Stages Of Drug Discovery Research Articles

The Role of Deep Learning in Pharma: Revolutionizing Drug Discovery and Development.

- In recent years, the pharmaceutical industry has witnessed a transformative shift in the way drugs are discovered and developed, thanks to the advent of deep learning. This paper explores the profound impact of deep learning techniques on various stages of drug discovery and development, from target identification and lead optimization to clinical trials and personalized medicine. Deep learning, a subset of artificial intelligence, has demonstrated exceptional capabilities in handling complex biological data, including genomics, proteomics, and chemical informatics. It enables the integration of vast and diverse datasets, facilitating the identification of potential drug targets with unprecedented accuracy. Moreover, deep learning models can predict the binding affinity of drug candidates to specific target proteins, expediting the lead optimization process and reducing the need for costly experimental iterations. Deep learning algorithms enhance patient stratification and biomarker discovery, ultimately leading to more successful trials with higher patient response rates. Additionally, the ability to analyze real-world patient data aids in the identification of adverse events and the development of safer drugs.
 Perrsonalized medicine is another area greatly influenced by deep learning, as it allows for tailoring treatments to individual patients based on their unique genetic and clinical profiles. This promises to revolutionize patient care, optimizing therapeutic outcomes while minimizing adverse effects. Despite the remarkable advancements facilitated by deep learning, there are challenges to address, such as data privacy, interpretability of models, and regulatory considerations. This paper discusses these challenges and potential solutions. Deep learning has emerged as a powerful tool in the pharmaceutical industry, driving innovation, efficiency, and precision in drug discovery and development. Its integration into the drug development pipeline holds the promise of accelerating the delivery of safer and more effective therapies to patients worldwide, marking a significant milestone in the evolution of pharmaceutical science.

Qualify AI Drug Discovery Tools through FDA ISTAND Program to Model Responsible Drug Discovery AI and Mitigate Dual Use Concerns

Artificial intelligence (AI) is poised to revolutionize many fields of science and technology. One field that stands to benefit significantly is drug discovery, which is a time-consuming and expensive process. AI can predict compounds and some of their relevant characteristics, including their efficacy and toxicity. In doing so, AI can help refine the pool of potential compounds that progress in the drug discovery pipeline, while excluding those that will later likely prove to be too toxic or ineffective (Tran et al. 2023). Essentially, AI can make the early stages of drug discovery more efficient by helping to avoid unnecessary human clinical trials and prevent costly, late-stage failures (Tran et al. 2023). Yet, as drug design AI capabilities burgeon, so does the concern that these algorithms could be used for malicious purposes, such as harnessing AI to instead predict compounds that are both highly effective and highly toxic, posing biosecurity risks. Although concerns about the dual-use potential of AI are warranted, there is great potential for AI’s beneficial application in drug discovery, so eliminating the use of AI in this space altogether is undesirable. We recommend that the Food and Drug Administration (FDA) place a special call for submissions of drug design AI with safeguards in place to prevent dual-use to its Innovative Science and Technology Approaches for New Drugs (ISTAND) Pilot Program. This would allow the FDA to open up a line of communication with drug design AI creators, educate the broader public on the potential for dual-use of these technologies and emphasize the need for safeguards, and select a drug design AI that models responsible AI applications for the field at large.

Monitoring the Effects Brought By a Drug on Myocardial Cells Using Field Effect Transistor (FET) Based Sensing Platform

