TBD

The discovery of a new drug is a challenging process that develops through a series of subsequent stages both in the discovery and development phase. In the discovery phase, for projects aiming at small molecule therapeutics, the process starts with the identification of hits that can be evolved to quality lead compounds, which will be the starting point for obtaining one or more preclinical development candidates. In the process referred to as Hit-to-Lead, hits are usually modified in order to increase potency on the target and build selectivity against possible off-targets. In the last two decades, however, it has been increasingly recognized that quality leads should also possess drug-like properties in order to avoid problems, or even failure, in the development phase. To achieve this goal, it is important that the compounds undergo characterization of drug-like properties such as solubility, stability in biological fluids and metabolic stability already at the Hit-to-Lead stage. More extensive evaluation of drug-likeness will be conducted during Lead Optimization in order to identify the most suitable compound(s) to progress to preclinical development. The assessment of compounds’ drug-likeness requires testing them in a number of different in vitro assays, followed eventually by dosing in animals to determine pharmacokinetics. To do so, it is necessary to develop proper analytical methods to detect compounds and measure their concentration in various media and biological fluids. The results of those assays will help the medicinal chemists in the discovery of compounds with the highest chance to become a drug. The lecture will provide an overview of the process from hits to drugs with a focus on the discovery stages.

The primary goal of any quality system is to provide the framework and methodology to ensure that the data are fit for purpose and to give assurance that the process under which the data have been generated are completely accountable, transparent, and reproducible.
There has been increasing evidence in recent years that research in life sciences is lacking in reproducibility and data quality. This raises the need for effective systems to improve data integrity in the evolving non-GxP research environment. The entire landscape of research is changing radically. Innovation, the lifeblood of scientific endeavour, is very costly and standards in research that were the norm yesterday will not be acceptable in the increasingly regulated and competitive world of tomorrow.

Thus, a non-GxP Research standard should focuses on data integrity and research reproducibility. The rigor and frequency of its application need to be adapted to the research phase to which it is applied: in early discovery, focus is laid on innovation, protection of intellectual property and data integrity.

The aim of this presentation is to provide the basics for implementing a quality management system in a non-regulated research organisation in order to guarantee Data Integrity and Data Quality. The critical elements that need to be considered to ensure a successful implementation of research quality standards in both industry and academia are presented and discussed. The quality standard proposed is founded on data integrity principles named ALCOA+ and good research practices and contains basic quality system elements, which are common to most laboratories. A pragmatic and risk-based quality system and associated assessment process is presented to ensure reproducibility and data quality of experimental results while making best use of the resources.

The entire process of introducing a new drug to the market can be considered as a race against time: it typically takes more than 10 years for its discovery and development, and only a few years are available for the return of the investments, i.e. the time lapse from market introduction to expiry of the patent application. In addition, the total cost for the development of a new drug is very high (averaging to 2.6 million USD, with an increase of 8.5% per year),¹ and so is the risk of failure for a candidate drug entering in clinical trials: the overall likelihood of approval starting from Clinical Phase I over the last decade is only 7.9%.² The above aspects take to a very competitive landscape among companies involved in drug discovery, and to the need for methods and technologies aimed at improving overall efficiency and reducing risks.

Perhaps the main issue is the still unmet need to reduce the attrition rate of potential new drugs entering the clinic, mainly dealt with the careful selection of the strongest candidates during the phases of lead generation and optimization, and the acquisition of early knowledge of the potential critical points of a research program. This in fact contributes in preventing the high costs of a drug candidate failure during clinical phase, and in concentrating resources on the projects with the most probable success.

In this effort, after a period starting from the early 2000s, during which the expectations from introduction of combinatorial chemistry and high throughput screening had not been followed by significant results, Medicinal Chemistry has essentially returned to more traditional synthetic methods, but with improvements gained from the combinatorial experience.³ The latter consist not only in changes of synthesis philosophy (use of robust chemical reactions amenable to diverse functionalizations, more attention to drug-like physico-chemical properties, synthesis of single compounds or small and more focused libraries), but also, end even more, in product purification and analysis, often constituting a previous significant time constrain. The main changes introduced can be summarized as follows:
• Use of information-rich and specific analytical techniques, to minimize analytical workload while maintaining full information on sample identity and purity directly from the crude reaction products: LC-MS (or-high resolution MS), high-field NMR.
• Increased use of open access instrumentation: the synthetic chemist analyzes his own products with standardized methods and automated software, minimizing the return time of results.
• Use of automated and integrated platforms for sample analysis and purification, such as mass-directed preparative HPLC.

All these aspects significantly contribute in speeding up the delivery of products for pharmacological screening, as well as generation of lead compounds and small-scale drug candidates.

This presentation will give an overview of the current analytical techniques in Medicinal Chemistry, particularly LC-MS, largely considered as the workhorse in this field, and its integration with other techniques, with practical examples and applications.

References
1. J.A. DiMasi et al., J. Health Economics, 47 (2016) pp. 20-33. DOI:10.1016/j.jhealeco.2016.01.012
2. D. Thomas, D. Chancellor, S. Chaudhuri et al., Clinical Development Success Rates and Contributing Factors 2011–2020, https://pharmaintelligence.informa.com/~/media/informa-shop-window/pharma/2021/files/reports/2021-clinical-development-success-rates-2011-2020-v17.pdf, accessed 15/07/2021.
3. I. B. Campbell et al., Medicinal chemistry in drug discovery in big pharma: past, present and future, Drug Discovery Today 23 (2018) pp. 219-234. DOI: 10.1016/j.drudis.2017.10.007

NMR spectroscopy is now well recognized as a powerful tool in the hit identification and hit-to-lead optimization phases of drug discovery projects. Although currently NMR cannot compete with other biophysical techniques in term of throughput, NMR has some distinctive features that make it very appealing. The rich information content obtained with NMR, combined with the reliable identification and binding affinity measurement of the hits, avoid squandering time and resources in the pursuit of unsuitable compounds. Furthermore, the possibility of detecting very weak affinity ligands, makes it particularly suitable in Fragment Based Drug Discovery (FBDD) projects and in tackling difficult biological targets with low ligandability. Several observable NMR parameters can be used for identifying and characterizing the interactions of a ligand with the receptor. Most of the experiments utilize 1H NMR spectroscopy. However, it has been shown that improvements in the detection and efficiency can be achieved utilizing 19F NMR spectroscopy.¹ These 19F NMR-based screening methodologies can be classified in three groups based on the detected signals. (i) The protein-based method, known as PrOF (Protein Observed Fluorine),² utilizes proteins labelled with selected fluorinated amino acids and monitors chemical shift perturbations of the protein 19F NMR signals in the presence of a binder. (ii) The ligand-based method, known as FAXS (Fluorine, chemical shift Anisotropy and eXchange for Screening)1, monitors changes in transverse relaxation and/or chemical shift of the ligand 19F NMR signal(s). It can be performed in direct or competition format. The former requires a library of fluorinated molecules whereas the latter requires only one fluorinated suitable reporter molecule. (iii) The substrate-based method, known as n-FABS (n Fluorine Atoms for Biochemical Screening),³ is a functional assay and utilizes substrates carrying a fluorinated group. It measures the area of the 19F NMR signals of the substrate and product of the enzymatic reaction. Its flexibility allows a broad range of screening applications, from purified enzymes to intact living cells. The principles of these methodologies will be presented along with selected applications in drug discovery projects.

References

1. C. Dalvit and A. Vulpetti J. Med. Chem. 62, 2218-2244 (2019). DOI: 10.1021/acs.jmedchem.8b01210
2. K.E. Arntson and W.C.K. Pomerantz J. Med. Chem. 59, 5158–5171 (2016). DOI: 10.1021/acs.jmedchem.5b01447
3. C. Dalvit, M. Veronesi and A. Vulpetti J. Biomol. NMR 74, 613-631 (2020). DOI: 10.1007/s10858-020-00311-3

Protein kinases are enzymes that regulate biological activities by pinpoint phosphorylation of specific amino acid residues of key proteins. Through this process, they modulate multiple cellular functions including growth and differentiation. Abnormal activities in this set of enzymes can result in diseases and can constitute the driver for tumor insurgence and progression. Kinases share high homology in the ATP-binding pocket; however, the presence of subtle differences can be exploited to obtain chemical entities with high potency and selectivity toward a single member within a kinase family. In recent years, several kinase inhibitors reached clinical approval in oncological target therapy thanks to a vast plethora of assays that have been developed to accelerate the discovery of new lead candidates.

At NMS, the road towards the development of the optimal lead candidate begins with the identification of a target kinase acting as a driver in a specific tumor context. The selected target protein is then expressed, purified and characterized for its activity. A crucial step of the process consists in the development of a screening assay that can reliably quantify target inhibition. The assay is then fully automated and a high-throughput screening (HTS) campaign on proprietary chemical compound libraries is run to identify a primary hit for the target. After the initial hit identification, a careful work of chemical optimization is needed to improve compound potency, selectivity and ADME properties. During development, a special focus is given to the selectivity within the kinome tree. Small molecules are routinely screened against a platform encompassing over 100 kinases to determine cross-reactivity and define a comprehensive selectivity profile. Orthogonal methodologies such as enzyme activity and inhibition studies, homogeneous time-resolved fluorescence assay (HTRF), surface plasmon resonance (SPR), differential scanning fluorimetry (DFS) and bioluminescent resonance energy transfer (nanoBRET) are all used to gain information on the compound under investigation and to characterize different properties of the lead candidate. In this presentation different examples of how hit compounds are developed will be presented, focusing especially on how selectivity within the kinase enzyme family is explored and how different technologies are applied to attain a lead candidate that can further advance in the drug development process.

