Unlabeled Compounds Research Articles

Assessing Popper Purity—Implications for the Regulation and Recreational Use of Alkyl Nitrites

Alkyl nitrites (“poppers”) are a diverse class of volatile chemical compounds with a varied legal and medical history. Though once commonly prescribed to treat angina, popper use is now almost exclusively recreational. Currently, poppers are widely available and sold legally under labels like “solvent cleaner”, despite marketing suggesting they are meant to be consumed. As a result, there is little incentive for producers to implement robust quality controls to protect users. In this study, nine common popper brands were analyzed using hydrogen-1 and carbon-13 nuclear magnetic resonance spectroscopy to assess the presence of impurities. Physical labels on all nine samples indicated the contents were “pure” isobutyl nitrite, despite contradictory online marketing in several cases. Spectral results showed isobutyl nitrite was present in all popper samples. However, there was evidence that various unlabeled compounds were also present in all samples. The identity and concentration of these contaminants were not clear, but the seemingly ubiquitous presence of impurities and lack of consistency in the tested samples are concerning and may represent a threat to users’ health. We hope the results of this study draw attention to the potential dangers of recreational popper use and the need to reassess how these compounds are regulated.

Pitfalls in the Detection of Volatiles Associated with Heated Tobacco and e-Vapor Products When Using PTR-TOF-MS.

We investigated the applicability of proton transfer reaction-time-of-flight mass spectrometry (PTR-TOF-MS) for quantitative analysis of mixtures comprising glycerin, acetol, glycidol, acetaldehyde, acetone, and propylene glycol. While PTR-TOF-MS offers real-time simultaneous determination, the method selectivity is limited when analyzing compounds with identical elemental compositions or when labile compounds present in the mixture produce fragments that generate overlapping ions with other matrix components. In this study, we observed significant fragmentation of glycerin, acetol, glycidol, and propylene glycol during protonation via hydronium ions (H3O+). Nevertheless, specific ions generated by glycerin (m/z 93.055) and propylene glycol (m/z 77.060) enabled their selective detection. To thoroughly investigate the selectivity of the method, various mixtures containing both isotope-labeled and unlabeled compounds were utilized. The experimental findings demonstrated that when samples contained high levels of glycerin, it was not feasible to perform time-resolved analysis in H3O+ mode for acetaldehyde, acetol, and glycidol. To overcome the observed selectivity limitations associated with the H3O+ reagent ions, alternative ionization modes were investigated. The ammonium ion mode proved appropriate for analyzing propylene glycol (m/z 94.086) and acetone (m/z 76.076) mixtures. Concerning the nitric oxide mode, specific m/z were identified for acetaldehyde (m/z 43.018), acetone (m/z 88.039), glycidol (m/z 73.028), and propylene glycol (m/z 75.044). It was concluded that considering the presence of multiple product ions and the potential influence of other compounds, it is crucial to conduct a thorough selectivity assessment when employing PTR-TOF-MS as the sole method for analyzing compounds in complex matrices of unknown composition.

Development and Characterization of a Novel Carbon-11 Labeled Positron Emission Tomography Radiotracer for Neuroimaging of Sirtuin 1 with Benzoxazine-Based Compounds.

Sirtuins (SIRTs) comprise a group of histone deacetylase enzymes crucial for regulating metabolic pathways and contributing significantly to various disease mechanisms. Sirtuin 1 (SIRT1), among the seven known mammalian homologs, is extensively investigated and understood, playing a key role in neurodegenerative disorders and cancer. This study focuses on potential as a therapeutic target for conditions such as Parkinson's disease (PD), Huntington's disease (HD), and Alzheimer's disease (AD). Utilizing positron emission tomography (PET) as a noninvasive molecular imaging modality, we aimed to expedite the validation of a promising sirtuin 1 inhibitor for clinical trials. However, the absence of a validated sirtuin 1 PET radiotracer impedes clinical translation. We present the development of [11C]1, and 11C-labeled benzoxazine-based derivative, as a lead imaging probe. The radiosynthesis of [11C]1 resulted in a radiochemical yield of 31 ± 4%. Baseline studies demonstrated that [11C]1 exhibited excellent blood-brain barrier (BBB) penetration capability, with uniform accumulation throughout various brain regions. Self-blocking studies revealed that introducing an unlabeled compound 1, effectively blocking sirtuin 1, led to a substantial reduction in whole-brain uptake, emphasizing the in vivo specificity of [11C]1 for sirtuin 1. The development of [11C]1 provides a valuable tool for noninvasive imaging investigations in rodent models with aberrant sirtuin 1 expression. This novel radiotracer holds promise for advancing our understanding of sirtuin 1's role in disease mechanisms and may facilitate the validation of sirtuin 1 inhibitors in clinical trials.

