7196 research outputs found
Sort by
Human iPSC-derived retinal organoid model for in vitro toxicity screening
No full textDrug toxicity screening on retina is essential for the development of safe therapies for a large number of diseases, whilst preserving visual acuity and function. To this end, retinal organoids derived from human pluripotent stem cells (hPSCs) provide a suitable screening platform due to their similarity to human retina and the ease of generation in large-scale formats. In this study, two hPSC cell lines were differrentiated to retinal organoids, which comprised all key retinal cell types in multiple nuclear and synaptic layers. Single cell RNA-Seq of retinal organoids at day 200 of differentiation indicated the maintenance of retinal ganglion cells and development of bipolar cells: both cell types segregated into several subtypes. Ketorolac, Digoxin, Thioridazine, Sildenafil, Ethanol and Methanol were selected as key drugs to screen on retinal organoids because of the retinotoxic effects described in literature. Exposure of the hPSC-derived retinal organoids to Diogxin, Thioridazine and Sildenafil exposure resulted in photoreceptor cell death, while Digoxin and Thioridazine additionally affected all other cell types, including Müller glia cells. Ethanol and Methanol exposure caused an increase in the fraction of retinal ganglion cells. All drug treatments caused activation of astrocytes, indicated by dendrites sprouting into neuroepithelium The ability to resond to light was preserved in organoids although the number of active retinal ganglion cells decreased after drug exposure. These data indicate similar drug effects in organoids to those reported in in vivo models and/or in humans, thus providing the first robust experimental evidence of their suitability for toxicological studies
Splitting chemical structure data sets for federated privacy-preserving machine learning.
No full textWith the increase in applications of machine learning methods in drug design and related fields, the challenge of designing sound test sets becomes more and more prominent. The goal of this challenge is to have a realistic split of chemical structures (compounds) between training, validation and test set such that the performance on the test set is meaningful to infer the performance in a prospective application. This challenge is by its own very interesting and relevant, but is even more complex in a federated machine learning approach where multiple partners jointly train a model under privacy-preserving conditions where chemical structures must not be shared between the different participating parties. In this work we discuss three methods which provide a splitting of a data set and are applicable in a federated privacy-preserving setting, namely: a. locality-sensitive hashing (LSH), b. sphere exclusion clustering, c. scaffold-based binning (scaffold network). For evaluation of these splitting methods we consider the following quality criteria (compared to random splitting): bias in prediction performance, classification label and data imbalance, similarity distance between the test and training set compounds. The main findings of the paper are a. both sphere exclusion clustering and scaffold-based binning result in high quality splitting of the data sets, b. in terms of compute costs sphere exclusion clustering is very expensive in the case of federated privacy-preserving setting
Post-traumatic stress disorder: When fear strikes the heart
No full textDr. O'Donnell has been invited to conduct an interview for an article in the Harvard Heart Letter to summarize salient points from his recent article in JAMA Cardiology about PTSD and Cardiovascular Disease. The article that is the focus of the interview is: https://jamanetwork.com/journals/jamacardiology/fullarticle/2781972?resultClick=1. The interview will be conducted by Julie Corliss the Executive Editor of the Harvard Heart Letter ([email protected] phone 781-367-9310). Dr. O'Donnell may be quoted but will not actually write the article or serve as a co-author. The JAMA Cardiology article that is the basis of the interview does not pertain to Novartis research and no Novartis business will be discussed in the interview. The article was published as "Post-traumatic stress disorder: When fear strikes the heart" in Volume 32 Number 3 November 2021 Harvard Heart Letter
Relative Binding Free-Energy Calculations at Lipid-Exposed Sites: Deciphering Hot Spots.