The early process of drug discovery can be simplified in two major approaches, depicted in the figure below:

1. a target-based in silico approach, which requires reliable structural information on the protein target;
2. a ligand based HTS approach, which uses chemical compound collections.

At Axxam we have implemented a state-of-the art facility for the HTS-based approach, with several diversified automated screening stations, including robotic devices for liquid handling and plate assembling and readers for different types of measurements, such as: luminescence, fluorescence, FRET, current amplitude, gene expression, image analysis. With both the in-silico and HTS approaches, the Hit-to-Lead phase begins once the screening campaign has been completed along with the identification and validation of the most promising positive hits.

The aim of the Hit-to-Lead phase is to refine each hit series towards more potent and selective compounds, possessing sufficient properties to be able to examine their efficacy in the next step of in vivo preclinical testing.

This preclinical phase typically consists of intensive Structure-Activity Relationship (SAR) investigations around each core compound structure, with assays carried out to establish the magnitude of activity (efficacy and efficiency) and selectivity of each compound.

The effectiveness of an HTS campaign substantially depends on the quality of the assays and the compound collection used: biology and chemistry must work in harmony! Axxam has a strong and consolidated expertise in the early drug discovery phases from HTS to Hit-to-Lead, concerning the biology aspects.

The primary assay, used in HT and focused on the molecular target, must achieve the quality parameters of meaningfulness, reliability, and robustness, but should also be low cost and suitable for miniaturization, to allow for a reliable and sustainable screening of hundreds of thousands of molecules. On the other side, the assays used in the Hit-to-Lead phase can be more expensive and adaptable to a lower throughput, but they should be as close possible to the native situation and should provide as much information as possible, to validate the pharmacological improvement of the chemical entities.

In my lecture I will describe examples of cell-free and cell-based in vitro assays, developed in our laboratories, for different classes of targets and different final purposes, used in the HTS and Hit-to-Lead phases, as well as how we handle the flow of activities, to guarantee the best turnaround for generating the results.

The high attrition rate during the drug development phase is still representing a key challenge to address for the pharmaceutical industries.
Poor pharmacokinetic (ADME) profiles of the lead compounds was one of the major causes of failure in Phase I clinical trials. In the recent years it seemed that ADME issues were significantly addressed although the lack of efficacy and /or toxicity of the novel chemical entities is nevertheless the primary source of trial failure.
In the last two decades we assisted in a profound paradigmatic shift in the drug discovery process which looked at the optimization of most of the ADMET parameters very early throughout the discovery process by using several surrogate in vitro assays leading to the so called multi-parametric lead optimization strategy.
There are several assays that can be run in different formats during the lifetime of a drug discovery project. However, there is no a rule of thumb indicating which tests and at which stage have to be applied. Nevertheless, there are set of physico-chemical and biochemical assays that are generally though essential to be incorporated very early in most of the drug discovery projects.
A review of the largely used in vitro ADMET assays will be discussed describing advantages and possible issues associated with the interpretation of the associated data. Suggestions on ADME assays to be employed during lead optimization and pre-IND phase will be discussed also in the light of the most recent FDA and EMA guidelines.

Drug Metabolism and Pharmacokinetics (DMPK) is a fundamental aspect within the drug discovery and development process. The key role of DMPK in drug discovery is to participate in the design and selection of the lead compound by balancing the DMPK properties associated with potency, selectivity and safety. Furthermore, DMPK gives additional insight into the processes that control the absorption, distribution, metabolism and excretion (ADME) of the drug candidate (and its metabolites) with the aim of providing a mechanistic understanding of the pharmacokinetics, pharmacodynamics and safety of the drug candidate.

Liquid chromatography/mass spectrometry (LC/MS) is the reference analytical tool used in DMPK studies thanks to its selectivity, low sample volume requirements, sensitivity, and speed. Since the introduction of Atmospheric Pressure Ionization (API) sources such as Electrospray Ionization (ESI), LC-MS is the most used analytical tool for both quantitative and qualitative DMPK analyses. The aim of the seminar is to go over the main methodological and instrumental MS solutions adopted in recent years in DMPK studies. In particular, analytical strategies adapted for metabolite identification and characterization based on triple quadrupole MS analyzers such as precursor ion and neutral-loss experiments will be shown, together with the applications of ion traps for MSn and high-resolution MS systems (HRMS). Furthermore, chemical strategies based on derivatization, isotopic fingerprinting and H/D exchange experiments will be presented.

LC-MS strategies for absolute- and semi-quantitative analysis will also be discussed. In particular, applications of LC-Triple quadrupoles (TQMS), which became the workhorse for quantitative bioanalysis in the 1990s, will be considered, together with the most recent advantage of hybrid systems such as quadrupole-TOF and quadrupole-orbitrap which, beside being the gold standard for qualitative analyses, now also offer the right performance for quantitative analyses, thanks to sensitivity, dynamic range, resolution, accuracy and scan‐to‐scan reproducibility, making them a worthwhile alternative to TQMS systems.

Finally, analytical methods to predict compounds which can potentially induce idiosyncratic reactions by forming protein covalent adducts will be discussed. Also in this regard different MS strategies will be discussed such as the neutral loss scanning based on a triple quadrupole mass spectrometer which was the first LC/MS method developed for detecting reactive metabolites trapped by GSH in studies of bioactivation of drugs or other xenobiotics.

Methodological insights together with application and case studies will be considered in the seminar.

Metabolite identification plays a crucial role in drug development, leading to a constant need to evaluate new technologies that can improve data quality or increase efficiencies. The metabolite identification challenges that technologies such as IMS can address depend on the drug development stage.

Liquid chromatography–mass spectrometry (LC–MS) helps to address these challenges by separating compounds in mixtures using two dimensions, namely retention time (RT) and mass-to-charge ratio (m/z). Careful and reproducible analytical measurement of both properties usually allows the DMPK scientist to determine what overall biotransformations have occurred and what the structures look like over many samples or studies. Different acquisition methods are also employed, such as data dependent and data-independent acquisition (DDA and DIA, respectively). DIA methods, such MSE rapidly alternates between full scan low collision and high collision energies, allowing both precursor (drug and metabolite) and product ion high resolution accurate mass measurements to be obtained. IMS has now evolved to afford an added dimension of separation to metabolite identification, based on an ion’s size and shape, along with an accurate mass-to-charge ratio, as it passes through a gas such as nitrogen. This property can be used to generate a collision cross section (CCS) value, correlated to its drift time (DT), that describes the ion’s overall size (influenced by structure, charge, conformation, and interaction with the gas). The development of IMS is decades-old, but its availability in combination with commercialized LC–MS instruments has led to increased interest in its relevance for a range of applications, including metabolite identification. Ion mobility supported on commercial LC–MS platforms now includes traveling wave ion mobility (TWIMS), drift tube ion mobility (DTIMS), and trapped ion mobility (TIMS), all of which generate CCS values that can be stored in customized libraries. Spectral clarity is improved by using DT to align retention time data and filter out nonpertinent ions, which is especially important when looking for metabolite fragment ions in biological matrices of preclinical toxicity models. In addition, the precise CCS value, which is matrix and ion-concentration independent, can be used to track isomeric metabolites across chromatographic conditions and resolve coeluting compounds throughout the entire drug development life cycle. This ability to more confidently detect, characterize, and track metabolites across time and geographies without worrying about varying chromatographic conditions greatly eases the burden of the biotransformation scientist.

LC–MS in combination with IMS generates high-quality data that directly impacts the confidence of later conclusions; however, software is needed to easily manage, analyze, and store these complex data sets efficiently. Fortunately, several solutions have been developed that incorporate IMS into software packages at each stage of development and across molecule types.

Conclusion. New technologies for metabolite identification workflows, such as ion mobility, are needed to generate the best data, within the needed time frame, at the appropriate phase of the drug discovery and development process. IMS delivers this capability by adding another separation dimension that increases confidence in identification and structural elucidation, distinguishes coeluting metabolites, and tracks metabolites across different matrices and chromatographic conditions using CCS values. When combined with the effective software, ion mobility can improve efficiencies and workflows to deliver real value to the metabolite identification laboratory.

In the hit-to-lead drug discovery stage, several experiments for ADME assessment such as preliminary PK, tissue distribution in rodents and plasma protein binding require compound quantitation in biological matrices. The preferred analytical technique for small molecules is LC-MS/MS. The method should be developed with adequate sensitivity, selectivity and linearity even though often without the use of Internal Standard (IS). A tiered approach is adopted based on the method application.