Predicting lifespan-extending chemical compounds for C. elegans with machine learning and biologically interpretable features.

Recently, there has been a growing interest in the development of pharmacological interventions targeting ageing, as well as in the use of machine learning for analysing ageing-related data. In this work, we use machine learning methods to analyse data from DrugAge, a database of chemical compounds (including drugs) modulating lifespan in model organisms. To this end, we created four types of datasets for predicting whether or not a compound extends the lifespan of C. elegans (the most frequent model organism in DrugAge), using four different types of predictive biological features, based on: compound-protein interactions, interactions between compounds and proteins encoded by ageing-related genes, and two types of terms annotated for proteins targeted by the compounds, namely Gene Ontology (GO) terms and physiology terms from the WormBase's Phenotype Ontology. To analyse these datasets, we used a combination of feature selection methods in a data pre-processing phase and the well-established random forest algorithm for learning predictive models from the selected features. In addition, we interpreted the most important features in the two best models in light of the biology of ageing. One noteworthy feature was the GO term "Glutathione metabolic process", which plays an important role in cellular redox homeostasis and detoxification. We also predicted the most promising novel compounds for extending lifespan from a list of previously unlabelled compounds. These include nitroprusside, which is used as an antihypertensive medication. Overall, our work opens avenues for future work in employing machine learning to predict novel life-extending compounds.

Development of a PET Tracer for OGA with Improved Kinetics in the Living Brain.

O-GlcNAcylation is thought to play a role in the development of tau pathology in Alzheimer's disease because of its ability to modulate tau's aggregation propensity. O-GlcNAcylation is regulated by 2 enzymes: O-GlcNAc transferase and O-GlcNAcase (OGA). Development of a PET tracer would therefore be an essential tool for developing therapeutic small-molecule inhibitors of OGA, enabling clinical testing of target engagement and dose selection. Methods: A collection of small-molecule compounds was screened for inhibitory activity and high-affinity binding to OGA, as well as favorable PET tracer attributes (multidrug resistance protein 1 efflux, central nervous system PET multiparameter optimization, etc.). Two lead compounds with high affinity and selectivity for OGA were selected for further profiling, including OGA binding to tissue homogenate using a radioligand competition binding assay. In vivo pharmacokinetics were established using a microdosing approach with unlabeled compounds in rats. In vivo imaging studies were performed in rodents and nonhuman primates (NHPs) with 11C-labeled compounds. Results: Two selected candidates, BIO-735 and BIO-578, displayed promising attributes invitro. After radiolabeling with tritium, [3H]BIO-735 and [3H]BIO-578 binding in rodent brain homogenates demonstrated dissociation constants of 0.6 and 2.3 nM, respectively. Binding was inhibited, concentration-dependently, by homologous compounds and thiamet G, a well-characterized and structurally diverse OGA inhibitor. Imaging studies in rats and NHPs showed both tracers had high uptake in the brain and inhibition of binding to OGA in the presence of a nonradioactive compound. However, only BIO-578 demonstrated reversible binding kinetics within the time frame of a PET study with a 11C-labeled molecule to enable quantification using kinetic modeling. Specificity of tracer uptake was confirmed with a 10 mg/kg blocking dose of thiamet G. Conclusion: We describe the development and testing of 2 11C PET tracers targeting the protein OGA. The lead compound BIO-578 demonstrated high affinity and selectivity for OGA in rodent and human postmortem brain tissue, leading to its further testing in NHPs. NHP PET imaging studies showed that the tracer had excellent brain kinetics, with full inhibition of specific binding by thiamet G. These results suggest that the tracer [11C]BIO-578 is well suited for further characterization in humans.

Demonstration of the Early Cardiac Bioavailability of a Non-Specific Cell-Targeted Peptide Using Radionuclide-Based Imaging In Vivo.

