Image Reconstruction Using Supervised Learning in Wearable Electrical Impedance Tomography of the Thorax.

Electrical impedance tomography (EIT) is a non-invasive technique for visualizing the internal structure of a human body. Capacitively coupled electrical impedance tomography (CCEIT) is a new contactless EIT technique that can potentially be used as a wearable device. Recent studies have shown that a machine learning-based approach is very promising for EIT image reconstruction. Most of the studies concern models containing up to 22 electrodes and focus on using different artificial neural network models, from simple shallow networks to complex convolutional networks. However, the use of convolutional networks in image reconstruction with a higher number of electrodes requires further investigation. In this work, two different architectures of artificial networks were used for CCEIT image reconstruction: a fully connected deep neural network and a conditional generative adversarial network (cGAN). The training dataset was generated by the numerical simulation of a thorax phantom with healthy and illness-affected lungs. Three kinds of illnesses, pneumothorax, pleural effusion, and hydropneumothorax, were modeled using the electrical properties of the tissues. The thorax phantom included the heart, aorta, spine, and lungs. The sensor with 32 area electrodes was used in the numerical model. The ECTsim custom-designed toolbox for Matlab was used to solve the forward problem and measurement simulation. Two artificial neural networks were trained with supervision for image reconstruction. Reconstruction quality was compared between those networks and one-step algebraic reconstruction methods such as linear back projection and pseudoinverse with Tikhonov regularization. This evaluation was based on pixel-to-pixel metrics such as root-mean-square error, structural similarity index, 2D correlation coefficient, and peak signal-to-noise ratio. Additionally, the diagnostic value measured by the ROC AUC metric was used to assess the image quality. The results showed that obtaining information about regional lung function (regions affected by pneumothorax or pleural effusion) is possible using image reconstruction based on supervised learning and deep neural networks in EIT. The results obtained using cGAN are strongly better than those obtained using a fully connected network, especially in the case of noisy measurement data. However, diagnostic value estimation showed that even algebraic methods allow us to obtain satisfactory results.

Design, synthesis and biological evaluation of hybrid of ubenimex-fluorouracil for hepatocellular carcinoma therapy.

In our previous study, we discovered a ubenimex-fluorouracil (5FU) conjugates BC-02, which displays significant in vivo anti-tumor activity, however, the instability of BC-02 in plasma limits its further development as a drug candidate. Herein, we designed and synthesized four novel ubenimex-5FU conjugates by optimizing the linkers between ubenimex and 5FU based on BC-02. Representative compound 20 is more stable than BC-02 in human plasma and displays about 100 times higher CD13 inhibitory activity than the positive control ubenimex. Meanwhile, the antiproliferative activity of 20 was comparable with 5FU in vitro. The preliminary mechanism study indicated that compound 20 exhibited significant anti-invasion and anti-angiogenesis activities in vitro. Furthermore, compound 20 obviously inhibits tumor growth and metastasis in vivo and prolong the survival time of tumor-bearing mice. Our study may have an important implication reference for the design of more druglike mutual prodrug, and compound 20 can be used as a lead compound for further design and development.

Discovery of benzamide-based PI3K/HDAC dual inhibitors with marked pro-apoptosis activity in lymphoma cells.

Inhibition of PI3K and histone deacetylase (HDAC) activity simultaneously using a single molecule appears to be a promising approach for cancer treatment. Current PI3K/HDAC dual inhibitors commonly use hydroxamate moiety as zinc binding group, which lack HDAC isoform selectivity and have potential genotoxicity. In this study, a novel series of benzamide-based PI3K/HDAC dual inhibitors were rationally designed and synthesized. Representative compound PH14 showed potent inhibitory activity toward PI3Kα and HDAC3, with IC50 values of 20.3 nM and 24.5 nM, respectively. This was further supported by the blockage of AKT phosphorylation and an increase in acetylated histone H3 levels in Western blot study. The advantage of simultaneously targeting PI3Kα and HDAC is not only reflected in the significant antiproliferative activity, but also in its ability to promote the apoptosis in Jeko-1 cells. Moreover, PH14 had weak inhibitory effects on CYP450 enzymes and hERG. In the pharmacokinetic study, the administration of 1 mg/kg of PH14 the administration of 1 mg/kg of PH14 resulted in a t1/2 of 10 h and an AUC (0-∞) of 2772 h ng/mL. Our study may provide ideas for the further development of novel HDAC/PI3K dual inhibitors.

Synthesis and bioactivity evaluation of ferrocene-based hydroxamic acids as selective histone deacetylase 6 inhibitors.

Histone deacetylase 6 (HDAC6) is involved in multiple regulatory processes and emerges as a promising target for treating cancer and neurodegenerative diseases. Benefited from the unique sandwich conformation of ferrocene, a series of ferrocene-based hydroxamic acids have been developed as novel HDAC6 inhibitors in this paper, especially the two ansa-ferrocenyl complexes with IC50s at the nanomolar level. [3]-Ferrocenophane hydroxamic acid analog II-5 displays the most potent inhibitory activity on HDAC6 and establishes remarkable selectivity towards other HDAC isoforms. Compound II-5 dose-dependently induces accumulation of acetylated α-tubulin while having a negligible effect on the level of acetylated Histone H3, confirming its isoform selectivity. Further biological evaluation of II-5 on cancer cells corroborates its antiproliferative effect, which mainly contributed to the induction of cellular apoptosis. It is worth noting that compound II-5 demonstrates an optimal profile on human plasma stability. These results strengthen ferrocene's unique role in developing selective protein inhibitors and indicate that compound II-5 may be a suitable lead for further evaluation and development for treating HDAC6-associated disorders and diseases.

Al-induced CsUGT84J2 enhances flavonol and auxin accumulation to promote root growth in tea plants.

