Communications Chemistry最新文献

The transformation of ritonavir form 1 into a less soluble form 2 is the most notorious example of the risks associated with crystal polymorphism in pharmaceuticals. Since then, significant advancements have occurred in the field of theoretical crystal structure prediction, which forecasts the potential polymorphs of a molecule and their stability ranking. However, a question remains whether in silico modeling would have predicted the ritonavir disaster and informed appropriate action. Furthermore, the experimental landscape of ritonavir remains incomplete as no solution of form 4 has been deposited. Here, we show that CSP would have foreseen the existence of more stable then-unfound form 2 of ritonavir at room temperature. From a risk standpoint, the threat posed by this polymorph would have been considered severe due to its unique conformational and structural characteristics, combined with the formulation's low tolerance for solubility reduction. This would have prompted additional work that could have averted the crisis. Furthermore, we determined the crystal structure of form 4 of ritonavir by three-dimensional electron diffraction, combined with in silico modeling and experimental powder X-ray diffraction, revealing a disordered motif and proving it is thermodynamically unstable.

Lanthanide luminescent materials have found a wide range of use in various fields. While most applications utilize intrashell f-f transitions, f-d transitions have recently emerged as a novel approach to develop molecular luminescent materials. Unlike Laporte-forbidden f-f transitions, f-d transitions are Laporte-allowed and thus not dependent on chromophoric ligands. Moreover, the involvement of 5d orbitals renders it possible to modulate the emission wavelength by tuning the lanthanide-ligand interaction. Herein, we report a series of three-coordinated cerium(III) complexes with a trigonal planar geometry supported by bulky 2,6-disubstituted phenoxide ligands. These complexes were fully characterized by X-ray crystallography and ¹H NMR spectroscopy. In addition, we investigated the photophysical properties of these compounds that reveal strong d-f emissions with the photoluminescent quantum yield up to 62% and the CIE value of (0.46, 0.51). This work opens a new avenue to regulate the emission wavelength of lanthanide luminescent molecules through ligand field engineering.

Characterization of protein-ligand interactions is essential for the pre-clinical development of drug candidates and Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS) has emerged as a valuable tool in this process. HDX-MS has predominantly been employed with high affinity compounds with only a few examples of its application for weaker binders such as fragments. Nevertheless, HDX-MS usage could be instrumental in Fragment-Based Drug Discovery (FBDD) programs. In this work, the drug-target protein Cyclophilin D (CypD) was used as a model to explore the boundaries of fragments binding characterization by HDX-MS (fHDX-MS). We performed a systematic study on the optimal conditions for fHDX-MS execution and found that fragments with binding affinities in the double-digit mM range are still amenable to fHDX-MS. We observed that, despite the intrinsic low resolution of HDX-MS, fragments binding sites that partially overlap can still be distinguished. Overall, this study shows that fHDX-MS can be a useful method for FBDD.

Phosphinines, or phosphabenzenes, exhibit distinctive electronic properties yet remain underexplored due to the challenges associated with their selective functionalization. We present herein the straightforward functionalization of λ⁵-phosphinine derivatives using organometallic strategies. Halogen-zinc and -magnesium exchanges were successfully performed employing Et₂Zn·2Oamyl or i-PrMgCl·LiCl species under smooth reaction conditions. Such method allowed access to a wide range of sophisticated architectures, photophysical studies of which demonstrated interesting fluorescence properties. With the possibility of using such fluorescence in biomarking, λ⁵-phosphinines were grafted on a few glycosides, nucleosides and pharmaceutically relevant moieties.

Nuclear magnetism is typically investigated by perturbing the spin system with radio frequency pulses, but low polarization and detection using induction coils limit direct access to the longitudinal magnetization. The hyperpolarization technique SABRE-SHEATH requires ultra-low magnetic fields for spin order transfer; consequently, SQUID sensors with a frequency-independent sensitivity are well-suited for unperturbed detection in this regime. We demonstrate direct observation of hyperpolarization build up (T_B) and spin lattice relaxation (T₁) in [1-¹³C]pyruvate, hyperpolarized with SABRE-SHEATH at 150 nT and 500 nT. The values for T_B of 36 s and 26 s and T₁ of 40 s and 43 s, respectively, suggests a shift in dominant polarization transfer efficacy or complexes, highlighting the method's merit in characterizing hyperpolarization pathways. Moreover, as demand for hyperpolarized probes in metabolic imaging continues to grow, the exceptional time resolution makes direct detection a valuable tool for understanding and optimizing polarization dynamics and reactor designs.