A dedicated time-saving workflow supports the high-throughput preliminary PK assessments service available at Accelera: a generic protein precipitation sample preparation combined with a column switching/mobile phase test is used for chromatographic method development. Chromatographic methods with isocratic/ gradient elution and gradient wash, run times< 5 min (or <4 min with UHPLC) and k’>2 are selected. Moreover, the use of surrogate matrix and reduced matrix volume for extraction comply with the 3Rs rules.

The LC-MS/MS quantitative analytical methods that support ADME in vitro studies such as plasma Protein Binding (PPB) are often in the latest lead optimization stage: the LLOQ of the method must be adjusted to be suitable to determine the free drug at the lowest concentration tested in multispecies equilibrium dialysis experiments. LC-MS/MS quantitative analytical methods are also applicable to tissue distribution studies performed in rodents during the discovery phase. Homogenizers, sonicators, microdismembrators, hydrolysis are used for sample preparation of matrices such as brain, lung, liver, eyes, intestine and tumor. At Accelera a bead beaten homogenizer (Precellys Evolution) is used to prepare 24 samples in parallel, in a rapid (few minutes) and easy way. This technique is routinely applied for mouse and rat brain drug quantification alongside plasma preliminary PK. It is also applied successfully to tissues like kidney, skin, heart and liver. Complex tissue matrices might need a higher specificity of the analytical method to prevent potential interferents from affecting the determination of the analyte.

Case studies of methods developed in our laboratories to support potential anti-cancer drug candidates in PK, PPB and tissue distribution experiments are presented. Triple quadrupole instruments API5000, API5500 and API6500 coupled with HPLC were used for analyte determination within different concentration ranges. The whole bioanalytical workflow for sample and data management, from the study design to reporting, was supported by Watson LIMS software.

Bioanalysis in support to the drug discovery processes has specific peculiarities that are different from those related to the drug development support. Rapid turnaround, hundreds of compound to be tested per month, and the absence of any previous information on stability, in vivo exposure, analytical challenges and metabolism represent a unique challenging situation where speed of analysis should meet reliability of the data. The lack of a proper internal standard represent another important limitation for LC-MSMS methods that may be more susceptible of matrix effect issues. In this presentation it will be described a general overview of the sample preparation techniques, the matrix effect common issues and how to deal with them, the method development approach and metabolism aspect to be considered when using different ionization techniques.
In the last decades the number of biologics and multi domain drugs such as ADCs in the discovery pipeline of BioTech and Pharma companies increased significantly posing new challenges for better driving the drug design process. At the same time LCMS technology continued to evolve improving the instrument sensibility and enabling the application of this technology platform to the bioanalysis of PK studies as an alternative to the well-established ELISA. In particular the possibility to develop generic assays without the need of specific reagents together with the multiplexing capability of the LCMS is an attractive approach in the Drug Discovery environment. An overview of the therapeutic proteins quantitative analysis using LC-MS/MS will be also discussed.

Designing and selecting appropriate lead candidates is key in reducing clinical attrition and costs associated with the development of new drugs. Early formulation work is performed across these phases ensuring a smooth transition between Discovery and Development phases. In this contest analytical support is focused mainly into solid form optimization screening, solubility and stability profiling and on overall biopharmaceutical evaluation to select the appropriate attributes for identified best lead candidates. This includes not only experimental protocols, appropriate criteria selection but also reinforced interactions and close communication between department experts for data cross –review and ensure key decision making and mitigation plans established at the stage of candidate entry into development“.

Small-molecule drug discovery remains an enormous challenge in which various compounds characteristics, as the potency, selectivity, in vivo efficacy and safety, need to be fine-tuned to provide novel active pharmaceutical ingredients. To this aim, the ability to rapidly prepare compound collections for medicinal chemistry learning cycles and the capacity to optimize processing methods of lead products for clinical studies, are crucial for the success rate of drug discovery programs.¹ In this context, continuous flow chemistry and related technologies are considered valuable tools to solve synthesis limitations in terms of compound throughput, quality and scalability, and have shown great potential in uncovering and developing novel drug candidates.²
In this lesson, case studies on the use of enabling chemical technologies to expedite hit-to-lead explorations and optimize chemical processes are discussed. The scope is to illustrate how innovative approaches as flow chemistry and automation can be exploited for streamlined compound synthesis and characterization. Particular emphasis will be given to the integration of flow synthesizers with process control devices, in-line analysis, and downstream operations (Figure 1) to realize systems capable of reaction optimization, multistep continuous synthesis, and parallel/sequential compound library preparation, that can be coupled with bio-assays and predictive computational tools and accelerate medicinal chemistry discovery cycles.

Figure 1

References
1. Blakemore, D. C.; Castro, L.; Churcher, I.; Rees, D. C.; Thomas, A. W.; Wilson, D. M.; Wood, A. Organic synthesis provides opportunities to transform drug discovery. Nat. Chem. 2018, 10, 383−394. DOI: 10.1038/s41557-018-0021-z
2. Gioiello, A.; Piccinno, A.; Lozza, A. M.; Cerra B. The Medicinal Chemistry in the Era of Machines and Automation: Recent Advances in Continuous Flow Technology. J. Med. Chem. 2020, 63, 6624−6647. DOI: 10.1021/acs.jmedchem.9b01956

Laquinimod is a medicine developed by Active Biotech and Teva for the treatment of multiple sclerosis (MS) and additional indications of other autoimmune and neurodegenerative diseases.

As a part of development, several polydeuterated analogues of Laquinimod were prepared. During analytical method development we observed noticeable reduction of retention time with increase of deuterium atoms in the molecule. D₅-Laquinimod (five hydrogen atoms replaced with deuterium) elutes about 0.2 min earlier than the non-deuterated (RT is about 15.7 min), while D₁₆-Laquinimod (sixteen hydrogen atoms replaced with deuterium) elutes more than 1.0 min earlier.

The difference of chromatographic behavior of deuterated compounds, as compared to non-deuterated prototypes, is discussed in relation with the differences between C H and C D bonds. Deuterium atom is smaller than hydrogen; consequently, C D bonds are shorter than C H, which results in smaller "molar volume" and reduced lipophilicity of polydeuterated molecules. RP chromatography is based, mainly, on hydrophobic interactions. Therefore retention of compounds with lower lipophilicity should be weaker. As a result, in RP LC, molecules containing heavier isotopes of hydrogen elute earlier. This was suggested as a major contribution to the observed "Isotope effect".

Introduction. My PhD research project focused on development and application of analytical methodologies in the field of Alzheimer’s Disease (AD) drug discovery.[1] AD is a neurodegenerative disease clinically characterized by progressive memory loss, cognitive dysfunction and behavioural change, and is the most common type of dementia. The major neuropathological hallmarks of AD are amyloid plaques, consisting of β-amyloid peptide, and neurofibrillary tangles (NFTs) mostly composed of paired helical filaments of hyperphosphorylated tau. Nowadays, AD is considered a global health concern for two main reasons: firstly, due to the gradual aging of the population, its incidence is estimated to increase; secondly, the approved drugs only alleviate the symptoms and delay the cognitive decline. The drugs currently used in the treatment are an antagonist of glutamate NMDA receptor (Memantine) and inhibitors of the enzyme acetylcholinesterase (Galantamine, Rivastigmine and Donepezil). Recently a controversial drug (Aducanumab) which is a human IgG1 anti-Aβ monoclonal antibody specific for β-amyloid oligomers and fibrils, passed successfully all the clinical trials. Until recently, AD has had no treatment that can be considered disease-modifying, because the therapies in use do not affect its incidence or the known AD altered pathways. Besides, late-stage clinical trials of new drugs failed due to the multifactorial nature of AD. Thus, a promising strategy might be designing multi-target directed ligands (MTDLs) which could interact simultaneously on different pathways.

In this context, the main goals of my Ph.D. thesis were to develop advanced analytical methodologies suitable for characterizing new potential MTDLs. This characterization would occur both in the first discovery stage, i.e. at the high-throughput screening level on isolated targets, and in the development phase in order to evaluate bioavailability and pharmacokinetics in vivo. The main targets involved in the AD pathogenesis taken into consideration were two enzymes, glycogen-synthase kinase 3 β (GSK-3β) and histone deacetylases (HDAC). GSK-3β is responsible for the hyperphosphorylated tau proteins involved in the intraneuronal neurofibrillary tangles, while HDAC for the histone deacetylation.[2][3] Both GSK-3β and class I–II HDAC inhibitors are able to shift microglia from the neurotoxic activation (proinflammatory) to the neuroprotective (anti-inflammatory) phenotype. Moreover, the combined inhibition of GSK-3β and HDACs induces synergistic neuroprotective effects, potentially resulting in an improved therapeutic selectivity. My study also focused on another target, i.e. the β-amyloid (Aβ) aggregation process.[4] This aggregation starts incidentally with the formation of oligomers species that are reorganized into protofibrils and fibrils, the latter found in amyloid plaques. Oligomers accumulated in the brain of AD patients are suggested to be the most toxic species for the cells. By permeabilizing cellular membranes, they can initiate a series of events leading to cell dysfunction and apoptosis. Due to the critical role of Aβ aggregation process in the development and progression of AD, many studies have been carried out on the characterization of molecules capable of contrasting this process. Furthermore, the Aβ cascade is strictly connected to other AD pathological events, including the formation of neurofibrillary tangles (NFTs), inflammatory reactions, increased oxidative stress and mitochondrial dysfunction, which are all factors of cell death and dementia.