: Ion channels are the membrane proteins that modulate the active flow of ions between cellular environment. These therapeutic targets are expressed in every active cell and can be used as important targets for cellular changes. The ion channels play an important role in different physiological processes including blood regulation, cell proliferation, nerve relaxation, etc. Various cardiac and neurological diseases have been reported to be related to ion channel regulation. They are brought through by the ion channel mutations. Several techniques are currently deployed for monitoring drug effectiveness of ion channels. Among these techniques, patch clamp technology is considered as a golden standard for monitoring the ion channel function. Unfortunately, the conventional patch clamp technologies are not enough to be employed in the early stages of drug discovery. Similarly, serval fluorescence detection methods and combination technologies have been developed for calcium ion channel detection. These assays are not suitable for large scale detection due to less selectivity and sensitivity.At present, field-effect transistor-based biosensors (BioFETs) have been widely employed in various applications due to their non-invasive detection, and high sensitivity. In this work, we adopted an extended gated EDL-FET platform for rapid drug screening on myocardial cell lines (H9c2). The drug effectiveness was studied by tested nifedipine on myocardial cells which were mounted over an EDL-FET platform. Two different surface conditions were studied using Ca2+ treatment and the capability the of EDL-FET sensor as a drug screening platform was discussed.Surface functionalization and Measurements An extended gate chip was used as the sensor array. Before probe immobilization, the sensor electrode was treated with oxygen plasma. Cell attachment over the sensor platform was facilitated using a glycoprotein- Fibronectin. Traut’s reagent was used to modify fibronectin to initiate the covalent attachment to the sensor surface. Optical images and electrical readings were taken to ensure proper functionalization of the fibronectin-attached surface before the cells were seeded over it. Myocardial cells (H9c2 cells), removed from their culture media, were incubated over the functionalized electrodes for ~2 hours following which, electrical measurements were taken to establish a baseline reading using the BioFET device. Different concentrations of nifedipine [0 to 50 μM] and calcium ions [18 μM to 1.8 mM] were used to make testing solutions, which were used for monitoring various changes occurring on the ion channels of the myocardial cells. Figure 1

Emerging microfluidic platforms for crystallization process development

The microfluidic platform offers an exclusive environment compatible with different analytical tools to investigate crystallization processes. Moreover, microfluidic devices allow precise and convenient control of crystallization conditions at nano or meso scales, making them suitable for pharmaceutical and energy applications. These miniaturized screening devices can be effective during drug discovery stages for screening active ingredients available on micro/milligram scales. In this review, we discuss the advancements and applications of microfluidics platforms in the development of crystallization processes, including measurement of crystal growth and nucleation rates, screening of co-crystals, polymorphs and morphology, synthesis of nanocrystals, and determination of metastable zones and solubility. In addition, the latest microfluidics setups for high-throughput screening and lab-on-chip enhancements with sensors and automation are presented in this review. Lastly, an outlook on how microfluidic platforms can be adapted to improve workflow in crystallization process development is presented.

Employing Molecular Conformations for Ligand-Based Virtual Screening with Equivariant Graph Neural Network and Deep Multiple Instance Learning.

Ligand-based virtual screening (LBVS) is a promising approach for rapid and low-cost screening of potentially bioactive molecules in the early stage of drug discovery. Compared with traditional similarity-based machine learning methods, deep learning frameworks for LBVS can more effectively extract high-order molecule structure representations from molecular fingerprints or structures. However, the 3D conformation of a molecule largely influences its bioactivity and physical properties, and has rarely been considered in previous deep learning-based LBVS methods. Moreover, the relative bioactivity benchmark dataset is still lacking. To address these issues, we introduce a novel end-to-end deep learning architecture trained from molecular conformers for LBVS. We first extracted molecule conformers from multiple public molecular bioactivity data and consolidated them into a large-scale bioactivity benchmark dataset, which totally includes millions of endpoints and molecules corresponding to 954 targets. Then, we devised a deep learning-based LBVS called EquiVS to learn molecule representations from conformers for bioactivity prediction. Specifically, graph convolutional network (GCN) and equivariant graph neural network (EGNN) are sequentially stacked to learn high-order molecule-level and conformer-level representations, followed with attention-based deep multiple-instance learning (MIL) to aggregate these representations and then predict the potential bioactivity for the query molecule on a given target. We conducted various experiments to validate the data quality of our benchmark dataset, and confirmed EquiVS achieved better performance compared with 10 traditional machine learning or deep learning-based LBVS methods. Further ablation studies demonstrate the significant contribution of molecular conformation for bioactivity prediction, as well as the reasonability and non-redundancy of deep learning architecture in EquiVS. Finally, a model interpretation case study on CDK2 shows the potential of EquiVS in optimal conformer discovery. The overall study shows that our proposed benchmark dataset and EquiVS method have promising prospects in virtual screening applications.