The cardiac bioavailability of peptide drugs that inhibit harmful intracellular protein-protein interactions in cardiovascular diseases remains a challenging task in drug development. This study investigates whether a non-specific cell-targeted peptide drug is available in a timely manner at its intended biological destination, the heart, using a combined stepwise nuclear molecular imaging approach. An octapeptide (heart8P) was covalently coupled with the trans-activator of transcription (TAT) protein transduction domain residues 48-59 of human immunodeficiency virus-1 (TAT-heart8P) for efficient internalization into mammalian cells. The pharmacokinetics of TAT-heart8P were evaluated in dogs and rats. The cellular internalization of TAT-heart8P-Cy(5.5) was examined on cardiomyocytes. The real-time cardiac delivery of 68Ga-NODAGA-TAT-heart8P was tested in mice under physiological and pathological conditions. Pharmacokinetic studies of TAT-heart8P in dogs and rats revealed a fast blood clearance, high tissue distribution, and high extraction by the liver. TAT-heart-8P-Cy(5.5) was rapidly internalized in mouse and human cardiomyocytes. Correspondingly, organ uptake of hydrophilic 68Ga-NODAGA-TAT-heart8P occurred rapidly after injection with an initial cardiac bioavailability already 10 min post-injection. The saturable cardiac uptake was revailed by the pre-injection of the unlabeled compound. The cardiac uptake of 68Ga-NODAGA-TAT-heart8P did not change in a model of cell membrane toxicity. This study provides a sequential stepwise workflow to evaluate the cardiac delivery of a hydrophilic, non-specific cell-targeting peptide. 68Ga-NODAGA-TAT-heart8P showed rapid accumulation in the target tissue early after injection. The implementation of PET/CT radionuclide-based imaging methodology as a means to assess effective and temporal cardiac uptake represents a useful and critical application in drug development and pharmacological research and can be extended to the evaluation of comparable drug candidates.

A novel BRET-based assay to investigate binding and residence time of unmodified ligands to the human lysosomal ion channel TRPML1 in intact cells

Here, we report a bioluminescence resonance energy transfer (BRET) assay as a novel way to investigate the binding of unlabeled ligands to the human transient receptor potential mucolipin 1 (hTRPML1), a lysosomal ion channel involved in several genetic diseases and cancer progression. This novel BRET assay can be used to determine equilibrium and kinetic binding parameters of unlabeled compounds to hTRPML1 using intact human-derived cells, thus complementing the information obtained using functional assays based on ion channel activation. We expect this new BRET assay to expedite the identification and optimization of cell-permeable ligands that interact with hTRPML1 within the physiologically relevant environment of lysosomes.

Does Perdeuteration Increase the Polarity of Janus Face Cycloalkanes?

AbstractStimulated by a suggestion of the late Professor Jack D. Dunitz, that perdeuterated Janus face cycloalkanes may be more polar than their unlabelled forms, the deuterated isotopologue of all cis‐1,2,3,4,5,6‐hexafluorocyclohexane ([2H6]‐1a) and all cis‐1,2,3,4‐tetrafluorocyclopentane ([2H6]‐3a) were prepared. Computation at the B3LYP−D3 level indicated that [2H6]‐1a is not more polar than its protio form 1, however perdeuterated cyclopentane [2H6]‐3a may indeed be more polar than 3, although the magnitude is predicted to be small. None‐the‐less retention time analysis on a reverse phase GC/MS column of an add‐mix of 3 and [2H6]‐3a gave some indication that the per‐deuterated isotopologue 3a was detected marginally before the unlabelled compound consistent with increased polarity associated with perdeuteration.

Exploring chemical space for lead identification by propagating on chemical similarity network

MotivationLead identification is a fundamental step to prioritize candidate compounds for downstream drug discovery process. Machine learning (ML) and deep learning (DL) approaches are widely used to identify lead compounds using both chemical property and experimental information. However, ML or DL methods rarely consider compound similarity information directly since ML and DL models use abstract representation of molecules for model construction. Alternatively, data mining approaches are also used to explore chemical space with drug candidates by screening undesirable compounds. A major challenge for data mining approaches is to develop efficient data mining methods that search large chemical space for desirable lead compounds with low false positive rate. ResultsIn this work, we developed a network propagation (NP) based data mining method for lead identification that performs search on an ensemble of chemical similarity networks. We compiled 14 fingerprint-based similarity networks. Given a target protein of interest, we use a deep learning-based drug target interaction model to narrow down compound candidates and then we use network propagation to prioritize drug candidates that are highly correlated with drug activity score such as IC50. In an extensive experiment with BindingDB, we showed that our approach successfully discovered intentionally unlabeled compounds for given targets. To further demonstrate the prediction power of our approach, we identified 24 candidate leads for CLK1. Two out of five synthesizable candidates were experimentally validated in binding assays. In conclusion, our framework can be very useful for lead identification from very large compound databases such as ZINC.