Characterization of new potential active molecules in AD drug discovery.

The method was applied for the characterization of the First-in-Class GSK-3β/ HDAC dual Inhibitors designed by the medicinal chemistry group supervised by Professor Andrea Milelli, Department for Life Quality Studies, University of Bologna. The in vitro high-throughput luminescent assay was able to select among the new series of molecules a promising hit compound with an inhibitory activity towards GSK-3β in the low micromolar range (3.19 ± 0.08 μM ). This compound induces an increase in histone acetylation and a reduction of tau phosphorylation. From its biochemical characterization it was demonstrated that it is non-toxic in SH-SY5Y neuroblastoma cell line. Moreover, it promotes neurogenesis and shows immunomodulatory effects. For concentrations ‹100 μM, it proved no lethality in a wt-zebrafish model and it has a high water solubility. For all these reasons this GSK-3β/HDAC dual inhibitor could be considered interesting to develop an innovative disease-modifying agent.[6]

The luminescent kinase assay was also suitable to screen natural compounds. Notably, different alkaloids isolated at the Department of Pharmaceutical Botany, Charles University in Hradec Králové (Czech Republic) from various Amaryllidaceae plant species were characterized towards GSK-3β. This class of natural molecules proved to be interesting in the research of MTDL for being active on various targets given its many properties (antibacterial, antimalarial, antitumor and AChE inhibitors). In particular one of the drugs currently approved for AD (Galantamine) is an amaryllidaceae alkaloid. After a preliminary screening at single test concentration (50 µM), the most promising compounds were the alkaloids with a homolycorine and lycorine structure, whose inhibitory potency was then evaluated with IC50 values in the micromolar range. It has already been possible to determine some SARs (hydroxyl substitution in position 2 and tetrahydrofuran ring opening) but it will be necessary to characterize a wider range of natural and semi-synthetic compounds for further investigation.[7] Aggregation of the Aβ peptide, in particular Aβ42, is another crucial process in AD pathogenesis. Recent studies have proved the ability of carbon monoxide-releasing molecules (CORMs) to protect neurons and astrocytes from Aβ peptide toxicity. In fact, CORMs are able to carry and release controlled levels of CO that exert anti-inflammatory and anti-apoptotic activities at physiologically relevant concentrations. In order to investigate the reactivity of CORM-2 and CORM-3 with Aβ42 in vitro and the potential inhibition of its aggregation a multi-methodological approach has been applied. The ESI-MS analysis allowed the detection of stable Aβ42/CORMs adducts, involving the addition of the Ru(CO)2 portion of CORMs at histidine residues on the Aβ42 skeleton.

Furthermore, thioflavin T fluorescence assay showed that CORMs through formation of stable adducts with Aβ42 are able to inhibit its aggregation. The inhibitory effect was confirmed also by MS analysis. Moreover, CD studies of Aβ42 spectra recorded in the absence and in the presence of CORM-3 at a 1:1 molar ratio showed the ability of CORM-3 to stabilize the peptide in its soluble, unordered conformation. This stabilization delayed its misfolding and aggregation. The multi-methodological study developed allowed to deeply investigate the neuroprotective mechanism of CORMs, that can be considered promising active molecules towards amyloid aggregation and its toxicity.[8]

A lipidomic approach to investigate Aß42 peptide toxicity on neuronal cells in view of Alzheimer’s Disease drug discovery. Given the central role of the amyloid pathway in the multifactorial nature of AD, I dedicated large part of my research to further explore the amyloid aggregation process and its mechanism of toxicity. Different studies highlighted connections between amyloid aggregation and AD extensive oxidative stress.[9] Interestingly, β- amyloid aggregation and lipid peroxidation are strictly related.[10] A possible explanation for this connection could be ascribed to the hydrophobic nature of the oligomeric peptide which makes it able to intercalate into the neuronal lipid bilayer. We therefore aimed at clarifying the mechanism of Aß42 aggregation species toxicity through a lipidomic study on the most important classes of lipids, which we identified for the first time in differentiated SH-SY5Y neuroblastoma cells: Phosphatidylcholines, Lyso-Phosphatidylcholines, Ether-Phosphatidylcholines, Phosphatidylethanolamines, Sphingomyelins and Triacylglycerols. In fact, many mechanisms of toxicity have been proposed as a consequence of the effects of oxidative stress on poly unsaturated fatty acids (PUFA), included in lipids structures.[11] Part of the work was carried out during the period I spent at the Institute of Pharmaceutical Science in Tuebingen (Germany) in the research group of Professor Laemmerhofer, whose field of expertise is grounded in lipidomic analysis. For this purpose, we optimised an LC-MS method to determine the lipid profile of differentiated human neuroblastoma derived SH-SY5Y cells. Cells were treated with increasing concentrations of Aß42 (2-50 µM) at different times of incubation, then the lipid extraction was carried out with IPA: H2O (90:10 v/v). LC-MS analyses of extracts were performed by RP-UHPLC system coupled to a high-resolution quadrupole TOF mass spectrometer in comprehensive data-independent SWATH acquisition mode.[12][13] Data processing was achieved via MS-DIAL (version 4.24). Each lipid class profiling in SHSY5Y cells treated with Aβ42, was compared to those obtained for the untreated cells to identify some distinctive alterations in their lipidomic pattern. This approach was found suitable to underline some peculiar lipid alterations due to Aβ42 different aggregation species and to explore deeply the cellular response mechanisms to the toxic stimulus. Further studies will be conducted to identify some specific PC and PE oxidized species, which are useful as biomarkers. Thus, the present method can also be applied to underline the mechanism of action of amyloid aggregation inhibitors, able not only to inhibit the aggregation process, but also to reduce lipid alterations, in view of AD new drugs discovery.

Development and validation of analytical methods for the pharmacokinetic profile of nutraceutical formulations. In view of taking into consideration prevention aids of AD, a correlation between antioxidants and polyunsaturated fatty acids intake from diet have been demonstrated to play a key role in the protection of cognitive performance.[14] In light of these premises, polyphenols are explored for their properties as interesting active agents able to prevent the oxidative stress as risk factor in the early altered pathways of the pathogenesis of AD. Therefore, in order to clarify their pharmacokinetics, a new chromatographic method by Ultra High Performance liquid chromatographic (UHPLC) technology, has been developed and validated for the determination of polydatin and resveratrol, as potential metabolite, in human plasma.

Polydatin is the natural precursor of resveratrol combined with glucose. Their antioxidant properties are known. Moreover, Polydatin demonstrated to have higher scavenging activity against hydroxyl radicals than resveratrol both in vitro and in vivo. This can be ascribed to its higher oral bioavailability. Thus, the metabolism of polydatin in vivo has been investigated through many studies. In rats it was showed that polydatin undergoes extensive first-pass deglycosylation and glucoronidation. Therefore, polydatin is primarily metabolized to resveratrol in the small intestine and liver. Then, resveratrol is metabolized into glucoronidated form. Resveratrol and glucoronidated resveratrol are mainly excreted by the kidneys.[15] In this view, dietary supplements containing polydatin are gaining great interest. Since no studies were performed in humans, it was necessary to develop analytical methods for the characterization of polydatin pharmacokinetic. This could determine the right posology and elucidate the polydatin metabolism in humans. For these reasons, my study focused on developing an Ultra High Performance liquid chromatographic (UHPLC) method for the determination of polydatin and resveratrol, as potential metabolite, in human plasma.[16] After the optimization of the chromatographic conditions, the method has been validated on spiked human plasma samples. The optimized extraction allowed to obtain analytes recovery up to 98,48 ± 4,03%. Then, the isocratic elution in reversed phase mode provides a fast separation of polydatin and resveratrol (less than 10.0 min). Chromatographic analysis was performed on a C18, 10 cm x 3.0 mm, 2.7 µm stationary phase, by using triethanolamine phosphate solution (0.1 M, pH= 3.7) and ACN 85:15 (v/v) as mobile phase at a flow rate of 0.5mL/min. The UV detector was set at 306 nm for the analysis of both polydatin and resveratrol. The method resulted sensitive with a limit of detection (LoD) and the limit of quantification (LoQ) for polydatin in plasma samples found to be 7,82 ± 0,38 nM and 26,06 ± 1,28 nM respectively. The method was found to be accurate and precise with a coefficient for intra- and inter-day variation below 5%. The method was then applied to the analysis of plasma samples from orally treated volunteers with nutritional supplements containing polydatin. The polydatin distribution after dietary supplements (containing 40 or 75 mg of polydatin) administrated to 6 healthy volunteers was analysed in blood samples from all volunteers for one month. One urine sample was also analysed by employing the same UHPLC method. The results suggested that a dose higher than 75 mg/day has to be administered in order to determine a concentration in the blood suitable to exert its properties.