Ex-vivo drug screening of surgically resected glioma stem cells to replace murine avatars and provide personalise cancer therapy for glioblastoma patients.

With diminishing returns and high clinical failure rates from traditional preclinical and animal-based drug discovery strategies, more emphasis is being placed on alternative drug discovery platforms. Ex vivo approaches represent a departure from both more traditional preclinical animal-based models and clinical-based strategies and aim to address intra-tumoural and inter-patient variability at an earlier stage of drug discovery. Additionally, these approaches could also offer precise treatment stratification for patients within a week of tumour resection in order to direct tailored therapy. One tumour group that could significantly benefit from such ex vivo approaches are high-grade gliomas, which exhibit extensive heterogeneity, cellular plasticity and therapy-resistant glioma stem cell (GSC) niches. Historic use of murine-based preclinical models for these tumours has largely failed to generate new therapies, resulting in relatively stagnant and unacceptable survival rates of around 12-15 months post-diagnosis over the last 50 years. The near universal use of DNA damaging chemoradiotherapy after surgical resection within standard-of-care (SoC) therapy regimens provides an opportunity to improve current treatments if we can identify efficient drug combinations in preclinical models that better reflect the complex inter-/intra-tumour heterogeneity, GSC plasticity and inherent DNA damage resistance mechanisms. We have therefore developed and optimised a high-throughput ex vivo drug screening platform; GliExP, which maintains GSC populations using immediately dissociated fresh surgical tissue. As a proof-of-concept for GliExP, we have optimised SoC therapy responses and screened 30+ small molecule therapeutics and preclinical compounds against tumours from 18 different patients, including multi-region spatial heterogeneity sampling from several individual tumours. Our data therefore provides a strong basis to build upon GliExP to incorporate combination-based oncology therapeutics in tandem with SoC therapies as an important preclinical alternative to murine models (reduction and replacement) to triage experimental therapeutics for clinical translation and deliver rapid identification of effective treatment strategies for individual gliomas.

How can we discover developable antibody-based biotherapeutics?

Antibody-based biotherapeutics have emerged as a successful class of pharmaceuticals despite significant challenges and risks to their discovery and development. This review discusses the most frequently encountered hurdles in the research and development (R&D) of antibody-based biotherapeutics and proposes a conceptual framework called biopharmaceutical informatics. Our vision advocates for the syncretic use of computation and experimentation at every stage of biologic drug discovery, considering developability (manufacturability, safety, efficacy, and pharmacology) of potential drug candidates from the earliest stages of the drug discovery phase. The computational advances in recent years allow for more precise formulation of disease concepts, rapid identification, and validation of targets suitable for therapeutic intervention and discovery of potential biotherapeutics that can agonize or antagonize them. Furthermore, computational methods for de novo and epitope-specific antibody design are increasingly being developed, opening novel computationally driven opportunities for biologic drug discovery. Here, we review the opportunities and limitations of emerging computational approaches for optimizing antigens to generate robust immune responses, in silico generation of antibody sequences, discovery of potential antibody binders through virtual screening, assessment of hits, identification of lead drug candidates and their affinity maturation, and optimization for developability. The adoption of biopharmaceutical informatics across all aspects of drug discovery and development cycles should help bring affordable and effective biotherapeutics to patients more quickly.

Strategies for MALDI-MS method development to investigate different pharmaceutical drug modalities

High-throughput screening (HTS) is an ultra-high throughput tool that allows the identification of diverse, pharmaceutically active compounds in the early stages of drug discovery, by enabling chemists to screen reactions at micromole/nanomole scales. To eliminate the bottleneck arising from analysis of large libraries, Matrix Assisted Laser Desorption Ionization Mass Spectrometry (MALDI-MS) can be utilized as a compatible, ultra-fast, label-free, analysis tool. Upon integrating this technique with suitable liquid handling robots that can accurately spot matrices and analytes on a MALDI plate, chemical reactions can be screened at a sampling rate of <1 s per sample. While earlier efforts to improve MALDI-HTS applications were geared towards the development of automation and data processing tools, a systematic method development study across different drug modalities was needed to understand the contribution of different parameters such as MALDI matrices, choice of matrices, additives, and solvent composition, on MALDI-MS signal intensities of analytes. The current study will demonstrate the importance of initial MALDI-MS method development on small molecules, peptides, antibodies, and antibody drug conjugates. This study is designed as a starting point for MALDI-MS signal optimization, which can eventually lead to the development of effective MALDI-HTS methods for various modalities, thus paving the way for a wide range of applications, including but not limited to conformational analysis.

