Two-Step Machine Learning Approach for Charge-Transfer Coupling with Structurally Diverse Data.

Electronic coupling is important in determining charge-transfer rates and dynamics. Coupling strength is sensitive to both intermolecular, e.g., orientation or distance, and intramolecular degrees of freedom. Hence, it is challenging to build an accurate machine learning model to predict electronic coupling of molecular pairs, especially for those derived from the amorphous phase, for which intermolecular configurations are much more diverse than those derived from crystals. In this work, we devise a new prediction algorithm that employs two consecutive KRR models. The first model predicts molecular orbitals (MOs) from structural variation for each fragment, and coupling is further predicted by using the overlap integral included in a second model. With our two-step procedure, we achieved mean absolute errors of 0.27 meV for an ethylene dimer and 1.99 meV for a naphthalene pair, much improved accuracy amounting to 14-fold and 3-fold error reductions, respectively. In addition, MOs from the first model can also be the starting point to obtain other quantum chemical properties from atomistic structures. This approach is also compatible with a MO predictor with sufficient accuracy.

Distilling coarse-grained representations of molecular electronic structure with continuously gated message passing.

Bottom-up methods for coarse-grained (CG) molecular modeling are critically needed to establish rigorous links between atomistic reference data and reduced molecular representations. For a target molecule, the ideal reduced CG representation is a function of both the conformational ensemble of the system and the target physical observable(s) to be reproduced at the CG resolution. However, there is an absence of algorithms for selecting CG representations of molecules from which complex properties, including molecular electronic structure, can be accurately modeled. We introduce continuously gated message passing (CGMP), a graph neural network (GNN) method for atomically decomposing molecular electronic structure sampled over conformational ensembles. CGMP integrates 3D-invariant GNNs and a novel gated message passing system to continuously reduce the atomic degrees of freedom accessible for electronic predictions, resulting in a one-shot importance ranking of atoms contributing to a target molecular property. Moreover, CGMP provides the first approach by which to quantify the degeneracy of "good" CG representations conditioned on specific prediction targets, facilitating the development of more transferable CG representations. We further show how CGMP can be used to highlight multiatom correlations, illuminating a path to developing CG electronic Hamiltonians in terms of interpretable collective variables for arbitrarily complex molecules.

SERPING1 Reduces Cell Migration via ERK-MMP2-MMP-9 Cascade in Sorafenib- Resistant Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is the most common primary hepatic malignant tumor, and it ranks 2nd in terms of mortality rate among all malignancies in Taiwan. Sorafenib is a multiple tyrosine kinase inhibitor that suppresses tumor cell proliferation and angiogenesis around tumors via different pathways. However, the survival outcome of advanced HCC patients treated with sorafenib is still unsatisfactory. Unfortunately, there are no clinically applicable biomarkers to predict sorafenib therapeutic efficiency in HCC thus far. We found that serpin peptidase inhibitor, clade G, member 1 (SERPING1) is highly associated with overall and recurrence-free survival rates in HCC patients and is also highly correlated with several clinical parameters. SERPING1 expression was increased with sorafenib in both the HCC cell extract and conditioned medium, which was also observed in sorafenib-resistant HepG2 and Huh7 cells. Sorafenib decreased cell viability and migration, which was similar to the effect of SERPING1 in HCC progression. Moreover, sorafenib inhibited both MMP-2 and MMP-9 activity and enhanced the expression of p-ERK in HCC cells. In summary, sorafenib reduces HCC cancer progression might through the p-ERK-MMP-2-MMP-9 cascade via upregulation of SERPING1. In the present study, the roles and molecular mechanisms of SERPING1 and its value as a marker for predicting sorafenib resistance and progression in HCC patients were examined. The results of the present study provide a deep understanding of the roles of SERPING1 in HCC sorafenib resistance, which can be applied to develop early diagnosis and prognosis evaluation methods and establish novel therapeutic targets for specifically treating HCC.