Conclusions. The analytical methodologies investigated in my PhD thesis resulted able to support AD drug discovery. In fact, they reliably and quickly identify true positive hit compounds, and perform structure activity relationship in view of the development and optimization of a lead structure with an improved activity and bioavailability. The new multimethodological approaches described give more innovative insights about the mechanism of toxicity Aβ aggregation that is crucial in AD pathogenesis and the combination of advance analytical methodology characterize new potential aggregation inhibitors in a comprehensive way. The identification and the relative quantification of some lipid altered species due to Aβ toxicity could be useful in the characterization of new biomarkers for the early detection of AD and in the therapeutic monitoring. Moreover, the methods developed showed to be suitable for the pharmacokinetic study of nutraceuticals formulation useful in AD prevention.

References
[1] De Simone A., Naldi M., Tedesco D., Bartolini M., Davani L., Andrisano V. J Pharm Biomed Anal. 2020, 178: 112899
[2] Lauretti E, Dincer O., Praticò D. Biochim Biophys Acta Mol Cell Res. 2020, 1867: 118664
[3] De Simone A., Milelli A. ChemMedChem. 2019, 14: 1067-1073
[4] Castellani RJ, Plascencia-Villa G, Perry G. Lab Invest. 2019, 99: 958-970
[5] Baki A.,Bielik A., Molnár L., Szendrei G., Keserü G. M. Assay Drug Dev Technol. 2007, 5: 75-83
[6] De Simone A., La Pietra V., Betari N., Petragnani N., Conte M., Daniele S., Pietrobono D., Martini C., Petralla S., Casadei R., Davani L., Frabetti F., Russomanno P., Novellino E., Montanari S., Tumiatti V., Ballerini P., Sarno F., Nebbioso A., Altucci L., Monti B., Andrisano V., Milelli A. ACS Med. Chem. Lett. 2019, 10: 469-474
[7] Hulcová D., Breiterová K., Siatka T., Klímová K., Davani L., Afratová M., Hostalkova A., De Simone A., Andrisano V., Cahlíková L. Molecules 2018, 23: 719
[8] De Simone A., Naldi M., Tedesco D., Milelli A., Bartolini M., Davani L., Widera D., Dallas M. L., Andrisano, V. ACS Omega 2019, 4: 12308-12318
[9] Allan Butterfield D., Boyd-Kimball D. J Alzheimers Dis. 2018, 62: 1345-1367.
[10] Butterfield D. A., Lauderback C. M. Free Radic. Biol. Med. 2002, 32: 1050-60
[11] Kao YC., Ho PC., Tu YK., Jou IM., Tsai KJ. Int J Mol Sci. 2020, 21:1505
[12] Calderón C., Rubarth L., Cebo M., Merfort I., Lämmerhofer M. Proteomics 2020, 20: e1900113
[13] Drotleff B.,Illison J., Schlotterbeck J., Lukowski R., Lämmerhofer, M. Anal. Chim. Acta 2019, 1086: 90-102
[14] Bhatti G. K., Reddy A. P., Reddy P. H., Bhatti J. S. Front Aging Neurosci. 2020, 11: 369
[15] Wang J., Zhu X., Peng Y., Zha W., Feng D., Zhu Y., Wan P., Qi H., He J., Zhou J., Sun J. Biomed. Chromatogr. 2012, 26: 1371-6
[16] Montanari S., Davani L., Tumiatti V., Dimilta M., Gaddi A. V., De Simone A., Andrisano V. J. Pharm. Biomed. Anal. 2021, 198: 113985

Today, following the biorefinery concept, several research groups and companies are strongly encouraged to focalize their effort in building innovative circular-based economy model, based on the exploitation of agro-food by-products as a source of innovative polymers or bioactive compounds for food and pharmaceutical fields [1]. Cereals are the most cultivated crops in Lombardy and among them maize and its pigmented varieties are characterized by a very complex chemical composition, rich in polyphenols and caroteinoids, representing potential good sources of bioactive compounds [2]. Based on the above mentioned, the aim of this PhD project was the development of innovative nutraceutical ingredients in collaboration with FlaNat Research s.r.l, obtained from agro-food wastes of typical Lombard crops and produced by the company. The first by-product considered was a purple maize variety (Moradyn) cob. Moradyn purple cobs were extracted for 3h at 50 °C with an hydroalcoholic solution (H2O:EtOH, 50:50 v:v), and the extract was freeze dried and stored for further analysis at 20 °C [3]. Moradyn purple cobs extract (MCE) was chemically characterized by RP-HPLC-DAD-ESI/MSn using data dependent acquisition mode. Twenty compounds were detected and identified according to the comparison of their selectivity, UV-vis spectra, pseudo-molecular ions and fragmentation patterns data with reference pure standard compounds, if commercially available; otherwise, with data reported in literature. Among them, three main anthocyanins were identified: cyanidin-3-O-glucoside (C3G), perlagonidin-3-O-glucoside, and peonidin-3-O-glucoside, confirming literature data [4,5]; moreover, MCE was rich in flavonols, mainly quercetin, kaempferol, and isorhamnetin derivatives. Anthocyanins are known for their strong antioxidant and hypoglycemic activities [2, 6, 7] and considering that anthocyanin fraction (AF), represented 28% (w/w) of the phytocomplex, it was isolated by solid phase extraction technique, freeze dried, and tested for its hypoglicemic and antiglycative properties, which are strictly correlated with type II diabetes. The healthy properties of AF were compared with those of MCE and both the samples were tested at different concentrations, ranging from 0.05 to 1mg/mL. The bioactivity screening was carried out using three different antiglycative assays, each monitoring a single phase of Maillard’s reaction (early, middle, and last phase) and α -glucosidase inhibitory test. Results showed that AF at the concentration 0.5mg/mL totally inhibited α -glucosidase, while MCE reached the same activity value only when at 1mg/mL; however, MCE at 1mgl/mL exerted a stronger antiglycative activity than the isolated anthocyanin fraction. So, even if AF represented the most active class of compounds in MCE, it is well known that the healthy effect of AF can be strongly enhanced by the synergy with other MCE polyphenols [8]. Therefore, MCE was selected for further experiments. MCE, thanks to its chemical composition, could be potentially used as nutraceutical ingredient to prevent typical risk factors related to age related chronic diseases. However, the efficacy of such kind of phytochemicals is related to their stability and bioaccessibility [9]. So, in order to fully investigate MCE efficacy and quality, a bioaccessibility study and a solid-state stability investigation were performed. MCE bioaccessibility was evaluated by submitting it to an in vitro static simulated digestion protocol [10], monitoring the gradual reduction (by HPLC) of fifteen markers present in the extract at each digestion phases. Gastrointestinal pH and enzymatic conditions strongly affected polyphenols bioaccessibility, since their concentration strongly decrease at each digestion phase [3]. The effects of digestion on MCE bioactivity were evaluated by monitoring its hypoglycemic and antiglycative properties at the end of each digestion step. These experiments confirmed that digestion strongly affected MCE quality and efficacy: in fact, no antiglycative or hypoglicemic activity was detected at the end of the similated digestion [3]. In order to preserve the biological properties, MCE was added with Arabic gum (80/20 w/w) so mimicking a typical industrial scaled-up of an ingredient and an accelerated stability study to assess its solid-state stability was performed. A preliminary physico-chemical characterization on the raw materials and their mixtures was explored in order to evaluate if 20% Arabic gum (w/w) could modify MCE behaviour when exposed to high temperatures. The calorimetric profiles were acquired in the range 80 °C – 300 °C. MCE profile presented three large endothermic peaks at 90 °C, 160 °C, and 225 °C, due to the decomposition of some of its different constituents, since the cooling curve showed no peak. The first peak registered in MCE profile appeared as the superimposition of two processes with different kinetics, the first one, faster, starting at around 0 °C and the second one, slower, starting at about 50 °C. Moreover, DSC analysis was also performed on C3G, which is the most abundant anthocyanins in AF, and it was observed that the first peak registered in MCE calorimetric curve was also present in C3G profile, suggesting that anthocyanins could be the first components to be decomposed and hence the most instable of MCE constituents. For this reason, C3G was chosen as a stability marker for MCE mixtures and its decomposition temperature as the best temperature range to study the behaviour of the mixtures under thermal stress. The calorimetric curve of Arabic gum showed a very large and asymmetric signal starting at -35 °C, centred at 60 °C and not ended at 110 °C, due to irreversible processes happening in the matrix. Finally, the calorimetric curves of the mixtures MCE – 20% Arabic gum (w/w) and C3G – 20% Arabic gum (w/w) were superimposable in shape and characteristic temperatures to those of the main components up to 150 °C, with no effect and by-side reactions attributable to the gum presence in the examined temperature range. To further assess the compatibility between C3G and Arabic gum, IR spectra of the pure component and of the 20% mixture were acquired. Therefore, based on the physico-chemical characterization of the mixture MCE-20% Arabic gum and its main components, this mixture was considered suitable to carry-out the experiments aimed to assess MCE solid-state stability, always using C3G as reference marker [11]. MCE – 20% Arabic gum mixtures solid-state stability was evaluated by applying an innovative approach based on Accellerated Stability Assessment Program (ASAP) [10]. This approach allowed to rapidly and accurately estimate MCE-20% Arabic gum shelf-life by estimating its isoconversion time (which is the time at which marker compounds reached a fixed specification limit of degradation, when exposed to specific temperature and relative humidity – RH- value conditions) from the degradation rate constant (k), which, in turn, was extrapolated from a mathematical model based on MCE (equation 1) [11, 12].