No full textRelative binding free-energy (RBFE) calculations are experiencing resurgence in the computer-aided drug design of novel small molecules due to performance gains allowed by cutting-edge molecular mechanic force fields and computer hardware. Application of RBFE to soluble proteins is becoming a routine, while recent studies outline necessary steps to successfully apply RBFE at the orthosteric site of membrane-embedded G-protein-coupled receptors (GPCRs). In this work, we apply RBFE to a congeneric series of antagonists that bind to a lipid-exposed, extra-helical site of the P2Y1 receptor. We find promising performance of RBFE, such that it may be applied in a predictive manner on drug discovery programs targeting lipid-exposed sites. Further, by the application of the microkinetic model, binding at a lipid-exposed site can be split into (1) membrane partitioning of the drug molecule followed by (2) binding at the extra-helical site. We find that RBFE can be applied to calculate the free energy of each step, allowing the uncoupling of observed binding free energy from the influence of membrane affinity. This protocol may be used to identify binding hot spots at extra-helical sites and guide drug discovery programs toward optimizing intrinsic activity at the target
Inhibition of MDM2 promotes antitumor responses in p53 wild-type cancer cells through their interaction with the immune and stromal microenvironment
No full textP53 is a transcription factor that plays a central role in guarding the genomic stability of cells through cell-cycle arrest or induction of apoptosis. However, the effects of p53 in antitumor immunity are poorly understood. To investigate the role of p53 in controlling tumor-immune cell cross-talk, we studied murine syngeneic models treated with HDM201, a potent and selective second-generation MDM2 inhibitor. In response to HDM201 treatment, the percentage of dendritic cells increased, including the CD103þ antigen cross-presenting subset. Furthermore, HDM201 increased the percentage of TbetþEomesþ CD8þ T cells and the CD8þ/Treg ratio within the tumor. These immunophenotypic changes were eliminated with the knockout of p53 in tumor cells. Enhanced expression of CD80 on tumor cells was observed in vitro and in vivo, which coincided with T-cell-mediated tumor cell killing. Combining HDM201 with PD-1 or PD-L1 blockade increased the number of complete tumor regressions. Responding mice developed durable, antigen-specific memory T cells and rejected subsequent tumor implantation. Importantly, antitumor activity of HDM201 in combination with PD-1/PD-L1 blockade was abrogated in p53-mutated and knockout syngeneic tumor models, indicating the effect of HDM201 on the tumor is required for triggering antitumor immunity. Taken together, these results demonstrate that MDM2 inhibition triggers adaptive immunity, which is further enhanced by blockade of PD-1/PD-L1 pathway, thereby providing a rationale for combining MDM2 inhibitors and checkpoint blocking antibodies in patients with wild-type p53 tumors
Microglial inclusions and neurofilament light chain release follow neuronal α-synuclein lesions in long-term brain slice cultures
No full textBackground: Proteopathic brain lesions are a hallmark of many age-related neurodegenerative diseases including synucleinopathies and develop at least a decade before the onset of clinical symptoms. Thus, understanding of the initiation and propagation of such lesions is key for developing therapeutics to delay or halt disease progression. Methods: Alpha-synuclein (αS) inclusions were induced in long-term murine and human slice cultures by seeded aggregation. An αS seed-recognizing human antibody was tested for blocking seeding and/or spreading of the αS lesions. Release of neurofilament light chain (NfL) into the culture medium was assessed. Results: To study initial stages of α-synucleinopathies, we induced αS inclusions in murine hippocampal slice cultures by seeded aggregation. Induction of αS inclusions in neurons was apparent as early as 1week post-seeding, followed by the occurrence of microglial inclusions in vicinity of the neuronal lesions at 2–3 weeks. The amount of αS inclusions was dependent on the type of αS seed and on the culture’s genetic background (wildtype vs A53T-αS genotype). Formation of αS inclusions could be monitored by neurofilament light chain protein release into the culture medium, a fluid biomarker of neurodegeneration commonly used in clinical settings. Local microinjection of αS seeds resulted in spreading of αS inclusions to neuronally connected hippocampal subregions, and seeding and spreading could be inhibited by an αS seed-recognizing human antibody. We then applied parameters of the murine cultures to surgical resection-derived adult human long-term neocortical slice cultures from 22 to 61-year-old donors. Similarly, in these human slice cultures, proof-of-principle induction of αS lesions was achieved at 1week post-seeding in combination with viral A53T-αS expressions. Conclusion: The successful translation of these brain cultures from mouse to human with the first reported induction of human αS lesions in a true adult human brain environment underlines the potential of this model to study proteopathic lesions in intact mouse and now even aged human brain environments