Anterior gradient 2 induces resistance to sorafenib via endoplasmic reticulum stress regulation in hepatocellular carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) accounts for almost 80% of all liver cancer cases and is the sixth most common cancer and the second most common cause of cancer-related death worldwide. The survival rate of sorafenib-treated advanced HCC patients is still unsatisfactory. Unfortunately, no useful biomarkers have been verified to predict sorafenib efficacy in HCC. RESULTS: We assessed a sorafenib resistance-related microarray dataset and found that anterior gradient 2 (AGR2) is highly associated with overall and recurrence-free survival and with several clinical parameters in HCC. However, the mechanisms underlying the role of AGR2 in sorafenib resistance and HCC progression remain unknown. We found that sorafenib induces AGR2 secretion via posttranslational modification and that AGR2 plays a critical role in sorafenib-regulated cell viability and endoplasmic reticulum (ER) stress and induces apoptosis in sorafenib-sensitive cells. In sorafenib-sensitive cells, sorafenib downregulates intracellular AGR2 and conversely induces AGR2 secretion, which suppresses its regulation of ER stress and cell survival. In contrast, AGR2 is highly intracellularly expressed in sorafenib-resistant cells, which supports ER homeostasis and cell survival. We suggest that AGR2 regulates ER stress to influence HCC progression and sorafenib resistance. CONCLUSIONS: This is the first study to report that AGR2 can modulate ER homeostasis via the IRE1α-XBP1 cascade to regulate HCC progression and sorafenib resistance. Elucidation of the predictive value of AGR2 and its molecular and cellular mechanisms in sorafenib resistance could provide additional options for HCC treatment.

MFI2 upregulation promotes malignant progression through EGF/FAK signaling in oral cavity squamous cell carcinoma.

Oral squamous cell carcinoma (OSCC) is the predominant histological type of the head and neck squamous cell carcinoma (HNSCC). By comparing the differentially expressed genes (DEGs) in OSCC-TCGA patients with copy number variations (CNVs) that we identify in OSCC-OncoScan dataset, we herein identified 37 dysregulated candidate genes. Among these potential candidate genes, 26 have been previously reported as dysregulated proteins or genes in HNSCC. Among 11 novel candidates, the overall survival analysis revealed that melanotransferrin (MFI2) is the most significant prognostic molecular in OSCC-TCGA patients. Another independent Taiwanese cohort confirmed that higher MFI2 transcript levels were significantly associated with poor prognosis. Mechanistically, we found that knockdown of MFI2 reduced cell viability, migration and invasion via modulating EGF/FAK signaling in OSCC cells. Collectively, our results support a mechanistic understanding of a novel role for MFI2 in promoting cell invasiveness in OSCC.

Machine-learned dynamic disorder of electron transfer coupling.

Electron transfer (ET) is a fundamental process in chemistry and biochemistry, and electronic coupling is an important determinant of the rate of ET. However, the electronic coupling is sensitive to many nuclear degrees of freedom, particularly those involved in intermolecular movements, making its characterization challenging. As a result, dynamic disorder in electron transfer coupling has rarely been investigated, hindering our understanding of charge transport dynamics in complex chemical and biological systems. In this work, we employed molecular dynamic simulations and machine-learning models to study dynamic disorder in the coupling of hole transfer between neighboring ethylene and naphthalene dimer. Our results reveal that low-frequency modes dominate these dynamics, resulting primarily from intermolecular movements such as rotation and translation. Interestingly, we observed an increasing contribution of translational motion as temperature increased. Moreover, we found that coupling is sub-Ohmic in its spectral density character, with cut-off frequencies in the range of 102 cm-1. Machine-learning models allow direct study of dynamics of electronic coupling in charge transport with sufficient ensemble trajectories, providing further new insights into charge transporting dynamics.

Nuclear accumulation of KPNA2 impacts radioresistance through positive regulation of the PLSCR1-STAT1 loop in lung adenocarcinoma.