Toward in vivo proof of binding of 18F-labeled inhibitor [18F]TRACK to peripheral tropomyosin receptor kinases

BackgroundTropomyosin receptor kinases (TrkA, TrkB, TrkC) are a family of tyrosine kinases primarily expressed in neuronal cells of the brain. Identification of oncogenic alterations in Trk expression as a driver in multiple tumor types has increased interest in their role in human cancers. Recently, first- and second-generation 11C and 18F-labeled Trk inhibitors, e.g., [18F]TRACK, have been developed. The goal of the present study was to analyze the direct interaction of [18F]TRACK with peripheral Trk receptors in vivo to prove its specificity for use as a functional imaging probe.MethodsIn vitro uptake and competition experiments were carried out using the colorectal cancer cell line KM12. Dynamic PET experiments were performed with [18F]TRACK, either alone or in the presence of amitriptyline, an activator of Trk, entrectinib, a Trk inhibitor, or unlabeled reference compound TRACK in KM12 tumor-bearing athymic nude mice as well as B6129SF2/J and corresponding B6;129S2-Ntrk2tm1Bbd/J mice. Western blot and immunohistochemistry experiments were done with KM12 tumors, brown adipose tissue (BAT), and brain tissue samples.ResultsUptake of [18F]TRACK was increasing over time reaching 208 ± 72% radioactivity per mg protein (n = 6/2) after 60 min incubation time. Entrectinib and TRACK competitively blocked [18F]TRACK uptake in vitro (IC50 30.9 ± 3.6 and 29.4 ± 9.4 nM; both n = 6/2). [18F]TRACK showed uptake into KM12 tumors (SUVmean,60 min 0.43 ± 0.03; n = 6). Tumor-to-muscle ratio reached 0.9 (60 min) and 1.2 (120 min). In TrkB expressing BAT, [18F]TRACK uptake reached SUVmean,60 min 1.32 ± 0.08 (n = 7). Activation of Trk through amitriptyline resulted in a significant radioactivity increase of 21% in KM12 tumor (SUVmean,60 min from 0.53 ± 0.01 to 0.43 ± 0.03; n = 6; p < 0.05) and of 21% in BAT (SUVmean,60 min from 1.32 ± 0.08; n = 5 to 1.59 ± 0.07; n = 6; p < 0.05) respectively. Immunohistochemistry showed TrkB > TrkA expression on BAT fat cells, but TrkA > TrkB in whole brain. WB analysis showed sevenfold higher TrkB expression in BAT versus KM12 tumor tissue.ConclusionThe present data show that radiotracer [18F]TRACK can target peripheral Trk receptors in human KM12 colon cancer as well as brown adipose tissue as confirmed through in vitro and in vivo blocking experiments. Higher TrkB versus TrkA protein expression was detected in brown adipose tissue of mice confirming a peripheral functional role of brain-derived neurotrophic factor in adipose tissue.

A deep learning approach to discover cyclin-dependent kinases 12 (CDK12) inhibitors in breast cancer.

