MASS SPECTROMETRIC STRATEGIES FOR THE STUDY OF PLANT EXTRACTS BIOAVAILABILITY, BIOACTIVITY AND MECHANISMS OF ACTION

1. Introduction and aim of the work Drug discovery in phytomedicine has in the past been mainly focused on the isolation and characterization of new bioactive compounds from natural products. Several NCE (new chemical entities) have been isolated from plants and they are now the active principles of many drugs able to treat and prevent different kinds of diseases. This drug discovery approach is aimed at the determination of the single "active principle" in plants, based on the assumption that a plant has one or more ingredients which determine its therapeutic effects. Beside NCE derived from plants and herbs, there is another important approach which assumes that a synergy of all ingredients of plants will bring about the maximum of therapeutic efficacy [1]. There are new forms of registered plant-derived medicines (phytomedicines) that are not single chemical entities but a complex mixture of active and inert ingredients derived form a crude extraction. However this approach has long been limited since adequate methods to standardize complex plant mixtures as well as to rationalize complex modes of actions were lacking. Moreover ADMET studies were limited due to the complexity of the phytomedicines. Hence most of the information that is usually retrieved for NCE during the drug discovery stage, such as the ADME profile and the mechanism of action was often not obtained for such complex natural derivatives, limiting their efficacy and application in therapy. Recently, thanks to the advent of novel MS techniques and to the commercial availability of high resolution MS analysers, the opportunity to determine the ADME profiles of plant extracts and to explore their mode of action has become possible. Advanced analytical techniques play an increasingly important role in the characterization, identification and quantification of plant extract compounds, not only in the context of their natural source but also in biological fluids to study their bioavailability and to discover the active compounds. Mass spectrometry has become one of the main standard techniques in this field because of the high sensitivity and specificity of the available mass analyzers. Based on these premises, the aim of my PhD work has been to set-up and apply state of the art MS strategies to better understand the mechanisms of action of some crude plant extracts and in particular tannins and rice extracts as well as to define the absorption and PK profile of cranberry and bilberry standardized extracts which are widely used as therapeutic agents. 2. Set-up and application of MS methods to elucidate biological activities and mechanisms of action of plant extracts During the first part of my Ph.D program, I have used MS strategies to investigate the ability of plant extracts to act as i) sequestering agents of reactive carbonyl species, toxic lipid peroxidation products involved in the pathogenetic mechanisms of several inflammatory based disorders and ii) as protein precipitation agents (tannin effect) using bradykinin, a pro-inflammatory mediator, as protein target. 2.1 Set-up of an isotopic labelling procedure for the characterization of HNE-sequestering agents in natural extracts and its application for the identification of anthocyanidins in black rice giant germ Reactive carbonyl species (RCS) are cytotoxic molecules deriving from the oxidation of sugars and lipids. When the human system undergoes stress condition, the physiological detoxification pathways are not efficient enough to inhibit these reactive molecules. RCS are electrophilic compounds that can easily react with the nucleophilic centers of proteins leading to the generation of Advanced Glycation End Products (AGEs) and Advanced Lipoxidation End Products (ALEs) [2]. These products are involved in many chronic diseases like diabetes, atherosclerosis and neurodegeneration [3]. Plant extracts represent a source of active compounds with a potential RCS quenching ability. The mass spectrometric method we have recently developed [4] was used to screen the ability of water-soluble rice extracts to act as inhibitors of RCS-induced protein carbonylation. Ubiquitin, used as a protein model, was incubated with one of the most reactive RCS, 4-hydroxy-trans-2-nonenal (HNE), and with increasing concentrations of these extracts for 24 h. The amount of modified ubiquitin was determined by HRMS. The inhibitory effect (IC50) of the different extracts was evaluated by measuring the extent of ubiquitin modification in the absence and the presence of the inhibitors. Black rice with giant germ proved to be the most active (IC50 ≈ 29.1 mg/mL). An isotopic signature profile method was applied for the identification of the putative bioactive compounds: the strategy takes advantage of the characteristic isotopic ion cluster produced by the mixture of HNE:2H5-HNE mixed at a 1:1 stoichiometric ratio. The identification of possible bioactive components was obtained by using available databases. Among the list of components identified some anthocyanidins and some aminoacids were present. The capacity of these compounds to act as HNE sequestering agents was tested by HPLC-UV analysis, measuring free HNE not quenched by the molecules tested. The results were expressed as carnosine units. These results support the hypothesis that certain bioactive components sequester RCS and this kind of rice extract can be adopted in dietary strategy for health benefits. 