lnK=24.209-[(7626.6*1/T) + (0.033*HR %)] (1)

Up to now, stability studies of natural extracts had not generally carried out using isoconversion approach even if this methodology allowed to estimate the degradation entity in a shorter time than the classical protocol and with better accuracy. Moreover, this approach let to estimate storage stability of a natural extract consisting of a high number of compounds without generating a big dataset. In fact, by selecting the marker compound which was the most sensitive to degradation parameters and extrapolating its isoconversion time, it was possible to predict the overall phytocomplex shelf-life. So, by applying this protocol MCE shelf-life at 25 °C and 30% RH was in the range 27-33 days [9]. This range was to much short for a commercial ingredient. Therefore, considering the results obtained from both the stability and bioaccessibility studies the need of a new MCE formulation with a suitable stabilizer agent was evident, in order to prolong MCE shelf-life and ameliorate its efficacy. The second ingredient developed during this PhD project was obtained from Camelina Sativa L. seeds cake, which is a Lombard by-product crop rich in polysaccharides (CSKP) and, therefore, a potential suitable carrier for MCE [13-14]. CSKP was extracted using water and purified by adding ethanol at 4°C. Both the chemical composition (defined by FT-IR, NMR, and SEC techniques) and the rheological properties of CSKP were evaluated. A new ingredient based on CSKP and MCE at different ratios (30/70, 50/50, and 25/75 w/w) was formulated and encapsulation efficacy was evaluated, using C3G as a reference marker. C3G was successfully encapsulated both in presence of 50% and 75% CSKP, since the encapsulation efficacy values were higher than 60% and 80%, respectively; therefore, these two formulations were selected to perform further bioaccessibility and stability studies. Both MCE- 50% CSKP and MCE- 75% CSKP formulation were submitted to in vitro digestion process and anthocyanins release kinetic was monitored at different timing point by HPLC analysis. Results indicated that the release trend was similar: C3G was mainly released after the oral phase, probably due to CSKP swelling and enzymatic degradation, while its release rate was kept constant during the gastric phase and partially decreased during duodenal phase. However, in presence of 75% CSKP C3G abundance registered at the end of gastric and duodenal phases was higher than in presence of 50% CSKP, suggesting that MCE- 75% CSKP strongly enhanced MCE bioaccessibility and made it able to overcome digestion environment, since all the marker compounds were still present in a bioaccessible form at duodenal level. Finally, MCE- 50% CSKP and MCE- 75% CSKP were submitted to solid-state stability study following the above mentioned approach. The data obtained from the stability study showed that only MCE- 75% CSKP successfully prolonged MCE shelf-life, which was estimated to be >400 days at 25 °C-30% RH and therefore consistent with the market need. Considering the overall results, MCE- 75% CSKP was selected for further experiments aimed to evaluate hypoglycemic and antiglycative activity in vitro. Moreover, CSKP hypocholesterolemic activity was further tested by assessing its cholesterol absorption capacity. The bile acids binding assay was carried-out using a human primary bile acids mixture (BA) as standards in order to mimic bile healthy human BA composition [15]. Thus, CSKP was submitted to in vitro digestion protocol [10] using BA as bile salts during the duodenal phase, then digestate was centrifuged and free BA was quantified in supernatant by HPLC analysis [16]. CSKP was tested at 4 different concentrations (0.75, 2.5, 5 and 10 mg/ml) and compared to Cholestyramine (positive control). Results showed that CSKP had good bile acid absorption property, since it totally bound BA already at the concentration of 5 mg/mL, similarly to the positive control [15]. In conclusion, this PhD project focused on the development of two different botanical ingredients with different bioactivities from Lombard agro-food by-products in collaboration with FlaNat Research: MCE and CSKP; in addition, the formulation obtained using CSKP as carrier for MCE was an effective alternative to obtain good stability and higher efficacy.

References
[1] Baiano A. et al. Molecules, 2014, 19: 14821-14842
[2] Colombo R. et al. Molecules, 2021, 26: 199-241
[3] Ferron L. et al. Molecules, 2020, 25: 1958-1979
[4] Lao F. et al. Food Anal. Methods, 2016, 9: 1367–1380
[5] Li C. Y. et al. J. Agric. Food Chem., 2008, 56: 11413–11416
[6] Ma H. et al. Int. J. Mol. Sci., 2018, 19: 461-480
[7] Zhu F. Food Res. Int., 2018: 109, 232-249
[8] Comisso M. et al. Plos One, 2017, 12: 1-23
[9] McClements D. J. J. Food Sci., 2015, 80: 1602-1611
[10] Minekus M. et al. Food Funct., 2014, 5: 1113-1124
[11] Ferron L. et al. 2021, submitted to Foods
[12] Waterman K. C. 2011, AAPS PharmSciTech, 12: 932-937
[13] Li N. et al. J. Ind. Crop, 2016, 83: 268-274
[14] Sarv V. et al. J. Am. Oil Chem. Soc., 2017, 94: 1279–1285
[15] Jung H. et al. J. Agric. Food Chem., 2012, 60: 6217-6222
[16] Nauman S. et al. Int. J. Mol. Sci., 2018, 19: 2193-2208

In the early phases of drug discovery, in vitro metabolism of new chemical entities (NCE) is essential for understanding the metabolic, pharmacological, and toxicological profile of the drugs. Traditionally, metabolism studies are carried out in solution, which requires a large number of tests to trace the drug’s profile and results in the use of large volumes of microsomal fraction, cofactors, and extensive sample preparation before analysis. Thus, novel analytical strategies that could minimize costs and increase test efficiency are of paramount importance for in vitro/in vivo drug metabolism studies. In this context, the main goal is the development of a new method for in vitro metabolism study through the immobilization of enzymes from cytochrome P450. In addition to the immobilization of microsomes, this model meets a variety of green chemistry goals including reduced waste generation, lower consumption of biological materials, and reuse of the enzymatic system in various analyses.

Thus, Rat microsomal fractions were covalently immobilized on to magnetic beads (RatF-Mb), according to the protocol reported.¹ A Dohelert design experiment² was carried out to determine the immobilization conditions in which variables such as immobilization time, glutaraldehyde, and protein concentrations were evaluated. The ability to metabolize albendazole to its sulfoxide was used as a means to monitor the activity of produced RatF-Mb through a qualified LC-MS method. The RatF-Mb obtained through the designed experiment caused an increase of 67% in the production of albendazole sulfoxide (major metabolite), moreover the time of immobilization was reduced to 2h only. The recycle of RatF-Mb was tested and showed good efficiency providing conditions for reusing in 6 experimental cycles.

References
1. Vanzolini, K. L., Jiang, Z., Zhang, X., Vieira, L. C. C., Gonçalvez, A. C., Cardoso, C. L., Cass Q. B., Moaddel, R.; Talanta. 2013, 116, 647. DOI: 10.1016/j.talanta.2013.07.046
2. Zhou, B., Li, Y., Dillespie, J., He, G-Q., Horsley, R., Schwarz, P.; J. Agric. Food Chem. 2007, 55, 10141. DOI: 10.1021/jf0722957

G protein coupled receptors (GPCRs) are one of the most promising druggable targets, due to their involvement in the regulation of a range of physiological processes in humans.¹ Still as much as 120 GPCRs identified in human genome remain orphan with unknown endogenous ligands.² As other membrane proteins, GPCRs were shown to establish interactions with lipid constituents of cell membrane, suggesting specific lipid molecules to play a role of allosteric modulators.³ It strongly underlines importance of studying ligand binding with GPCRs in a native lipid environment in order to ensure preserving the native conformation of the receptor and to avoid alteration of functional behavior. Application of recently established detergent-less solubilization methods involving styrene–maleic acid (SMA) co-polymers enables extraction of membrane proteins from biological membranes along with a native annular lipid shell.⁴

Unfortunately, overexpression of human proteins in a native environment, requires the use of human cell culture system which has relatively low efficiency due to the limited scalability and slow growth rates. GPCR expression furthermore complicate the endeavor, since overexpression leads to toxic event to the cells.⁵

The presentation will show results of a joint Polish-Italian exchange project aimed at establishing a stable framework for functional studies on various GPCR receptors - optimization of SMALP solubilization protocol applied for low-yielding native expression systems. In the scope of the PhD work, I have designed and obtained recombinant human β2-adrenergic receptor (rhβ2AR), which was overexpressed in HEK293T adherent human cell line. Designed recombinant receptor contained appropriate peptide tags needed for identification and purification of the protein. Prepared plasmid construct was transfected into the HEK293T cells, which were cultured in large quantities in order to isolate plasma membrane fraction used for extraction of membrane proteins with a help of SMA polymer. One of the partial goals was to investigate parameters influencing the extraction yield when using SMA polymer solubilization method. Influence of buffer composition and properties on membrane solubilization efficiency as well as SMA-lipid particles size distribution was investigated with use of turbidimetry, dynamic light scattering (DLS) and size-exclusion chromatography (SEC). Results show that maintaining proper ionic strength throughout all of the stages of the solubilization and purification processes is crucial. Additionally, modified purification scheme using combined in-line method of size-exclusion chromatography (SEC) along with an immobilized metal affinity chromatography (IMAC) is to be presented, what allowed to further increase rhβ2AR yield by minimizing losses during the purification procedure. Presented results show, that in order to obtain SMA water-soluble variants of GPCRs using native expression systems, a careful optimization of each of the stages of preparation is needed; starting from the preparation of appropriate recombinant variant of the protein, followed by appropriate solubilization conditions and efficient purification scheme. Proper optimization of already published protocols might allow sufficient solubilization efficiency of GPCRs expressed in native, low-yielding systems along with annular lipid layer, which in turn can facilitate accurate ligand binding and functional studies.