e15086 Background: Although anti-HER2 tyrosine kinase inhibitors (TKIs) have significantly prolonged survival and improved prognosis in HER2-positive breast cancer patients, resistance is a constant obstacle leading to TKIs treatment failure and tumor progression. Our previous research shows dual inhibition of HER2/CDK12 will prominently benefit the outcomes of patients with HER2-positive breast cancer by sensitizing the tumors to anti-HER2 TKIs treatment. Unfortunately, there is no high selective CDK12 inhibitor in clinical. This study sought to accelerate the CDK12 inhibitor discovery by a deep learning approach. Methods: We developed a large-scale self-supervised graph neural network to learn 11 million unlabelled compounds for molecular representation. And the model was trained with 233,823 kinase domain sequences by drug-target interaction datasets from the BindingDB database. Then, this transformer architecture was used to predict potential CDK12 inhibitors. The candidate hits compounds with diverse skeletons were divided into 50 clusters by clustering analysis. The representative hits with high transform scores and rational binding mode in each cluster were detected in the homogeneous time-resolved fluorescence CDK12 kinase assay. Candidate clusters with half-maximal inhibitory concentrations (IC50) lower than 200 nM were selected into expanded validation and kinase selective panel detection. xCELLigence RTCA system was adopted to monitor cell growth and sensitivity of breast cancer cells to CDK12 inhibitors in a real-time manner. Results: Our predictive transfer learning model yielded satisfactory predictions on various targets including CDK12. We screened a total of 4,527,236 compounds and recommended 50 clusters of potential CDK12 inhibitors for further enzyme activity detection. The IC50 of seven cluster representative compounds is lower than 10 uM. Expanded screening of 500 compounds in these clusters discovered a lot of hits with highly CDK12 selective inhibition. Half of the new compounds are showed promising proliferation inhibition in breast cancer cell lines. Conclusions: Our study provides a highly efficient and end-to-end deep learning approach to discover highly selective CDK12 inhibitors in breast cancer.

Synthesis and characterization of the two enantiomers of a chiral sigma-1 receptor radioligand: (S)-(+)- and (R)-(-)-[18F]FBFP

Racemic [18F]FBFP ([18F]1) proved to be a potent σ1 receptor radiotracer with superior imaging properties. The pure enantiomers of unlabeled compounds (S)- and (R)-1 and the corresponding iodonium ylide precursors were synthesized and characterized. The two enantiomers (S)-1 and (R)-1 exhibited comparable high affinity for σ1 receptors and selectivity over σ2 receptors. The Ca2+ fluorescence assay indicated that (R)-1 behaved as an antagonist and (S)-1 as an agonist for σ1 receptors. The 18F-labeled enantiomers (S)- and (R)-[18F]1 were obtained in >99% enantiomeric purity from the corresponding enantiopure iodonium ylide precursors with radiochemical yield of 24.4% ± 2.6% and molar activity of 86–214 GBq/µmol. In ICR mice both (S)- and (R)-[18F]1 displayed comparable high brain uptake, brain-to-blood ratio, in vivo stability and binding specificity in the brain and peripheral organs. In micro-positron emission tomography (PET) imaging studies in rats, (S)-[18F]1 exhibited faster clearance from the brain than (R)-[18F]1, indicating different brain kinetics of the two enantiomers. Both (S)- and (R)-[18F]1 warrant further evaluation in primates to translate a single enantiomer with more suitable kinetics for imaging the σ1 receptors in humans.

Exploration of chemical space with partial labeled noisy student self-training and self-supervised graph embedding

BackgroundDrug discovery is time-consuming and costly. Machine learning, especially deep learning, shows great potential in quantitative structure–activity relationship (QSAR) modeling to accelerate drug discovery process and reduce its cost. A big challenge in developing robust and generalizable deep learning models for QSAR is the lack of a large amount of data with high-quality and balanced labels. To address this challenge, we developed a self-training method, Partially LAbeled Noisy Student (PLANS), and a novel self-supervised graph embedding, Graph-Isomorphism-Network Fingerprint (GINFP), for chemical compounds representations with substructure information using unlabeled data. The representations can be used for predicting chemical properties such as binding affinity, toxicity, and others. PLANS-GINFP allows us to exploit millions of unlabeled chemical compounds as well as labeled and partially labeled pharmacological data to improve the generalizability of neural network models.ResultsWe evaluated the performance of PLANS-GINFP for predicting Cytochrome P450 (CYP450) binding activity in a CYP450 dataset and chemical toxicity in the Tox21 dataset. The extensive benchmark studies demonstrated that PLANS-GINFP could significantly improve the performance in both cases by a large margin. Both PLANS-based self-training and GINFP-based self-supervised learning contribute to the performance improvement.ConclusionTo better exploit chemical structures as an input for machine learning algorithms, we proposed a self-supervised graph neural network-based embedding method that can encode substructure information. Furthermore, we developed a model agnostic self-training method, PLANS, that can be applied to any deep learning architectures to improve prediction accuracies. PLANS provided a way to better utilize partially labeled and unlabeled data. Comprehensive benchmark studies demonstrated their potentials in predicting drug metabolism and toxicity profiles using sparse, noisy, and imbalanced data. PLANS-GINFP could serve as a general solution to improve the predictive modeling for QSAR modeling.