2.2 Development of a direct ESI-MS method to measure the tannin precipitation effect of proline rich peptides Tannins are a heterogeneous class of polyphenols that are present in several plants and foods [5]. Their health benefits, such as antidiarrheal activity [6], are due to their ability to precipitate protein. Thus, it is important to evaluate tannin content in plant extracts to evaluate their potential use as pharmaceutical and nutraceuticals. The purpose of the present work is to set up a suitable method able to quantify the extent of tannin protein precipitation extent. Bradykinin (100 nmoles/mL), chosen as a model, was incubated with increasing concentrations of 1,2,3,4,6-pentagalloyl β-D-glucose and tannic acid chosen as reference of tannic compounds. Bradykinin not precipitated by tannins was determined by a mass spectrometer TSQ Quantum Triple Quadrupole equipped with an ESI source (direct infusion analysis). The results were expressed as PC50 (112.3 µM and 84.6 µM for tannic acid and 1,2,3,4,6-pentagalloyl β-D-glucose respectively). The type of tannin-protein interaction was also evaluated after precipitate solubilization. The involvement of proline residue in tannin-protein interaction was confirmed by repeating the experiment using a synthetized peptide (RR-9) characterized by the same bradykinin sequence, but having proline residues replaced by glycine residues: no interaction occurred between the peptide and tannins. 3. MS methods for the study of the ADME profiles of standardized cranberry and bilberry extracts ADME profile of plant extracts is a quite challenging approach due to the multicomponent nature and chemical complexity of such extracts, their various concentrations across a wide range, the complexity of their interactions and their complex degradation dynamics in vivo. Due to the complexity of both botanicals and biological samples (e.g., blood, urine, tissues), the analytical approaches to monitor the time-dependent concentration profiles of bioavailable plant molecules require a high sensitivity and specificity. The advent of high resolution MS analyzers has fulfilled these requirements, permitting their application in studying ADME profiles of plant extracts. In the second part of my Ph.D thesis I have applied MS techniques to study the ADME profile of cranberry and bilberry extracts in humans and rodents, respectively. I have used both on target and off target MS methods which have permitted a better understanding of the absorption and PK profile of these natural extracts. 3.1 Profiling Vaccinium Macrocarpon components and metabolites in human urine and urines ex-vivo effect on the reduction of C. Albicans adhesion The activity of Cranberry (Vaccinium Macrocarpon) in the prevention of Urinary Tract Infection (UTIs) has long been reported [7]. Controversial results concerning the bioactive compounds are due to the use of different dosages and non-standardized Cranberry extract treatments. In vitro studies reported the activity of PAC-A2 [8], but its involvement as the bioactive molecule is uncertain since it is not detected in human urine [9]. In this work a standardized Cranberry extract (Anthocran™) was administered to four healthy volunteers at a dosage found to be active in human studies (2 capsules/day of Anthocran™, 36 mg proanthocyanidins/capsule) for 7 days. The volunteers followed a diet poor in polyphenols 72 hours before and during the treatment. The urine samples were collected before the beginning of the capsule intake and 1, 2, 4, 6, 10, 12 and 24 hours after the last assumed. An HPLC-MS/MS method was set up for the analyses using an LTQ-XL-Orbitrap mass spectrometer. Two different types of data analyses were performed: on target and off target. The on target data analysis was followed by creating a database containing all the Cranberry extract components and these components were added to other compounds found in literature. The off target strategy consisted of searching all the ions not present or present at an intensity relative to noise in the pre-treatment sample. CFM-ID (competitive fragmentation modeling for metabolite identification) was used for the identification of the unknown ions. These strategies allowed the identification of 35 analytes including Cranberry components, known metabolites and also metabolites hitherto unreported in the literature. Urine collected at different time ranges after the last dosage of Anthocran™ were ex-vivo tested on the reduction of C. Albicans adhesion. Fractions collected after 1 and 12 hours were found effective, significantly reducing adhesion compared to the control (p < 0.001). Purified Anthocran™ components and metabolites identified in the two active urine fractions will be tested. 