Although Al is not necessary or even toxic to most plants, it is beneficial for the growth of tea plants. However, the mechanism through which Al promotes root growth in tea plants remains unclear. In the present study, we found that flavonol glycoside levels in tea roots increased following Al treatment, and the Al-induced UDP glycosyltransferase CsUGT84J2 was involved in this mechanism. Enzyme activity assays revealed that rCsUGT84J2 exhibited catalytic activity on multiple types of substrates, including phenolic acids, flavonols, and auxins in vitro. Furthermore, metabolic analysis with UPLC-QqQ-MS/MS revealed significantly increased flavonol and auxin glycoside accumulation in CsUGT84J2-overexpressing Arabidopsis thaliana. In addition, the expression of genes involved in the flavonol pathway as well as in the auxin metabolism, transport, and signaling pathways was remarkably enhanced. Additionally, lateral root growth and exogenous Al stress tolerance were significantly improved in transgenic A. thaliana. Moreover, gene expression and metabolic accumulation related to phenolic acids, flavonols, and auxin were upregulated in CsUGT84J2-overexpressing tea plants but downregulated in CsUGT84J2-silenced tea plants. In conclusion, Al treatment induced CsUGT84J2 expression, mediated flavonol and auxin glycosylation, and regulated endogenous auxin homeostasis in tea roots, thereby promoting the growth of tea plants. Our findings lay the foundation for studying the precise mechanisms through which Al promotes the growth of tea plants.

Grain-boundary-rich layered double hydroxides via a boron-assisted strategy for the oxygen evolution reaction.

This study reports a facile boron-assisted strategy to prepare NiFe LDHs with rich grain boundaries. The formation of these grain boundaries originates from the imperfect oriented attachment between primary LDH particles transformed from amorphous borides/borates. The obtained grain-boundary-rich NiFe LDHs exhibit excellent oxygen evolution reaction activity.

First-in-Class Hydrazide-Based HDAC6 Selective Inhibitor with Potent Oral Anti-Inflammatory Activity by Attenuating NLRP3 Inflammasome Activation.

In this study, we report the first highly selective HDAC6 inhibitor with hydrazide as the zinc-binding group (ZBG), which displays superior pharmacokinetic properties to the current hydroxamic acid inhibitors. Structure-activity relationship study reveals that ethyl group substituent hydrazide-based ZBG and cap group with more substantial rigidity and larger volume increase the HDAC6 selectivity of designed compounds. Representative inhibitor 35m exhibits potent HDAC6 inhibitory activity with an IC50 value of 0.019 µM. To our surprise, 35m establishes significant improvement in the pharmacokinetic property with much higher AUC0-inf (10292 ng·h/mL) and oral bioavailability (93.4%) than hydroximic acid-based HDAC6 inhibitors Tubastatin A and ACY-1215. Low-dose 35m remarkably decreases LPS-induced IL-1ß release both in vitro and in vivo by blocking the activation of NLRP3, indicating that 35m can be a potential orally active therapeutic agent for the treatment of NLRP3-related diseases.

Potent Hydrazide-Based HDAC Inhibitors with a Superior Pharmacokinetic Profile for Efficient Treatment of Acute Myeloid Leukemia In Vivo.

As "Michael acceptors" may induce promiscuous responses in mammalian cells by reacting with various proteins, we modified the cinnamamide of our previous hydrazide-based HDAC inhibitors (HDACIs) to deactivate the Michael reaction. Representative compound 11h is 2-5 times more potent than lead compound 17 in both HDAC inhibitory activity (IC50 = 0.43-3.01 nM) and cell-based antitumor assay (IC50 = 19.23-61.04 nM). The breakthrough in the pharmacokinetic profile of 11h (oral bioavailability: 112%) makes it a lead-in-class oral active agent, validated in the in vivo anti-AML study (4 mg/kg p.o., TGI = 78.9%). Accumulated AcHH3 and AcHH4 levels in tumor tissue directly correlate with the in vivo efficacy, as panobinostat with lower AcHH3 and AcHH4 levels than 11h displays limited activity. To the best of our knowledge, this work contributes the first report of in vivo antitumor activity of hydrazide-based HDACIs. The outstanding pharmacokinetic/pharmacodynamic and antitumor activity of 11h could potentially extend the clinical application of current HDACIs.

A GLUT1 inhibitor-based fluorogenic probe for Warburg effect-targeted drug screening and diagnostic imaging of hyperglycolytic cancers.

Increased expression of glucose transporters, especially GLUT1 has been proven to be involved in the Warburg effect. Therefore, GLUT1-targeted oncological approaches are being successfully employed for clinical tumor diagnostic imaging (e.g. the 18F-FDG/PET), drug delivery and novel anticancer drug development. Despite the long history of the Warburg effect-targeted cancer diagnosis, other than antibody labeling, there have been no imaging tools developed for direct detection of the GLUT1 expression. Herein, we report the new strategy of using a non-antibody GLUT1 binding probe for Warburg effect-based tumor detection and diagnostic imaging. By specifically inhibits the transport function of GLUT1, the newly designed fluorescent probe, CUM-5, was found to be a useful tool not only for sensitive GLUT1-mediated cancer cell detection, but also for cell-based high-throughput GLUT inhibitor screening. In in vivo studies, CUM-5 shows clear advantages including desirable tumor-to-normal tissue contrast and excellent tumor selectivity (Tm/Bkg and Tm/Torg), as well as high fluorescence stability (long response time) and ideal physiological biocompatibility. In particular, the GLUT1 inhibitor probe offers the potential use for glycolysis-based diagnostic imaging in triple-negative breast cancer which is claimed to have unsatisfactory results with FDG/PET diagnosis, thus remaining a highly metastatic and lethal disease with a need for sensitive and precise identification.