Is Cys(MTSL) the Best α-Amino Acid Residue to Electron Spin Labeling of Synthetically Accessible Peptide Molecules with Nitroxides?

Electron paramagnetic resonance spectroscopy, particularly its pulse technique double electron–electron resonance (DEER) (also termed PELDOR), is rapidly becoming an extremely useful tool for the experimental determination of side chain-to-side chain distances between free radicals in molecules fundamental for life, such as polypeptides. Among appropriate probes, the most popular are undoubtedly nitroxide electron spin labels. In this context, suitable biosynthetically derived, helical regions of proteins, along with synthetic peptides with amphiphilic properties and antibacterial activities, are the most extensively investigated compounds. A strict requirement for a precise distance measurement has been identified in a minimal dynamic flexibility of the two nitroxide-bearing α-amino acid side chains. To this end, in this study, we have experimentally compared in detail the side-chain mobility properties of the two currently most widely utilized residues, namely, Cys(MTSL) and 2,2,6,6-tetramethylpiperidine-1-oxyl-4-amino-4-carboxylic acid (TOAC). In particular, two double-labeled, chemically synthesized 20-mer peptide molecules have been adopted as appropriate templates for our investigation on the determination of the model intramolecular separations. These double-Cys(MTSL) and double-TOAC compounds are both analogues of the almost completely rigid backbone peptide ruler which we have envisaged and 3D structurally analyzed as our original, unlabeled compound. Here, we have clearly found that the TOAC side-chain labels are largely more 3D structurally restricted than the MTSL labels. From this result, we conclude that the TOAC residue offers more precise information than the Cys(MTSL) residue on the side chain-to-side chain distance distribution in synthetically accessible peptide molecules.

Time-course quantitative mapping of caffeine within the epidermis, using high-contrast pump-probe stimulated Raman scattering microscopy.

BackgroundAn assessment of the drug penetration and distribution profiles within the skin is essential in dermatology and cosmetology. Recent advances in label‐free imaging technologies have facilitated the direct detection of unlabeled compounds in tissues, with high resolution. However, it remains challenging to provide quantitative time‐course distribution maps of drugs within the complex skin tissue. The present study aims at acquiring the real‐time quantitative skin penetration profiles of topically applied caffeine, by means of a combination of pump–probe phase‐modulated stimulated Raman scattering (PM‐SRS) and confocal reflection microscopy. The recently developed PM‐SRS microscopy is a unique imaging tool that can minimize strong background signals through a pulse‐shaping technique, while providing high‐contrast images of small molecules in tissues.Materials and methodsReconstructed human skin epidermis models were used in order to analyze caffeine penetration in tissues. The penetration profiles of caffeine in an aqueous solution, an oil‐in‐water gel, and a water‐in‐oil gel were examined by combining PM‐SRS and confocal reflection microscopy.ResultsThe characteristic Raman signal of caffeine was directly detected in the skin model using PM‐SRS. Integrating PM‐SRS and confocal reflection microscopy allowed real‐time concentration maps of caffeine to be obtained from formulation samples, within the skin model. Compared with the conventional Raman detection method, PM‐SRS lowered the background tissue‐oriented signals and supplied high‐contrast images of caffeine.ConclusionWe successfully established real‐time skin penetration profiles of caffeine from different formulations. PM‐SRS microscopy proved to be a powerful, non‐invasive, and real‐time depth‐profile imaging technique for use in quantitative studies of topically applied drugs.

Fully unsupervised deep mode of action learning for phenotyping high-content cellular images.

The identification and discovery of phenotypes from high content screening images is a challenging task. Earlier works use image analysis pipelines to extract biological features, supervised training methods or generate features with neural networks pretrained on non-cellular images. We introduce a novel unsupervised deep learning algorithm to cluster cellular images with similar Mode-of-Action (MOA) together using only the images' pixel intensity values as input. It corrects for batch effect during training. Importantly, our method does not require the extraction of cell candidates and works from the entire images directly. The method achieves competitive results on the labeled subset of the BBBC021 dataset with an accuracy of 97.09% for correctly classifying the MOA by nearest neighbors matching. Importantly, we can train our approach on unannotated datasets. Therefore, our method can discover novel MOAs and annotate unlabeled compounds. The ability to train end-to-end on the full resolution images makes our method easy to apply and allows it to further distinguish treatments by their effect on proliferation. Our code is available at https://github.com/Novartis/UMM-Discovery. Supplementary data are available at Bioinformatics online.