The urgent need to phase out SF₆, an extremely potent greenhouse gas prevalent in electrical grids, drives the search for eco-friendly insulation alternatives. Trifluoromethanesulfonyl fluoride (CF₃SO₂F) emerges as a promising candidate due to its excellent properties. However, understanding its thermal decomposition pathways and products under operationally relevant conditions is critical for evaluating its environmental feasibility and mitigating potential risks upon accidental release or during fault events. This study investigates the thermal decomposition mechanisms of CF₃SO₂F using a deep learning potential that combines ab initio accuracy with empirical MD efficiency. By leveraging machine learning driven molecular dynamics, we systematically analyze the yields and components of decomposition products versus temperatures, gas mixing ratios, and buffer gas. The results reveal that the bond-breaking pathways are temperature-dependent, with both elevated temperatures and higher buffer gas mixing ratios promoting its decomposition. Elevated gas pressure enhances the decomposition process by increasing the collision frequency among reactant species. Additionally, N₂ exhibits an inhibitory effect on decomposition under high pressure compared to CO₂. Experimental validation via a thermal decomposition platform confirms characteristic decomposition products. These findings are pivotal for guiding the rational design and safe deployment of CF₃SO₂F to achieve substantial greenhouse gas mitigation in the power industry.

Compound-protein interaction (CPI) prediction plays a crucial role in drug discovery by aiding the identification of binding and affinities between small molecules and proteins. Current deep learning models rely heavily on sequence-based representations and suffer from a lack of labeled data, which restricts their accuracy and generalizability. To overcome these challenges, we propose GenSPARC (a model with Generalized Structure- and Property-Aware Representations of protein and chemical language models for CPI prediction), a deep learning model that leverages structure-aware protein representations derived from AlphaFold2 predictions and FoldSeek's three-dimensional interaction alphabet. Compound features were extracted using graph convolutional networks and a pretrained chemical language model, thereby ensuring comprehensive multimodal representation. An attention mechanism further enhanced interaction modeling by capturing intricate binding patterns. GenSPARC was validated successfully with multiple CPI benchmark datasets, demonstrating strong generalizability across challenging data splits and competitive results in virtual screening tasks. Therefore, GenSPARC will substantially advance artificial intelligence-driven drug discovery.

Understanding the fundamental principles of spin-electric coupling in molecules with hyperfine-coupled electronic and nuclear spins offers a route to electric field-based molecular quantum information. We recently addressed the electronic degrees of freedom in $\left[\mathrm{Tm}{\left\{N{\left({\mathrm{Si}}^i{\mathrm{Pr}}_3\right)}_2\right\}}_2\right]$. Here, we treat both electronic and I = 1/2 nuclear spins explicitly to investigate the possibility of electric field control of the nuclear degrees of freedom. Furthermore, since the hyperfine coupling breaks Kramers degeneracy and therefore spin-electric coupling arises at zeroth-order, we investigate if this the inclusion of the nuclear spin strongly influences the overall coupling. Transitions are classified as EPR-, NMR-, or mixed/forbidden character, revealing that EPR-like transitions couple more strongly to electric fields than NMR-like ones, as crystal-field modulation dominates over hyperfine modulation. The anisotropy of the electric field effect agrees with previous results, but magnetic-field orientation dependence is suppressed by zeroth-order spin-electric coupling. Dissipative spin-dynamics simulations show that experimentally feasible electric field strengths and relaxation times permit coherent manipulation of both the electronic and nuclear spins, demonstrating an experimentally viable pathway for electric field control in $\left[\mathrm{Tm}{\left\{N{\left({\mathrm{Si}}^i{\mathrm{Pr}}_3\right)}_2\right\}}_2\right]$.

Water molecules play a significant role in maintaining protein structural stability and facilitating molecular interactions. Accurate prediction of water molecule positions around protein structures is essential for understanding their biological roles and has significant implications for protein engineering and drug discovery. Here, we introduce SuperWater, a novel generative AI framework that integrates a score-based diffusion model with equivariant graph neural networks to predict water molecule placements around proteins with high accuracy. SuperWater surpasses existing methods, delivering state-of-the-art performance in both crystal water coverage and prediction precision, achieving water localization within 0.3  ± 0.06 Å of experimentally validated positions. We demonstrate the capabilities of SuperWater through case studies involving protein hydration, protein-ligand binding, and protein-protein binding sites. This framework can be adapted for various applications, including structural biology, binding site prediction, multi-body docking, and water-mediated drug design.

Metabolite alterations are linked to diseases, yet large-scale untargeted metabolomics remains constrained by challenges in signal detection and integration of diverse datasets for developing pre-trained generative models. Here, we introduce mzLearn, a data-driven MS¹ signal-detection and alignment method that runs from mzML files without user-set parameters. Across 15 public datasets, mzLearn detects 11,442 signals on average vs 7,100 (XCMS) and 4,655 (ASARI), with higher TP (89.0% vs 77.4% vs 49.6%) and lower FP (12.5% vs 17.3% vs 18.8%), while correcting instrument drifts across large cohorts without experimental QC samples. mzLearn detected 2,736 robust metabolite signals from 22 public studies (20,548 blood samples), enabling the development of pre-trained variational autoencoder for untargeted metabolomics. Learned metabolite representations reflected demographic data and when fine-tuned on unseen renal cell carcinoma data, improved risk stratification and overall survival predictions, while feature-importance analysis (SHAP) highlighted biologically plausible lipid and carnitine signals. By producing a consistent, high-quality MS¹ feature matrix at scale, mzLearn paves the way for developing pre-trained foundation models for untargeted metabolomics.