References
1. Fang Y. et al., Front. Pharmacol. 2015, 6: 295. DOI: 10.3389/fphar.2015.00295
2. Hauser AS. Et al., Br. J. Pharmacol. 2019, 177: 961. DOI: 10.1111/bph.14950
3. Dawaliby R. Et al., Nat. Chem. Biol. 2016, 12: 35. DOI: 10.1038/nchembio.1960
4. Lee SC. Et al., Nat. Protoc. 2016, 11: 1149. DOI: 10.1038/nprot.2016.070
5. Massote D., BBA Biomembranes. 2003, 1610: 77. DOI: 10.1016/S0005-2736(02)00720-4

Plants are rich sources of bioactive compounds used in pharmaceutical preparations and drugs. However, most of biologically interesting components are still not revealed. It concerns also herbs and preparations used in Ayurveda or Traditional Chinese Medicine (TCM) which are very popular alternative to synthetic drugs. Among others, Rhodiola rosea, Akebia quinata, Clitoria ternatea are known for their antibacterial, adaptogenic and anti-dementia properties. searching and investigation of new biologically active substances of plant origin is very important due to potential use of these substances in the treatment of civilization diseases, such as Alzheimer disease, diabetes, depression, osteoporosis and bacterial infections. Combination of effect directed analysis (EDA), TLC biodetection TLC screening is a method of choice in searching new drugs enabling analyses of many samples in parallel and the comparison of their activity directly on a TLC plate.^1-3 Other advantages of the method are the ability to analyze target compounds even in complex matrices, the simplicity in sample preparation and the complete solvent evaporation. Moreover, the additional advantage is connected with the possibility of identification and isolation of the target substances.

In our current research, TLC-DB and TLC screening were used for detection antioxidants, antimicrobials and enzyme inhibitors including those responsible for inhibiting acetylcholinesterase, butyrylcholinesterase, lipase, α-glucosidase, α-amylase and tyrosinase in the extracts of chosen plants. According to the literature: acetylcholinesterase inhibitors could be used in the Alzhemer`s disease treatment; α-glucosidase, α-amylase and butyrylcholinesterase inhibitors are potential drugs in the therapy of the type 2 diabetes; lipase inhibitors could be effective in the treatment of obesity; tyrosinase inhibitors – could be used as a treatment of some skin disorders related to melanin hyperpigmentation. The results show that the tested samples differ in their chemical composition and biological activity. The bioactive zones detected by TLC-DB were isolated from the plate and analyzed by mass spectrometry.

References
1. I.M. Choma, E.M Grzelak, J. Chromatogr. A 1218 (2012) 2684-91. DOI: 10.1016/j.chroma.2010.12.069
2. G.E. Morlock, W. Schwack, J. Chromatogr. A 1217 (2010) 6600–09. DOI: 10.1016/j.chroma.2010.04.058
3. A. Marston, J. Chromatogr. A 1218 (2012) 2676-83. DOI: 10.1016/j.chroma.2010.12.068

Introduction. The ‘omics sciences are currently in development offering a new and combined perspective of cellular and organismal environment. Among these, genomics and proteomics are, probably, the most developed and already provided a large amount of data. On the contrary, lipidomics is still an emerging field with ongoing development of analytical and bioinformatics methods necessary to address the structural and physiochemical diversity of lipids [1,2]. The importance to provide a strong methodological approach paired to a rigorous data interpretation is explained by the recent discovery of the lipids’ key role in many biological processes. Indeed, they act not only as passive structural components or energy depots but also as second messengers in signal transduction, regulators of inter-cellular interactions and of surface charge. Again, most of lipids are involved in many diseases such as diabetes, atherosclerosis, cancer, neurodegenerative and dermatological ones [2]. Mass spectrometry (MS) technique, thanks to the recent significant advances, is the most suitable analytical method in many of ‘omics sciences, including lipidomics, making easier their integration [3]. Nevertheless, mainly for lipidomics, careful optimization of protocols is needed to reach the complete coverage of all lipid classes (fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, sterol lipids, prenol lipids, saccharolipids and polyketides) with an acceptable level of resolution.

The aim of my research is to integrate lipidomics analyses with quantitative proteomics methodologies for a more complete, multi-omics vision of intra- and extra-cellular processes in several biological conditions (physiological status, drug exposure, pathological conditions). In particular, during my PhD I applied proteomics and/or lipidomics approaches to investigate: i) effects of carnosine on nude mice skin to prevent UV-A damage; ii) Optimization of a reliable, reproducible, and fast methodology based on high resolution MS for the characterization and relative quantification of fatty acid esters of hydroxy fatty acids (FAHFAs) in human white adipose tissue (WAT) of obese patients; iii) Evaluation of molecular effects of low-molecular-weight hyaluronic acid in human dermal fibroblasts through advanced quantitative proteomics and semi-quantitative lipidomics; iv) Analysis of lipidome and proteome profile changing induced by Y-oryzanol prevention treatment in obese rats.

Study of carnosine’s effect on nude mice skin to prevent UV-A damage. The skin is an important barrier against external attacks from bacteria, radicals, or radiations. UV-A radiations (315-400 nm) cause significant impairment of this barrier, inducing inflammation, oxidative stress, and wrinkle formation, thereby promoting photoaging [4,5]. Previous studies reported that carnosine, a potent antioxidant, and carbonyl scavenger agent, may prevent photoaging features in the skin of hairless mice exposed to UV-A radiations. Here a quantitative proteomic approach in high-resolution MS (HRMS, Orbitrap Fusion™ Tribrid™ Mass Spectrometer -Thermo Scientific) was used to analyze the changes evoked by carnosine in the skin proteome of hairless mice exposed to UV-A. This approach allowed to quantify more than 2480 proteins, among them consistent differences were observed for 89 proteins in UV-A exposed vs control unexposed skins, and 252 proteins in UV-A-exposed skin preventively treated by carnosine (UVAC) vs UV-A. Several functional pathways were altered in the skin of UV-A exposed hairless mice, including the integrin-linked kinase, calcium signaling, fibrogenesis, cell migration and filament formation, mitochondrial function and metabolism. On the contrary, pre-treatment with carnosine prevented from UV-A induced proteome alterations. In conclusion, this study emphasizes the potency of a proteomics approach to identify the consequences of UV-A radiations in the skin, and points out the capacity of carnosine to prevent the alterations of skin proteome evoked by UV-A.