R&D efficiency of leading pharmaceutical companies - a 20-year analysis
No full textComparative analysis of the R&D efficiency of 14 leading pharmaceutical companies for the years 1999-2018 shows that there is a close positive correlation between R&D spending and the two investigated R&D output parameters, approved NMEs and the cumulative impact factor of their publications. In other words, higher R&D investments (input) were associated with higher R&D output. Second, our analyses indicate that there are 'economies of scale' (size) in pharmaceutical R&D. In particular, R&D organizations with more than 14,000 employees benefit from their size, deliver increased outputs and, therefore, are more efficient and more competitive
Evaluation of the Absorption, Metabolism, and Excretion of a Single Oral 1 mg Dose of Tropifexor in Healthy Male Subjects and the Concentration Dependency of Tropifexor Metabolism
No full textTropifexor (NVP-LJN452) is a highly potent, selective, non-steroidal, non-bile acid farnesoid X receptor (FXR) agonist for the treatment of nonalcoholic steatohepatitis (NASH). Its absorption, metabolism, and excretion was studied following a 1-mg oral dose of [14C]tropifexor to four healthy male subjects. Mass balance was achieved with ~94% of the administered dose recovered in excreta through 312 hours collection period. Fecal excretion of tropifexor-related radioactivity played a major role (~65% of the total dose). Tropifexor reached a maximum blood concentration (Cmax) of 33.5 ng/mL with a median Tmax of 4 hours and was eliminated with a plasma elimination half-life (T1/2) of 13.5 hours. Unchanged tropifexor was the principal drug-related component found in plasma (~92% of total radioactivity). Two minor oxidative metabolites, M11.6 and M22.4, were observed in circulation. Tropifexor was eliminated predominantly via metabolism with >68% of the dose recovered as metabolites in excreta. Oxidative metabolism appeared to be the major clearance pathway of tropifexor. Metabolites containing multiple oxidative modifications and combined oxidation and glucuronidation were also observed in human excreta. The involvement of direct glucuronidation could not be ruled out, based on previous in vitro and nonclinical in vivo studies indicating its contribution to tropifexor clearance. The relative contribution of the oxidation and glucuronidation pathways appeared to be
dose-dependent upon further in vitro investigation. Due to these complexities and the instability of glucuronide metabolites in the gastrointestinal tract, the contribution of glucuronidation remained undefined in this study
An afucosylated anti-CD32b monoclonal antibody induced platelet-mediated adverse events in human Fcg receptor transgenic mouse model and its potential human translatability
No full textA monoclonal, afucosylated, anti-CD32b (FCGR2B) antibody, NVS32b, was developed internally as a therapeutic candidate for treatment of B-cell malignancies. To assess its safety and tolerability, a humanized transgenic (Tg) mouse model that reportedly expresses all human Fc gamma receptors (FCGRs) while lacking all mouse FCGRs was used. Prior to its use, the model was extensively characterized and found to express all human FCGRs in a pattern similar to humans, with some deviations, such as low CD32 expression on T cells, substantial individual variation in the transgene copy number, integration of additional human genes, and overall higher expression of all FCGRs on myeloid cells compared to human. Unexpectedly, NVS32b induced severe thrombosis in huFCGR mice. The mechanism and relevance for human was further investigated. Species differences, in the NVS32b-driven in vitro platelet binding, activation and aggregation were observed (CD32a-binding by Fc and CDR of NVS32b causing platelet activation in huFCGR mice opposing to CD32a-binding by Fc and off-target-binding by CDR of NVS32b only after platelet activation in human). Therefore huFCGR mice may not be fully predictive of the risk of NVS32b-induced thromboembolic events in the clinic. Nevertheless the program was terminated as a result of this potential safety liability. This model could be considered beneficial in the pre-clinical research of immunotherapies targeting or involving FCGRs. While potential biological implications resulting from the differences in theFCGR expression pattern in humans cannot be predicted, these deviations should be considered and further evaluated when using this huFCGR mouse model
Data Science in R&D: it is about culture!
No full textShort "insight" opinion article advertising a recent paper of ours (OAK #42658