An accurate and universal protein-small molecule batch docking solution using Autodock Vina

As an important theoretical computation method in computer-aided drug design, molecular docking has significantly shifted the paradigm of drug discovery. As one of the open-source docking software, Autodock Vina (Vina) is widely popular, but the lack of relevant experience and inappropriate docking parameters make it unable to perform optimally in practical application scenarios, which leads to potential failure risks in the early stage of drug discovery. In order to simplify the docking steps and determine the most appropriate docking parameters, a universal solution for rigid receptor docking using Vina has been proposed in this paper, and a user-friendly software for the entire process of molecular docking using Vina is developed. The case studies show that our docking solution is able to be applied to different docking scenarios to facilitate a more accurate, faster, and more convenient new drug discovery process.

CMMS-GCL: Cross-Modality Metabolic Stability Prediction with Graph Contrastive Learning.

Metabolic stability plays a crucial role in the early stages of drug discovery and development. Accurately modeling and predicting molecular metabolic stability has great potential for the efficient screening of drug candidates as well as the optimization of lead compounds. Considering wet-lab experiment is time-consuming, laborious, and expensive, in silico prediction of metabolic stability is an alternative choice. However, few computational methods have been developed to address this task. In addition, it remains a significant challenge to explain key functional groups determining metabolic stability. To address these issues, we develop a novel cross-modality graph contrastive learning model named CMMS-GCL for predicting the metabolic stability of drug candidates. In our framework, we design deep learning methods to extract features for molecules from two modality data, i.e., SMILES sequence and molecule graph. In particular, for the sequence data, we design a multi-head attention BiGRU-based encoder to preserve the context of symbols to learn sequence representations of molecules. For the graph data, we propose a graph contrastive learning-based encoder to learn structure representations by effectively capturing the consistencies between local and global structures. We further exploit fully connected neural networks to combine the sequence and structure representations for model training. Extensive experimental results on two datasets demonstrate that our CMMS-GCL consistently outperforms seven state-of-the-art methods. Furthermore, a collection of case studies on sequence data and statistical analyses of the graph structure module, strengthens the validation of the interpretability of crucial functional groups recognized by CMMS-GCL. Overall, CMMS-GCL can serve as an effective and interpretable tool for predicting metabolic stability, identifying critical functional groups, and thus facilitating the drug discovery process and lead compound optimization. The code and data underlying this article are freely available at https://github.com/dubingxue/CMMS-GCL. Supplementary data are available at Bioinformatics online.

G4-QuadScreen: A Computational Tool for Identifying Multi-Target-Directed Anticancer Leads against G-Quadruplex DNA.

The study presents 'G4-QuadScreen', a user-friendly computational tool for identifying MTDLs against G4s. Also, it offers a few hit MTDLs based on in silico and in vitro approaches. Multi-tasking QSAR models were developed using linear discriminant analysis and random forest machine learning techniques for predicting the responses of interest (G4 interaction, G4 stabilization, G4 selectivity, and cytotoxicity) considering the variations in the experimental conditions (e.g., G4 sequences, endpoints, cell lines, buffers, and assays). A virtual screening with G4-QuadScreen and molecular docking using YASARA (AutoDock-Vina) was performed. G4 activities were confirmed via FRET melting, FID, and cell viability assays. Validation metrics demonstrated the high discriminatory power and robustness of the models (the accuracy of all models is ~>90% for the training sets and ~>80% for the external sets). The experimental evaluations showed that ten screened MTDLs have the capacity to selectively stabilize multiple G4s. Three screened MTDLs induced a strong inhibitory effect on various human cancer cell lines. This pioneering computational study serves a tool to accelerate the search for new leads against G4s, reducing false positive outcomes in the early stages of drug discovery. The G4-QuadScreen tool is accessible on the ChemoPredictionSuite website.