Characterization and relative quantification of fatty acid esters of hydroxy fatty acids (FAHFAs) in human white adipose tissue (WAT) of obese patients. Elevated levels of circulating fatty acids (FA) are generally associated with diabetes and obesity. However, certain FA (dietary omega-3 fatty acids, endogenous palmitoleate etc.) have positive metabolic effects partially reverting these conditions. Similarly, a new class of endogenous lipids called fatty acid esters of hydroxy fatty acids (FAHFAs), present in adipose tissue, kidney, pancreas, and serum, have shown analogous biological activities. Indeed, studies on insulin-resistant individuals which have lower palmitic acid esters of hydroxystearic acid (PAHSA) concentrations in serum and adipose tissue, suggest that FAHFAs deficiency may contribute to metabolic diseases [6]. Furthermore, FAHFAs, particularly 9-PAHSA and 5-PAHSA, have shown anti-inflammatory and immunomodulatory effects [7]. Increasing evidence on the physiological roles of FAHFAs motivates a more extensive characterization of these lipids as possible biomarkers and therapeutic targets for pathological conditions such as diabetes or obesity. Nevertheless, the low concentration in human tissues paired to the large structure heterogeneity of this lipids class challenges current analytical methods for their accurate identification and quantification. Moreover, the major amount of FAHFAs in cells is incorporated into triacylglycerols (TGs) by esterification to the glycerol backbone, requiring lipolysis (i.e., TGs hydrolysis) to quantify free FAHFAs [8]. Considering the current analytical limitations, our principal goal was to optimize a reliable, reproducible, and fast methodology based on high resolution MS for the characterization and relative quantification of FAHFAs in human white adipose tissue (WAT) of obese patients. To facilitate the reliable identification of FAHFAs in a complex human matrix, first optimization of sample processing and instrumental parameters were performed using individual and pooled mixtures of FAHFAs standards (n=15). Then, the procedures were applied to the real samples i.e., human white adipose tissue (WAT, n=169) corresponded to abdominal visceral (VAT) and subcutaneous (SAT) fat depots of obese insulin sensitive patients (IS, n=34; BMI > 40kg/m²), obese insulin resistant patients (IR, n=47; BMI > 40kg/m²) and lean subjects used as controls (L, n=5; BMI ‹25kg/m²). Briefly, since TGs are the major reservoir of FAHFAs, at first, we identified the optimum conditions that would hydrolyze esterified FAHFAs from TGs without degradation. Due to low physiological abundance of FAHFAs in human tissue, amino-propyl–NH2 (Strata NH2) solid phase extraction (SPE) enrichment was performed to separate FAHFAs from FFA released by TG hydrolysis. Finally, considering the low abundance of endogenous FAHFAs in adipose tissue samples and the expected large amounts of FFAs released from hydrolyzed TGs, a C18- reverse phase (C18-RP) SPE allowed the separation of these two lipid classes. Instrumental analyses were conducted on high-resolution MS (HRMS) (UHPLC/Q-Exactive Plus™, Thermo Scientific) applying a parallel reaction monitoring (PRM) method. 61 FAHFAs were manually identified, 32 out of those considered for the final analyses being present in > 70% of the samples and at least in two groups. 16 of 32 showed a significant alteration between groups (p value < 0.05; fold change >1.5) supporting a different response based not only on the metabolic status (lean vs obese) but also on the type of adipose tissue (SAT vs VAT). To conclude, considering the novelty and complexity of FAHFAs research (so far mainly conducted in vivo), this work aimed for the first time to set up a reliable and reproducible protocol to study them in human adipose tissue giving new perspective for the treatment of metabolic diseases.

Evaluation of molecular effects of low-molecular-weight hyaluronic acid in human dermal fibroblasts through advanced quantitative proteomics and semi-quantitative lipidomics. Hyaluronic acid (HA) is a glycosaminoglycan physiologically synthesized by several human cell types, but it is also a very common ingredient of commercial products, from pharmaceuticals to cosmetics due to its widespread distribution in humans and its diversified physiochemical proprieties including biocompatibility, biodegradability, mucoadhesivity, viscoelasticity and hygroscopicity [8]. Despite its extended use and preliminary evidence showing even also opposite activities to the native form, the precise intra- and extra-cellular effects of HA at low-molecular-weight (LWM-HA) are currently unclear [9]. At this regard, our first aim was to identify in-depth and quantify proteome’s changes in normal human dermal fibroblasts after 24 hours treatment with 0.125, 0.25 and 0.50 % LMW-HA (20-50 kDa) respectively vs controls. To do this, a label-free quantitative proteomic approach based on high-resolution mass spectrometry was used (Dionex Ultimate 3000 nano-LC system -Sunnyvale CA, USA- connected to Orbitrap Fusion™ Tribrid™ Mass Spectrometer -Thermo Scientific, Bremen, Germany). Overall, 2328 proteins were identified of which 39 significantly altered by 0.125 %, 149 by 0.25 % and 496 by 0.50 % LMW-HA. Protein networking studies indicated that the biological effects involve the enhancement of intracellular activity at all concentrations, as well as the extracellular matrix reorganization, proteoglycans and collagen biosynthesis. Moreover, the cell’s wellness was confirmed, although mild inflammatory and immune responses were induced at the highest concentration. Influenced by the treatment also lipids metabolism related proteins, also in this case with 0.50% LMW-HA [10]. Then, the LWM-HA effects were investigated from the lipidome point of view. 1380 lipids unique by structure were detected by HRMS (UHPLC/Q-Exactive Plus™, Thermo Scientific) of which 477 lipids were filtered out by the data manual curation not matching the identification criteria (presence of class characteristics fragments, retention time, fragmentation pattern products at MS2 level) leaving 903 lipids (n=384 if considered at bulk level). Triacylglycerols (TGs), Ceramides (Cer) and Phosphatidylcholines (PCs) are the most represented classes (45.2%, 12.1% and 11.3% respectively). 563 out of 903 features significantly altered vs untreated control group based on One-way Anova test (adjusted p value ‹0.05, Fisher’s test) showing a clear cluster separation between 0.50% LMW-HA group and the remainders suggesting a cellular effect related strictly to this concentration as previously saw with the proteomics analyses. Ceramides and hexosy-1-ceramides, among the most important lipids classes from the biological and functional point of view both at the stratum corneum and deeper fibroblasts layer, resulted also among those most significantly affected by LMW-HA 0.50%. Although separated proteomics and lipidomics analyses allowed us to understand the cellular LWM-HA effects from different point of views, combining them provided an even further comprehension about its impact. At this regard, as final goal of this project the significant altered proteins (n= 495) and lipids through the 0.50 % LMW-HA treatment were integrated with Ingenuity Pathways Analysis software (IPA) resulting in 25 networks including 26 lipids belonging to different classes and biological functions such as mitochondrial activity, cell signaling, metabolism, and energy storage. The more complete comprehension of intra- and extra-cellular effects of LMW-HA here provided by an advanced analytical approach and integratomics analyses will be useful to further exploit its features and improve current formulations.

Analysis of lipidome and proteome profile changing induced by Y-oryzanol prevention treatment in obese rats. Grains are the most common staple food consumed worldwide and among them, rice (Oryza sativa) is the top one. It is very important so to consider its constituents, such as Y-oryzanol (Orz) that is present in the bran layer and comprises a mixture of ferulic acid esters and phytosterols. Orz has shown several biological activities and it’s often associated with fat and cholesterol-lowering, anti-inflammatory, anti-cancer, anti-diabetic effects and antioxidant ones by blocking lipid peroxidation [11,12]. It’s also recently become a great candidate in metabolic disorders like obesity, responsible of 13% of global deaths. The mechanisms under all these effects are still not completely clarified. Starting from observed proteomics changes lipids-related in heart samples (parallel project), our main goal was to describe and relatively quantify by HRMS (UHPLC/Q-Exactive™) the plasma lipidome changes in obese-induced rats (high-sugar [78%] and high-fat [22%] diet, n=16) w/wo preventive treatment with Orz (0.5% in the meal for 30 weeks) vs lean ones (n=16). The oxidized lipids profile could support the unmodified lipids analyses further than plasma proteomics ones to offer a more complete vision of obesity’s metabolism effects and Orz activity. Once spiked with Splash Lipidomix, (Avanti Polar Lipids, Alabaster, AL, USA) lipid extracts were obtained with liquid-liquid extraction using MTBE/MeOH/H2O (4:1.2:1, v/v + 0.1% BHT) method [13]. Lipid identification was based on LipidHunter vRC3_2 and Lipostar v1.3.2 software and the peak areas were manually integrated and curated using Skyline. Variation in lipid profile among groups was evaluated with MetaboAnalyst online software. 394 lipids unique by structure were identified mostly belonging to TGs followed to PCs. Once compared their concentrations among groups we saw, for TGs for example, a marked increase in the obese vs lean group but at the same time as Orz improved the conditions lowering their values respect to untreated obese as well as of CEs and ceramides as Cer(18:1/16:0), positively associated also with an increasing of cardiovascular risk [14]. Further changes were also observed in PLs classes with some PEs, LPEs, PIs lipid species upregulated in obese vs lean and decreased by Orz treatment. Opposite trend for some PCs, generally inversed correlated with obesity status. PC(18:0_22:5) for example, a marker used in obesity models [15], has shown a return to physiological-like conditions in the Orz treated group. As on-going steps, the analyses of oxidized lipids and plasma proteomics in to integrate the present results offering a more complete vision of Orz activity in obesity.

Conclusions. To conclude, all these works demonstrate how omics sciences (proteomics and lipidomics in this case) through high-resolution mass spectrometry can provide a detail description of many biological processes triggered by different stimulus (drugs, environmental stress, metabolic status etc.). An even more complete vision is obtained by their integration (integratomics), an approach still scarcely applied in the omics field although it could offer a strong contribution in the field of personalized medicine connecting several expertise and perspectives.

References
[1] Joyce AR Nature, 2006.
[2] Wenk, M. R. Cell 2010 143, 888–895
[3] Tomas Cajka and Oliver Fiehn Anal. Chem. 2016, 88, 524−545
[4] Fisher GJ, Cutis 2005; 75:5-8
[5] Zhang M. Biosci Rep 2020, 40:7
[6] Yore MM, Cell. 2014;159(2):318‐332
[7] Kolar MJ, Anal. Chem. 2018, 90, 8
[8] Fallacara A. Polymers, 2018, 10
[9] Snetkov P, Polymers, 2020, 11;12(8):1800
[10] Radrezza S., JPBA 2020 Jun;185:113199
[11] H. Masuzaki, J Diabetes Investig 2019
[12] Minatel IO Int J Mol Sci. 2016;17(8):1107
[13] Matyash Vet, J Lipid Res. 2008
[14] R. Laaksonen, European Heart Journal, 2016
[15] Pickens, C.A., Sci Rep 2017