Lung adenocarcinoma (ADC) is the predominant histological type of lung cancer, and radiotherapy is one of the current therapeutic strategies for lung cancer treatment. Unfortunately, biological complexity and cancer heterogeneity contribute to radioresistance development. Karyopherin α2 (KPNA2) is a member of the importin α family that mediates the nucleocytoplasmic transport of cargo proteins. KPNA2 overexpression is observed across cancer tissues of diverse origins. However, the role of KPNA2 in lung cancer radioresistance is unclear. Herein, we demonstrated that high expression of KPNA2 is positively correlated with radioresistance and cancer stem cell (CSC) properties in lung ADC cells. Radioresistant cells exhibited nuclear accumulation of KPNA2 and its cargos (OCT4 and c-MYC). Additionally, KPNA2 knockdown regulated CSC-related gene expression in radioresistant cells. Next-generation sequencing and bioinformatic analysis revealed that STAT1 activation and nuclear phospholipid scramblase 1 (PLSCR1) are involved in KPNA2-mediated radioresistance. Endogenous PLSCR1 interacting with KPNA2 and PLSCR1 knockdown suppressed the radioresistance induced by KPNA2 expression. Both STAT1 and PLSCR1 were found to be positively correlated with dysregulated KPNA2 in radioresistant cells and ADC tissues. We further demonstrated a potential positive feedback loop between PLSCR1 and STAT1 in radioresistant cells, and this PLSCR1-STAT1 loop modulates CSC characteristics. In addition, AKT1 knockdown attenuated the nuclear accumulation of KPNA2 in radioresistant lung cancer cells. Our results collectively support a mechanistic understanding of a novel role for KPNA2 in promoting radioresistance in lung ADC cells.

Effects of solvation shell relaxation on chain association mechanisms in poly(3-hexylthiophene) solutions.

Using poly(3-hexylthiophene) (P3HT) as a model conjugated polymer and atomistic molecular dynamics simulations with carefully verified force fields, we performed in-depth investigations of solvation shell properties of P3HT chains (15 repeating units per chain) in two representative groups of non-polar (or aprotic) organic solvents (better solvents: ortho-dichlorobenzene, bromobenzene, and chlorobenzene; poorer solvents: chloroform, para-xylene, and toluene). We demonstrated that solvation shell relaxation properties in P3HT solutions dictate the formation of regular π-π associations and, hence, crystallinity through the initial chain association and subsequent chain sliding. In contrast, the mean features of polymer-solvent interactions, including solvation free energy and radial distribution function, present little or no difference for all solvent media investigated. Better-solvent media were revealed to bear relatively large values of the first solvation shell relaxation time (τ1 ≫ 100 ps) as well as larger ratios of relaxation times for the first two solvation shells (τ1/τ2 > 2), and vice versa for poorer-solvent media (τ1 ≪ 100 ps and τ1/τ2 < 2). The linear hexyl side-chain unit was noted to substantially enlarge both quantities while notably reducing the solvation free energy as well. As discussed herein, these findings shed new light on the mechanistic features by which solvent quality impacts the degree of π-π association crucial for modern applications with crystalline conjugated polymers.

Chemotherapeutic Drug-Regulated Cytokines Might Influence Therapeutic Efficacy in HCC.

Hepatocellular carcinoma (HCC), the most common type of liver cancer, is the second leading cause of cancer-related mortality worldwide. Processes involved in HCC progression and development, including cell transformation, proliferation, metastasis, and angiogenesis, are inflammation-associated carcinogenic processes because most cases of HCC develop from chronic liver damage and inflammation. Inflammation has been demonstrated to be a crucial factor inducing tumor development in various cancers, including HCC. Cytokines play critical roles in inflammation to accelerate tumor invasion and metastasis by mediating the migration of immune cells into damaged tissues in response to proinflammatory stimuli. Currently, surgical resection followed by chemotherapy is the most common curative therapeutic regimen for HCC. However, after chemotherapy, drug resistance is clearly observed, and cytokine secretion is dysregulated. Various chemotherapeutic agents, including cisplatin, etoposide, and 5-fluorouracil, demonstrate even lower efficacy in HCC than in other cancers. Tumor resistance to chemotherapeutic drugs is the key limitation of curative treatment and is responsible for treatment failure and recurrence, thus limiting the ability to treat patients with advanced HCC. Therefore, the capability to counteract drug resistance would be a major clinical advancement. In this review, we provide an overview of links between chemotherapeutic agents and inflammatory cytokine secretion in HCC. These links might provide insight into overcoming inflammatory reactions and cytokine secretion, ultimately counteracting chemotherapeutic resistance.