Affinity Chromatographic Method for Determining Drug-Protein Interaction with Enhanced Speed Than Typical Frontal Analysis.

Revealing drug-protein interaction is highly important to select a drug candidate with improved drug-like properties in the early stages of drug discovery. This highlights the urgent need to develop assays that enable the analysis of drug-protein interaction with high speed. Herein, this purpose was realized by the development of an affinity chromatographic method with a two-fold higher speed than typical assays like frontal analysis and zonal elution. The method involved synthesis of a stationary phase by immobilizing poly(ADP-ribose) polymerase-1 (PARP1) onto macroporous silica gel through a one-step bioorthogonal reaction, characterization of mutual displacement interaction of two canonical drugs to the immobilized PARP1, determination of the interaction between three (iniparib, rucaparib, and olaparib) drugs and the protein, and validation of these parameters by typical frontal analysis. The numbers of binding sites on the column were (2.85 ± 0.05) × 10-7, (1.89 ± 0.71) × 10-6, and (1.49 ± 0.06) × 10-7 M for iniparib, rucaparib, and olaparib, respectively. On these sites, the association constants of the three drugs to the protein were (9.85 ± 0.56) × 104, (2.85 ± 0.34) × 104, and (1.07 ± 0.35) × 105 M-1. The determined parameters presented a good agreement with the calculation by typical frontal analyses, which indicated that the current continuous competitive frontal analysis method was reliable for determining drug-protein interaction. Application of the methods was achieved by screening tubeimosides I and II as the bioactive compounds against breast cancer in Bolbostemma paniculatum. Their mechanism may be the interference of DNA repair via down-regulating PARP1 and meiotic recombination 11 expressions, thus leading to oncogene mutations and death of cancer cells. The method was high speed since it allowed simultaneous determination of binding parameters between two drugs and a protein with a smaller number of experiments to be performed. Such a feature made the method an attractive alternative for high-speed analysis of drug-protein interaction or the other bindings in a binary system.

Machine Learning Models to Predict Cytochrome P450 2B6 Inhibitors and Substrates.

Cytochrome P450 2B6 (CYP2B6) is responsible for the metabolism of ∼7% of marketed drugs. The in vitro drug interaction studies guidance for industry issued by the FDA stipulates that drug sponsors need to evaluate whether the investigated drugs interact with the major drug-metabolizing P450s including CYP2B6. Therefore, there has been greater attention to the development of predictive models for CYP2B6 inhibitors and substrates. In this study, conventional machine learning and deep learning models were developed to predict CYP2B6 inhibitors and substrates. Our results showed that the best CYP2B6 inhibitor model yielded the AUC values of 0.95 and 0.75 with the 10-fold cross-validation and the test set, respectively, and the best CYP2B6 substrate model produced the AUC values of 0.93 and 0.90 with the 10-fold cross-validation and the test set, respectively. The generalization ability of the CYP2B6 inhibitor and substrate models was assessed by using the external validation sets. Several significant substructural fragments relevant to CYP2B6 inhibitors and substrates were detected via frequency substructure analysis and information gain. In addition, the applicability domain of the models was defined by employing a nonparametric method based on the probability density distribution. We anticipate that our results would be useful for the prediction of potential CYP2B6 inhibitors and substrates in the early stage of drug discovery.

