MeCP2-Induced Alternations of Transcript Levels and m6A Methylation in Human Retinal Pigment Epithelium Cells.

MeCP2 is a transcriptional regulator that is involved in epithelial-mesenchymal transition (EMT) and is highly expressed in proliferative vitreoretinopathy. m6A methylation is a critical post-transcriptional regulation in eukaryotic cells. However, the connection between MeCP2 and m6A methylation has not been revealed in retinal pigment epithelium (RPE), and the regulatory role of MeCP2 at the post-transcriptional level in an m6A-dependent manner is rarely investigated. In this study, we used sequencing to reveal differences in transcript levels and m6A abundance of individual genes in RPE cells after treatment with human recombinant protein MeCP2. The biological functions and processes of differential genes were further analyzed by bioinformatics. The results exhibited that after MeCP2 treatment, 65 genes were up-regulated and 43 genes were down-regulated at the transcription level, and 4 peaks were hypermethylated and 9,041 peaks were hypomethylated at the m6A modification level. Enrichment analysis found that differentially expressed genes were associated with organic acid metabolism, melanogenesis, and vascular smooth muscle contraction. In addition, differentially methylated genes were related to cell junction, RNA processing and metabolism, cell activity, actin cytoskeleton, and several signaling pathways associated with EMT. Further conjoint analysis indicated that the transcription and m6A levels of the EGR1, ELOVL2, and SFR1 genes were altered, and EGR1 is an essential transcription factor in the EMT process. The RNA levels and m6A levels of the three genes were verified by qPCR and m6A-IP-qPCR, respectively. Overall, this study preliminarily revealed the differential mapping of MeCP2-induced m6A modifications, which contributes to the study of the epigenetic and EMT mechanism in RPE cells.

Fusion of Multiple Spectra for Investigating Chemical Bonding Properties via Machine Learning.

Chemical bonding properties are crucial to understanding the chemical behavior of molecules. Spectroscopy is a versatile technical tool to study various microscopic properties, but its interpretation suffers from human biases and the loss of high-dimensional information. Here, we present a machine learning approach to predict diverse bonding properties, including the bond dissociation energy, bond length, and α-C connectivity of hydroxyls in organic molecules, by fusing multiple spectra with different physical mechanisms. Combining nuclear magnetic resonance and vibrational spectroscopy exhibits higher prediction accuracy than what they did separately. On the hold-out test data set, the models achieve a mean absolute error of 1.243 kcal/mol and 1.041 × 10-4 Å for BDE and bond length and an accuracy of 95.09% for hydroxyl α-C connectivity. Our models demonstrate strong extrapolation capabilities when they are transferred to different molecules, external electric fields, and solvation environments. These end-to-end models pave the way to investigating chemical bonding properties by using spectroscopic observables.

Ultrafast stimulated resonance Raman signatures of lithium polysulfides for shuttling effect characterization: An ab initio study.

The shuttling effect is a crucial obstacle to the practical deployment of lithium sulfur batteries (LSBs). This can be ascribed to the generation of lithium polysulfide (LiPS) redox intermediates that are soluble in the electrolyte. The detailed mechanism of the shuttling, including the chemical structures responsible for the loss of effective mass and the dynamics/kinetics of the redox reactions, are not clear so far. To obtain this microscopic information, characterization techniques with high spatial and temporal resolutions are required. Here, we propose that resonance Raman spectroscopy combined with ultrafast broadband pulses is a powerful tool to reveal the mechanism of the shuttling effect. By combining the chemical bond level spatial resolution of resonance Raman and the femtosecond scale temporal resolution of the ultrafast pulses, this novel technique holds the potential of capturing the spectroscopic fingerprints of the LiPS intermediates during the working stages of LSBs. Using ab initio simulations, we show that, in addition to the excitation energy selective enhancement, resonance Raman signals of different LiPS intermediates are also characteristic and distinguishable. These results will facilitate the real-time in situ monitoring of LiPS species and reveal the underlying mechanism of the shuttling effect.

A Machine-Learning Protocol for Ultraviolet Protein-Backbone Absorption Spectroscopy under Environmental Fluctuations.

Ultraviolet (UV) absorption spectra are commonly used for characterizing the global structure of proteins. However, the theoretical interpretation of UV spectra is hindered by the large number of required expensive ab initio calculations of excited states spanning a huge conformation space. We present a machine-learning (ML) protocol for far-UV (FUV) spectra of proteins, which can predict FUV spectra of proteins with comparable accuracy to density functional theory (DFT) calculations but with 3-4 orders of magnitude reduced computational cost. It further shows excellent predictive power and transferability that can be used to probe structural mutations and protein folding pathways.

Identifying Cu(ii)-amyloid peptide binding intermediates in the early stages of aggregation by resonance Raman spectroscopy: a simulation study.

The aggregation of amyloid beta (Aß) peptides plays a crucial role in the pathology and etiology of Alzheimer's disease. Experimental evidence shows that copper ion is an aggregation-prone species with the ability to coordinately bind to Aß and further induce the formation of neurotoxic Aß oligomers. However, the detailed structures of Cu(ii)-Aß complexes have not been illustrated, and the kinetics and dynamics of the Cu(ii) binding are not well understood. Two Cu(ii)-Aß complexes have been proposed to exist under physiological conditions, and another two might exist at higher pH values. By using ab initio simulations for the spontaneous resonance Raman and time domain stimulated resonance Raman spectroscopy signals, we obtained the characteristic Raman vibronic features of each complex. These signals contain rich structural information with high temporal resolution, enabling the characterization of transient states during the fast Cu-Aß binding and interconversion processes.