Increase in Akkermansiaceae in Gut Microbiota of Prostate Cancer-Bearing Mice.

Gut microbiota are reported to be associated with many diseases, including cancers. Several bacterial taxa have been shown to be associated with cancer development or response to treatment. However, longitudinal microbiota alterations during the development of cancers are relatively unexplored. To better understand how microbiota changes, we profiled the gut microbiota composition from prostate cancer-bearing mice and control mice at five different time points. Distinct gut microbiota differences were found between cancer-bearing mice and control mice. Akkermansiaceae was found to be significantly higher in the first three weeks in cancer-bearing mice, which implies its role in the early stage of cancer colonization. We also found that Bifidobacteriaceae and Enterococcaceae were more abundant in the second and last sampling week, respectively. The increments of Akkermansiaceae, Bifidobacteriaceae and Enterococcaceae were previously found to be associated with responses to immunotherapy, which suggests links between these bacteria families and cancers. Additionally, our function analysis showed that the bacterial taxa carrying steroid biosynthesis and butirosin and neomycin biosynthesis were increased, whereas those carrying naphthalene degradation decreased in cancer-bearing mice. Our work identified the bacteria taxa altered during prostate cancer progression and provided a resource of longitudinal microbiota profiles during cancer development in a mouse model.

Artificial neural networks for predicting charge transfer coupling.

Quantum chemistry calculations have been very useful in providing many key detailed properties and enhancing our understanding of molecular systems. However, such calculation, especially with ab initio models, can be time-consuming. For example, in the prediction of charge-transfer properties, it is often necessary to work with an ensemble of different thermally populated structures. A possible alternative to such calculations is to use a machine-learning based approach. In this work, we show that the general prediction of electronic coupling, a property that is very sensitive to intermolecular degrees of freedom, can be obtained with artificial neural networks, with improved performance as compared to the popular kernel ridge regression method. We propose strategies for optimizing the learning rate and batch size, improving model performance, and further evaluating models to ensure that the physical signatures of charge-transfer coupling are well reproduced. We also address the effect of feature representation as well as statistical insights obtained from the loss function and the data structure. Our results pave the way for designing a general strategy for training such neural-network models for accurate prediction.

Characterization of Copy Number Variations in Oral Cavity Squamous Cell Carcinoma Reveals a Novel Role for MLLT3 in Cell Invasiveness.

BACKGROUND: DNA copy number variations (CNVs) are a hallmark of cancer, and the current study aimed to demonstrate the profile of the CNVs for oral cavity squamous cell carcinoma (OSCC) and elucidate the clinicopathological associations and molecular mechanisms of a potential marker derived from CNVs, mixed-lineage leukemia translocated to chromosome 3 protein (MLLT3), in OSCC carcinogenesis. MATERIALS AND METHODS: CNVs in 37 OSCC tissue specimens were analyzed using a high-resolution microarray, the OncoScan array. Gene expression was analyzed by real-time polymerase chain reaction in 127 OSCC and normal tissue samples. Cell function assays included cell cycle, migration, invasion and chromatin immunoprecipitation assays. RESULTS: We found a novel copy number amplified region, chromosome 9p, encompassing MLLT3 via the comparison of our data set with six other OSCC genome-wide CNV data sets. MLLT3 overexpression was associated with poorer overall survival in patients with OSCC (p = .048). MLLT3 knockdown reduced cell migration and invasion. The reduced invasion ability in MLLT3-knockdown cells was rescued with double knockdown of MLLT3 and CBP/p300-interacting transactivator with ED rich carboxy-terminal domain 4 (CITED4; 21.0% vs. 61.5%). Knockdown of MLLT3 impaired disruptor of telomeric silencing-1-like (Dot1L)-associated hypermethylation in the promoter of the tumor suppressor, CITED4 (p < .001), and hence dysregulated HIF-1α-mediated genes (TWIST, MMP1, MMP2, VIM, and CDH1) in OSCC cells. CONCLUSION: We identified unique CNVs in tumors of Taiwanese patients with OSCC. Notably, MLLT3 overexpression is related to the poorer prognosis of patients with OSCC and is required for Dot1L-mediated transcriptional repression of CITED4, leading to dysregulation of HIF-1α-mediated genes. IMPLICATIONS FOR PRACTICE: This article reports unique copy number variations in oral cavity squamous cell carcinoma (OSCC) tumors of Taiwanese patients. Notably, MLLT3 overexpression is related to the poorer prognosis of patients with OSCC and is required for Dot1L-mediated transcriptional repression of CITED4, leading to dysregulation of HIF-1α-mediated genes.