Kinetic profiling of novel spirobenzo-oxazinepiperidinone derivatives as equilibrative nucleoside transporter 1 inhibitors

Evaluation of kinetic parameters of drug-target binding, kon, koff, and residence time (RT), in addition to the traditional in vitro parameter of affinity is receiving increasing attention in the early stages of drug discovery. Target binding kinetics emerges as a meaningful concept for the evaluation of a ligand’s duration of action and more generally drug efficacy and safety. We report the biological evaluation of a novel series of spirobenzo-oxazinepiperidinone derivatives as inhibitors of the human equilibrative nucleoside transporter 1 (hENT1, SLC29A1). The compounds were evaluated in radioligand binding experiments, i.e., displacement, competition association, and washout assays, to evaluate their affinity and binding kinetic parameters. We also linked these pharmacological parameters to the compounds’ chemical characteristics, and learned that separate moieties of the molecules governed target affinity and binding kinetics. Among the 29 compounds tested, 28 stood out with high affinity and a long residence time of 87 min. These findings reveal the importance of supplementing affinity data with binding kinetics at transport proteins such as hENT1.

From the Bush to the Brain: Preclinical Stages of Ethnobotanical Anti-Inflammatory and Neuroprotective Drug Discovery—An Australian Example

The Australian rainforest is a rich source of medicinal plants that have evolved in the face of dramatic environmental challenges over a million years due to its prolonged geographical isolation from other continents. The rainforest consists of an inherent richness of plant secondary metabolites that are the most intense in the rainforest. The search for more potent and more bioavailable compounds from other plant sources is ongoing, and our short review will outline the pathways from the discovery of bioactive plants to the structural identification of active compounds, testing for potency, and then neuroprotection in a triculture system, and finally, the validation in an appropriate neuro-inflammatory mouse model, using some examples from our current research. We will focus on neuroinflammation as a potential treatment target for neurodegenerative diseases including multiple sclerosis (MS), Parkinson’s (PD), and Alzheimer’s disease (AD) for these plant-derived, anti-inflammatory molecules and highlight cytokine suppressive anti-inflammatory drugs (CSAIDs) as a better alternative to conventional nonsteroidal anti-inflammatory drugs (NSAIDs) to treat neuroinflammatory disorders.

Artificial Intelligence in Drug Discovery and Delivery: Advancements and Applications

Artificial Intelligence (AI) has made significant progress in drug discovery and drug delivery and has become an active area of research. The use of AI in drug delivery has gained significant attention, with the development of new technologies and algorithms that enable more efficient drug delivery. The history of AI in drug discovery can be traced back to the 1960s, and since then, AI has been used in various stages of drug discovery, including target identification, lead optimization, and drug design. AI can aid in different stages of drug development, including drug discovery, formulation, and optimization. AlphaFold is an AI-powered deep learning system that uses neural networks to predict the structure of proteins to design drug molecules. The ultimate goal of using AI in pharmaceuticals is to minimize cost and time and achieve exceptional results without the need for laboratory presence. However, experimental validation is still necessary to verify the accuracy and reliability of the AI-generated results.

High-throughput screening of LogD by using a sample pooling approach based on the traditional shake flask method

A new approach for screening LogD is presented. The method is based on the shake flask method combined with rapid generic LC-MS/MS bioanalysis by using a sample pooling approach that enables high-throughput screening of LogD or LogP in the drug discovery stage. The method is evaluated by a comparison of measured LogD between single and pooled compounds for a test set of structurally diverse compounds with a wide range of LogD values (from −0.04 to 6.01). Test compounds include 10 commercially available drug standards along with 27 new chemical entities. A good correlation (RMSE = 0.21, R2 = 0.9879) of LogD between the single and pooled compounds was obtained, suggesting that at least 37 compounds can be simultaneously measured with acceptable accuracy. The sample pooling method significantly reduced the number of bioanalysis samples as compared to the single compound measurement by the conventional shake flask method. The impact of DMSO content on LogD measurement was also investigated and the result demonstrated that at least 0.5% DMSO was tolerated in this method. The current new development will facilitate the drug discovery process by more rapidly assessing the LogD or LogP of drug candidates.

Prediction of Sites of Metabolism of CYP3A4 Substrates Utilizing Docking-Derived Geometric Features.

