 166 Publication 4 167 Publication 4 168 Publication 4 169 Publication 4 170 Publication 4 171 Publication 4 172 Publication 5 173 6. Publication 5 Multiplex planar bioassay with reduced diffusio