Machine Learning for Predicting Electron Transfer Coupling.

Electron transfer coupling is a critical factor in determining electron transfer rates. This coupling strength can be sensitive to details in molecular geometries, especially intermolecular configurations. Thus, studying charge transporting behavior with a full first-principle approach demands a large amount of computation resources in quantum chemistry (QC) calculation. To address this issue, we developed a machine learning (ML) approach to evaluate electronic coupling. A prototypical ML model for an ethylene system was built by kernel ridge regression with Coulomb matrix representation. Since the performance of the ML models highly dependent on their building strategies, we systematically investigated the generality of the ML models, the choice of features and target labels. The best ML model trained with 40 000 samples achieved a mean absolute error of 3.5 meV and greater than 98% accuracy in predicting phases. The distance and orientation dependence of electronic coupling was successfully captured. Bypassing QC calculation, the ML model saved 10-104 times the computation cost. With the help of ML, reliable charge transport models and mechanisms can be further developed.

The correspondence between the conformational and chromophoric properties of amorphous conjugated polymers in mesoscale condensed systems.

For π-conjugated polymers, the notion of spectroscopic units or "chromophores" provides illuminating insights into the experimentally observed absorption/emission spectra and the mechanisms of energy/charge transfer. To date, however, no statistical analysis has revealed a direct correspondence between chromophoric and conformational properties-with the latter being fundamental to polymer semiconductors. Herein, we propose a "persistence length" calculation to re-evaluate chain conformation over a full conjugation length. The mesoscale condensed systems of MEH-PPV and MEH-PPV/C60 hybrid (system size ∼10 × 10 × 10 nm3) are utilized as two prototypical model systems, along with a full range of segmental lengths (2-20-mer) and five lowest singlet excited states to hint at the generality of the features presented. We demonstrate, for the first time, that two properly re-defined conformational factors that characterize chain folding and planarity, respectively, capture excellently the population distribution of chromophores in both systems investigated. In contrast, the conventional strategy of utilizing two adjacent monomer units to characterize (local) chain conformation results in only an inconspicuous correlation between the two, as previously reported. It is further shown that chain folding-and not chain planarity-is more relevant in capturing the associated oscillator strength for the first excited state, where the transient dipole moments are known to align with the chain conformation, although the corresponding excitation energy and exciton size seem relatively unaffected. The observed effects of C60 on the MEH-PPV adsorption spectra also agree with recent experimental trends. Overall, the present findings are expected to aid future multiscale computer simulations and spectroscopy-data interpretations for polymer semiconductors and their hybrid systems.

Quantitative proteomics reveals a novel role of karyopherin alpha 2 in cell migration through the regulation of vimentin-pErk protein complex levels in lung cancer.

Karyopherin alpha 2 (KPNA2) is overexpressed in various human cancers and is associated with cancer invasiveness and poor prognosis. Herein, to understand the essential role of KPNA2 protein complexes in cancer progression, we applied stable isotope labeling with amino acids in cell culture (SILAC)-based quantitative proteomic strategy combined with immunoprecipitation (IP) to investigate the differential KPNA2 protein complexes in lung adenocarcinoma cell lines with different invasiveness potentials. We found that 64 KPNA2-interaction proteins displayed a 2-fold difference in abundance between CL1-5 (high invasiveness) and CL1-0 (low invasiveness) cells. Pathway map analysis revealed that the formation of complexes containing KPNA2 and cytoskeleton-remodeling-related proteins, including actin, beta tubulin, tubulin heterodimers, vimentin, keratin 8, keratin 18, and plectin, was associated with cancer invasiveness. IP demonstrated that the levels of KPNA2-vimentin-pErk complexes were significantly higher in CL1-5 cells than in CL1-0 cells. The KPNA2-vimentin-pErk complex was also up-regulated in the advanced stage compared with the early-stage lung adenocarcinoma tissues. Importantly, the levels of pErk as well as cell migration ability were significantly reduced in KPNA2-knockdown cells; however, migration was restored by treatment with pErk phosphatase inhibitors. Collectively, our results demonstrate the usefulness of a SILAC-based proteomic strategy for identifying invasiveness-associated KPNA2 protein complexes and provide new insight into the KPNA2-mediated modulation of cell migration.