Cytochrome P450 3A4 (CYP3A4) is one of the major drug-metabolizing enzymes in the human body and is responsible for the metabolism of ∼50% of clinically used drugs. Therefore, the identification of the compound's sites of metabolism (SOMs) mediated by CYP3A4 is of utmost importance in the early stage of drug discovery and development. Herein, docking-based approaches incorporating geometric features were used for SOMs prediction of CYP3A4 substrates. The cross-docking poses of a relatively large data set containing 474 substrates were analyzed in depth, and a widely observed geometric pattern called the close proximity of SOMs was derived from the poses. On the basis of the close proximity, several structure-based models have been constructed, which demonstrated better performance than those structure-based models using the criterion of Fe-SOM distance. For further improving the prediction performance, the structure-based models were also combined with the well-known ligand-based model SMARTCyp. One combined model exhibited good performance on the SOMs prediction of an external substrate set containing kinase inhibitors, PROTACs, approved drugs, and some lead compounds.

An alternative biomimetic tool – Dual hydrophilic/reversed-phase interaction mode

During the early stages of drug discovery, it becomes increasingly important to acquire a basic physico-chemical and pharmacokinetic profile of drug candidates using efficient and cost-effective laboratory and computational models. In this study, the possibility of applying the dual hydrophilic interaction liquid chromatography/reversed-phase mode (HILIC/RP) as a alternative biomimetic tool for the preliminary profiling of imidazoline and piperazine-related compounds was investigated. Their retention behaviour on the mixed-mode HILIC/RP column was described as a dependence of the logarithm of the retention factor (logk) in relation to the volume fraction of the water component in the mobile phase (φ). The retention was quantified by integration of HILIC (logkH), RP (logkR) and the total HILIC/RP (logkA) area under the logk = f(φ) curve. The biomimetic capabilities of HILIC/RP mode were observed through physico-chemical and pharmacokinetic interpretation of chromatographic behaviour. The physico-chemical characteristics of the HILIC/RP mode were defined based on linear solvation energy relationships, genetic algorithm-multiple linear regression and Boruta-XGBoost predictive modeling. The created models showed reliable predictive performances for defining the logkA, logkR, and logkH values. The mixed-mode logkA, logkR, and logkH parameters correlate well with physico-chemical (e.g., lipophilicity indices) and basic pharmacokinetic properties of the examined compounds. The models created have confirmed that the simultaneous contribution of both HILIC/RP mechanisms can be used to accurately evaluate the key properties of newly synthesized imidazoline and piperazine-related compounds while reducing costs and selecting suitable candidates in the early stages of drug development and discovery.

Improved anti-tumor activity of fluorinated camptothecin derivatives 9-fluorocamptothecin and 7-ethyl-9-fluorocamptothecin on hepatocellular carcinoma by targeting topoisomerase I

Primary liver cancer is one of the most common malignant cancers of the digestive system that lacks effective chemotherapeutic drugs in clinical settings. Camptothecin (CPT) and its derivatives have been approved for cancer treatment; however, their application is limited by their systemic toxicity. For lead optimization in new drug discovery stages, fluorination is an effective and robust approach to increase the bioavailability and optimize the pharmacokinetics of candidate compounds, thereby improving their efficacy. To obtain new and highly active CPT derivatives, we designed, synthesized, and evaluated two new fluorinated CPT derivatives, 9-fluorocamptothecin (A1) and 7-ethyl-9-fluorocamptothecin (A2), in this study. In vitro, A1 and A2 exhibited more robust anti-tumor activity than topotecan (TPT) in various cancer cells, particularly hepatocellular carcinoma (HCC) cells. In vivo, A1 and A2 exhibited greater anti-tumor activity than TPT in both AKT/Met induced primary HCC mouse models and implanted HepG2 cell xenografts. Acute toxicity tests revealed that A1 and A2 were not lethal and did not cause significant body weight loss at high doses. Moreover, A1 and A2 exhibited no significant toxicity in the mouse liver, heart, lung, spleen, kidney, and hematopoietic systems at therapeutic doses. Mechanistically, A1 and A2 blocked HCC cell proliferation by inhibiting the enzymatic activity of Topo I, subsequently inducing DNA damage, cell cycle arrest, and apoptosis. In summary, our results indicate that fluorination improves the anti-tumor activity of CPT while decreasing its toxicity and highlight the application potential of fluorination products A1 and A2 in clinical settings.