3.2 Pharmacokinetic profile of bilberry anthocyanins in rats and the role of glucose transporters: LC-MS/MS and computational studies Anthocyanins are pigments widely present in plants and foods and they belong to the flavonoid class. Their health benefits such as antioxidant, anti-inflammatory and anticarcinogenic properties are well reported [10-12], leading to a potential use of these molecules in oxidative-based diseases. Several studies have shown that they are absorbed both in the stomach and in the small intestine [13]. One mechanism of absorption proposed there is the GLUT transporters, a hypothesis supported by an in vitro study on Caco-2 cells [14]. The purpose of this work was to better understand the role of GLUT transporters (GLUT2 and sGLT1) in anthocyanins absorption. A quantitative HPLC-MS/MS method was set up to determine the absorption of 15 anthocyanins present in a standardized bilberry extract (Mirtoselect®, 36% anthocyanins) in rats. The oral treatment was performed under fasted and fed conditions and, in fasting conditions with the co-somministration of glucose. The first group (Group 1) did notundergo any diet or starvation and was treated with Mirtoselect® 100 mg/kg. The second and the third groups were only allowed to drink water 18 h before the treatment and then they were administered respectively with Mirtoselect® 100 mg/kg (Group 2) and Mirtoselect® 100 mg/kg with 1 g/kg glucose (Group 3). The pharmacokinetic parameters were calculated with PK solver add-in program for Microsoft Excel) AUC, TMAX and CMAX. Computational studies were performed to fully understand the molecular recognition of anthocyanins by GLUT2 and sGLT1, to explain the different absorption of anthocyanins found in PK studies and to investigate the electrical form involved in the mechanism of transport. The interaction between anthocyanins and GLUT2 and sGLT1 are quite similar and generally, in the case of anthocyanins, involved: i) H-bonds from all sugar hydroxyl functions, ii) π-π stacking and H-bond from the 5H-chromen-5-one, iii) H-bonds from the 3,4-dihydroxyphenyl moiety. The two simulated transporters yield comparable models, so both proteins might be similarly involved in anthocyanin absorption. The key difference is the greater flexibility of GLUT2 which seems to be able to recognize all the electrical forms tested. 4. Conclusions In conclusion, the present Ph.D work has demonstrated that mass spectrometric strategies based on a high resolution MS analyzer can be successfully applied to better characterize the biological activities and PK profiles of natural extracts characterized by a complex composition. Hence, information usually obtained for pure compounds in the discovery stage can also be retrieved for more complex bioactive products such as plant extracts. Plant extracts which nowadays find a wide use as health products should be better characterized in terms of biological activity and PK thus assuring more efficacy and safety and the fact they are a complex matrix should not limit this information. References [1] Ulrich-Merzenich G, Panek D, Zeitler H, Vetter H, Wagner H, Indian J Exp Biol. 2010, 48: 208-19. [2] Vistoli G., De Maddis D., Cipak A., Zarkovic N., Carini M., Aldini G., Free Radic Res. 2013, 47: 3-27. [3] Nedić O., Rattan S.I., Grune T., Trougakos I.P., Free Radic Res 2013, 47: 28-38. [4] Colzani M., Criscuolo A., De Maddis D., Garzon D., Yeum K.J., Vistoli G., Carini M., Aldini G., J Pharm Biomed Anal. 2014, 91: 108-118. [5] Serrano J., Puupponen-Pimiä R., Dauer A., Aura A.M., Saura-Calixto F., Mol Nutr Food Res. 2009, 53: S310-29. [6] Qin Y., Wang J.B., Kong W.J., Zhao Y.L., Yang H.Y., Dai C.M., Fang F., Zhang L., Li B.C., Jin C., Xiao X.H., J Ethnopharmacol. 2011, 133: 1096-102. [7] Vasileiou, I., Katsargyris A., Theocharis S., Giaginis C., Nutr Res. 2013, 33 : 595-607. [8] Howell, A.B., Vorsa N., Der Marderosian A., Foo L.Y., N Engl J Med. 1998, 339 : 1085-6. [9] Valentova, K., Stejskal D., Bednar P., Vostalova J., Cíhalík C., Vecerova R., Koukalova D., Kolar M., Reichenbach R., Sknouril L., Ulrichova J., Simanek V., J Agric Food Chem. 2007, 55 : 3217-24. [10] Zafra-Stone S., Yasmin T., Bagchi M., Chatterjee A., Vinson J.A., Bagchi D., Mol Nutr Food Res. 2007, 51: 675-83. [11] Joseph S.V., Edirisinghe I., Burton-Freeman B.M., J Agric Food Chem. 2014, 62: 3886-903. [12] Lin B.W., Gong C.C., Song H.F., Cui Y.Y., Br J Pharmacol. 2016,174:1226-1243. [13] He J., Wallace T.C., Keatley K.E., Failla M.L., Giusti M.M., J. Agric. Food Chem., 2009, 7: 3141-3148. [14] Zou T.B., Feng D., Song G., Li H.W., Tang H.W., Ling W.H., Nutrients. 2014,6: 4165-77.