Quantitative proteomics reveals regulation of karyopherin subunit alpha-2 (KPNA2) and its potential novel cargo proteins in nonsmall cell lung cancer.

The process of nucleocytoplasmic shuttling is mediated by karyopherins. Dysregulated expression of karyopherins may trigger oncogenesis through aberrant distribution of cargo proteins. Karyopherin subunit alpha-2 (KPNA2) was previously identified as a potential biomarker for nonsmall cell lung cancer by integration of the cancer cell secretome and tissue transcriptome data sets. Knockdown of KPNA2 suppressed the proliferation and migration abilities of lung cancer cells. However, the precise molecular mechanisms underlying KPNA2 activity in cancer remain to be established. In the current study, we applied gene knockdown, subcellular fractionation, and stable isotope labeling by amino acids in cell culture-based quantitative proteomic strategies to systematically analyze the KPNA2-regulating protein profiles in an adenocarcinoma cell line. Interaction network analysis revealed that several KPNA2-regulating proteins are involved in the cell cycle, DNA metabolic process, cellular component movements and cell migration. Importantly, E2F1 was identified as a potential novel cargo of KPNA2 in the nuclear proteome. The mRNA levels of potential effectors of E2F1 measured using quantitative PCR indicated that E2F1 is one of the "master molecule" responses to KPNA2 knockdown. Immunofluorescence staining and immunoprecipitation assays disclosed co-localization and association between E2F1 and KPNA2. An in vitro protein binding assay further demonstrated that E2F1 interacts directly with KPNA2. Moreover, knockdown of KPNA2 led to subcellular redistribution of E2F1 in lung cancer cells. Our results collectively demonstrate the utility of quantitative proteomic approaches and provide a fundamental platform to further explore the biological roles of KPNA2 in nonsmall cell lung cancer.

Accessing the electronic structure of liquid crystalline semiconductors with bottom-up electronic coarse-graining.

Understanding the relationship between multiscale morphology and electronic structure is a grand challenge for semiconducting soft materials. Computational studies aimed at characterizing these relationships require the complex integration of quantum-chemical (QC) calculations, all-atom and coarse-grained (CG) molecular dynamics simulations, and back-mapping approaches. However, these methods pose substantial computational challenges that limit their application to the requisite length scales of soft material morphologies. Here, we demonstrate the bottom-up electronic coarse-graining (ECG) of morphology-dependent electronic structure in the liquid-crystal-forming semiconductor, 2-(4-methoxyphenyl)-7-octyl-benzothienobenzothiophene (BTBT). ECG is applied to construct density functional theory (DFT)-accurate valence band Hamiltonians of the isotropic and smectic liquid crystal (LC) phases using only the CG representation of BTBT. By bypassing the atomistic resolution and its prohibitive computational costs, ECG enables the first calculations of the morphology dependence of the electronic structure of charge carriers across LC phases at the ∼20 nm length scale, with robust statistical sampling. Kinetic Monte Carlo (kMC) simulations reveal a strong morphology dependence on zero-field charge mobility among different LC phases as well as the presence of two-molecule charge carriers that act as traps and hinder charge transport. We leverage these results to further evaluate the feasibility of developing mesoscopic, field-based ECG models in future works. The fully CG approach to electronic property predictions in LC semiconductors opens a new computational direction for designing electronic processes in soft materials at their characteristic length scales.

Non-negligible Outer-Shell Reorganization Energy for Charge Transfer in Nonpolar Systems.