Employing in vitro metabolism to guide design of F-labelled PET probes of novel α-synuclein binding bifunctional compounds

A challenge in the development of novel 18F-labelled positron emission tomography (PET) imaging probes is identification of metabolically stable sites to incorporate the 18F radioisotope. Metabolic loss of 18F from PET probes in vivo can lead to misleading biodistribution data as displaced 18F can accumulate in various tissues. In this study we report on in vitro hepatic microsomal metabolism of novel caffeine containing bifunctional compounds (C8-6-I, C8-6-N, C8-6-C8) that can prevent in vitro aggregation of α-synuclein, which is associated with the pathophysiology of Parkinson’s disease. The metabolic profile obtained guided us to synthesize stable isotope 19F-labelled analogues in which the fluorine was introduced at the metabolically stable N7 of the caffeine moiety. An in vitro hepatic microsomal metabolism study of the 19F-labelled analogues resulted in similar metabolites to the unlabelled compounds and demonstrated that the fluorine was metabolically stable, suggesting that these analogues are appropriate PET imaging probes. This straightforward in vitro strategy is valuable for avoiding costly stability failures when designing radiolabelled compounds for PET imaging.

Synthesis and Initial Characterization of a Selective, Pseudo-irreversible Inhibitor of Human Butyrylcholinesterase as PET Tracer.

The enzyme butyrylcholinesterase (BChE) represents a promising target for imaging probes to potentially enable early diagnosis of neurodegenerative diseases like Alzheimer's disease (AD) and to monitor disease progression in some forms of cancer. In this study, we present the design, facile synthesis, in vitro and preliminary ex vivo and in vivo evaluation of a morpholine‐based, selective inhibitor of human BChE as a positron emission tomography (PET) tracer with a pseudo‐irreversible binding mode. We demonstrate a novel protecting group strategy for 18F radiolabeling of carbamate precursors and show that the inhibitory potency as well as kinetic properties of our unlabeled reference compound were retained in comparison to the parent compound. In particular, the prolonged duration of enzyme inhibition of such a morpholinocarbamate motivated us to design a PET tracer, possibly enabling a precise mapping of BChE distribution.

Preclinical Evaluation of an 18F-Labeled SW-100 Derivative for PET Imaging of Histone Deacetylase 6 in the Brain.

Histone deacetylase 6 (HDAC6), an enzyme involved in protein degradation, exhibits several unique properties, such as cytoplasmic localization and ubiquitin binding. HDAC6 has emerged as an interesting therapeutic target in the treatment of neurodegenerative disorders such as Alzheimer's and Parkinson's diseases. Techniques enabling noninvasive HDAC6 imaging in the brain could enhance understanding of its pathologic role, but development of brain-penetrating radioligands for HDACs imaging by positron emission tomography (PET) remains challenging. Here, we report the synthesis and evaluation of an 18F-labeled tetrahydroquinoline derivative, [18F]2, based on the HDAC6 selective inhibitor SW-100 as a brain HDAC6 imaging radioligand. [18F]2 was synthesized via copper-mediated radiofluorination from an arylboronic precursor, followed by removal of the catalyst by solid-phase extraction and then hydroxamic acid formation. [18F]2 demonstrated good penetration and moderate stability in the mouse brain. In mouse plasma, however, [18F]2 was rapidly metabolized to a corresponding carboxylic acid form. Blocking studies in mice with unlabeled compound 2 and HDAC6 selective inhibitors, including tubastatin A and ACY-775, demonstrated that the HDAC6 inhibitors displaced over 80% of [18F]2 taken up in the brain, indicating selective binding of [18F]2. These results suggest that [18F]2 is a potentially useful PET radioligand for brain HDAC6 imaging.

An Efficient Synthesis of Optically Active [4-13C] Labelled Quorum Sensing Signal Autoinducer-2.

A new synthetic route for the quorum sensing signal Autoinducer-2 (AI-2) is described and used for the preparation of [4-13C]-AI-2 starting from [1-13C]-bromoacetic acid. The key step in this process was the enantioselective reduction of an intermediate ketone. This synthesis provides, selectively, both enantiomers of the labelled or unlabelled parent compound, (R) or (S)-4,5-dihydroxypentane-2,3-dione (DPD) and was used for an improved synthesis of [1-13C]-AI-2.