Many charge-transporting molecular systems function as ordered or disordered arrays of solid state materials composed of nonpolar (or weakly polar) molecules. Due to low dielectric constants for nonpolar systems, it is common to ignore the effects of outer-shell reorganization energy (λout). However, ignoring λout has not been properly supported and it can severely impact predictions and insights derived. Here, we estimate λout by two means: from experimental ultraviolet photoelectron spectra, in which vibronic progression in these spectra can be fitted with the widths of peaks determining the low-frequency component in reorganization energy, regarded to be closely associated with λout, and from molecular dynamic (MD) simulation of nonpolar molecules, in which disorder or fluctuation statistics for energies of charged molecules are calculated. An upper bound for λout was obtained as 505 and 549 meV for crystalline anthracene (140 K) and pentacene (50 K), respectively, by fitting of experimental data, and 212 and 170 meV, respectively, from MD simulations. These values are comparable to the inner-sphere reorganization energy (λin) arising from intramolecular vibration. With corresponding spectral density functions calculated, we found that λout is influenced both by low- and high-frequency dynamics, in which the former arises from constrained translational and rotational motions of surrounding molecules. In an amorphous state, about half of the λout's obtained are from high-frequency components, which is quite different from the conventional polar solvation. Moreover, crystalline systems exhibit super-Ohmic spectral density, whereas amorphous systems are sub-Ohmic.

Oxidative stress mediates nucleocytoplasmic shuttling of KPNA2 via AKT1-CDK1 axis-regulated S62 phosphorylation.

Karyopherin α 2 (KPNA2, importin α1), a transport factor shuttling between the nuclear and cytoplasmic compartments, is involved in the nuclear import of proteins and participates in cellular processes such as cell cycle regulation, apoptosis, and transcriptional regulation. However, it is still unclear which signaling regulates the nucleocytoplasmic distribution of KPNA2 in response to cellular stress. In this study, we report that oxidative stress increases nuclear retention of KPNA2 through alpha serine/threonine-protein kinase (AKT1)-mediated reduction of serine 62 (S62) phosphorylation. We first found that AKT1 activation was required for H2O2-induced nuclear accumulation of KPNA2. Immunoprecipitation and quantitative proteomic analysis revealed that the phosphorylation of KPNA2 at S62 was decreased under H2O2-induced oxidative stress. We showed that cyclin-dependent kinase 1 (CDK1), a kinase responsible for KPNA2 S62 phosphorylation, contributes to the localization of KPNA2 in the cytoplasm. AKT1 knockdown increased KPNA2 S62 phosphorylation and inhibited CDK1 activation. Furthermore, H2O2-induced AKT1 activation promoted nuclear KPNA2 interaction with nucleophosmin 1 (NPM1), resulting in attenuation of NPM1-mediated cyclin D1 gene transcription. Thus, we infer that the AKT1-CDK1 axis regulates the nucleocytoplasmic shuttling and function of KPNA2 through spatiotemporal regulation of KPNA2 S62 phosphorylation under oxidative stress conditions.

Identifying Coarse-Grained Representations for Electronic Predictions.

Coarse-grained (CG) simulations are an important computational tool in chemistry and materials science. Recently, systematic "bottom-up" CG models have been introduced to capture electronic structure variations of molecules and polymers at the CG resolution. However, the performance of these models is limited by the ability to select reduced representations that preserve electronic structure information, which remains a challenge. We propose two methods for (i) identifying important electronically coupled atomic degrees of freedom and (ii) scoring the efficacy of CG representations used in conjunction with CG electronic predictions. The first method is a physically motivated approach that incorporates nuclear vibrations and electronic structure derived from simple quantum chemical calculations. We complement this physically motivated approach with a machine learning technique based on the marginal contribution of nuclear degrees of freedom to electronic prediction accuracy using an equivariant graph neural network. By integrating these two approaches, we can both identify critical electronically coupled atomic coordinates and score the efficacy of arbitrary CG representations for making electronic predictions. We leverage this capability to make a connection between optimized CG representations and the future potential for "bottom-up" development of simplified model Hamiltonians incorporating nonlinear vibrational modes.