Immune checkpoint therapy-elicited sialylation of IgG antibodies impairs antitumorigenic type I interferon responses in hepatocellular carcinoma.

The reinvigoration of anti-tumor T cells in response to immune checkpoint blockade (ICB) therapy is well established. Whether and how ICB therapy manipulates antibody-mediated immune response in cancer environments, however, remains elusive. Using tandem mass spectrometric analysis of modification of immunoglobulin G (IgG) from hepatoma tissues, we identified a role of ICB therapy in catalyzing IgG sialylation in the Fc region. Effector T cells triggered sialylation of IgG via an interferon (IFN)-γ-ST6Gal-I-dependent pathway. DC-SIGN+ macrophages represented the main target cells of sialylated IgG. Upon interacting with sialylated IgG, DC-SIGN stimulated Raf-1-elicited elevation of ATF3, which inactivated cGAS-STING pathway and eliminated subsequent type-I-IFN-triggered antitumorigenic immunity. Although enhanced IgG sialylation in tumors predicted improved therapeutic outcomes for patients receiving ICB therapy, impeding IgG sialylation augmented antitumorigenic T cell immunity after ICB therapy. Thus, targeting antibody-based negative feedback action of ICB therapy has potential for improving efficacy of cancer immunotherapies.

miRNome targeting NF-κB signaling orchestrates macrophage-triggered cancer metastasis and recurrence.

Whether and how tumor intrinsic signature determines macrophage-elicited metastasis remain elusive. Here, we show, in detailed studies of data regarding 7,477 patients of 20 types of human cancers, that only 13.8% ± 2.6%/27.9% ± 3.03% of patients with high macrophage infiltration index exhibit early recurrence/vascular invasion. In parallel, although macrophages enhance the motility of various hepatoma cells, their enhancement intensity is significantly heterogeneous. We identify that the expression of malignant Dicer, a ribonuclease that cleaves miRNA precursors into mature miRNAs, determines macrophage-elicited metastasis. Mechanistically, the downregulation of Dicer in cancer cells leads to defects in miRNome targeting NF-κB signaling, which in turn enhances the ability of cancer cells to respond to macrophage-related inflammatory signals and ultimately promotes metastasis. Importantly, transporting miR-26b-5p, the most potential miRNA targeting NF-κB signaling in hepatocellular carcinoma, can effectively reverse macrophage-elicited metastasis of hepatoma in vivo. Our results provide insights into the crosstalk between Dicer-elicited miRNome and cancer immune microenvironments and suggest that strategies to remodel malignant cell miRNome may overcome pro-tumorigenic activities of inflammatory cells.

Unraveling the Structure-Property-Performance Relationships of Fused-Ring Nonfullerene Acceptors: Toward a C-Shaped ortho-Benzodipyrrole-Based Acceptor for Highly Efficient Organic Photovoltaics.

The high-performance Y6-based nonfullerene acceptors (NFAs) feature a C-shaped A-DA'D-A-type molecular architecture with a central electron-deficient thiadiazole (Tz) A' unit. In this work, we designed and synthesized a new A-D-A-type NFA, termed CB16, having a C-shaped ortho-benzodipyrrole-based skeleton of Y6 but with the Tz unit eliminated. When processed with nonhalogenated xylene without using any additives, the binary PM6:CB16 devices display a remarkable power conversion efficiency (PCE) of 18.32% with a high open-circuit voltage (Voc) of 0.92 V, surpassing the performance of the corresponding Y6-based devices. In contrast, similarly synthesized SB16, featuring an S-shaped para-benzodipyrrole-based skeleton, yields a low PCE of 0.15% due to the strong side-chain aggregation of SB16. The C-shaped A-DNBND-A skeleton in CB16 and the Y6-series NFAs constitutes the essential structural foundation for achieving exceptional device performance. The central Tz moiety or other A' units can be employed to finely adjust intermolecular interactions. The single-crystal X-ray structure reveals that ortho-benzodipyrrole-embedded A-DNBND-A plays an important role in the formation of a 3D elliptical network packing for efficient charge transport. Solution structures of the PM6:NFAs detected by small- and wide-angle X-ray scattering (SWAXS) indicate that removing the Tz unit in the C-shaped skeleton could reduce the self-packing of CB16, thereby enhancing the complexing and networking with PM6 in the spin-coating solution and the subsequent device film. Elucidating the structure-property-performance relationships of A-DA'D-A-type NFAs in this work paves the way for the future development of structurally simplified A-D-A-type NFAs.

Synthesis of Indole-Fused Pyrazino[1,2-a]quinazolinones by Copper(I)-Catalyzed Selective Hydroamination-Cyclization of Alkynyl-tethered Quinazolinones.

We described a copper(I)-catalyzed atom economic and selective hydroamination-cyclization of alkynyl-tethered quinazolinones to prepare a variety of indole-fused pyrazino[1,2-a]quinazolinones in good to excellent yields ranging from 39% to 99% under mild reaction conditions. Control experiments revealed that coordination-directed method of quinazolinone moiety with copper(I) was important for the selective hydroamination-cyclization of alkynes at the N1-atom instead of N3-atom of quinazolinone. The reaction could be easily performed at gram scales and some prepared indole-fused pyrazino[1,2-a]quinazolinones with donating groups on the indole moiety showed a distinct fluorescence emission wavelength with blue shift under the acid conditions.

Object detection: A novel AI technology for the diagnosis of hepatocyte ballooning.

Metabolic dysfunction-associated fatty liver disease (MAFLD) has reached epidemic proportions worldwide and is the most frequent cause of chronic liver disease in developed countries. Within the spectrum of liver disease in MAFLD, steatohepatitis is a progressive form of liver disease and hepatocyte ballooning (HB) is a cardinal pathological feature of steatohepatitis. The accurate and reproducible diagnosis of HB is therefore critical for the early detection and treatment of steatohepatitis. Currently, a diagnosis of HB relies on pathological examination by expert pathologists, which may be a time-consuming and subjective process. Hence, there has been interest in developing automated methods for diagnosing HB. This narrative review briefly discusses the development of artificial intelligence (AI) technology for diagnosing fatty liver disease pathology over the last 30 years and provides an overview of the current research status of AI algorithms for the identification of HB, including published articles on traditional machine learning algorithms and deep learning algorithms. This narrative review also provides a summary of object detection algorithms, including the principles, historical developments, and applications in the medical image analysis. The potential benefits of object detection algorithms for HB diagnosis (specifically those combined with a transformer architecture) are discussed, along with the future directions of object detection algorithms in HB diagnosis and the potential applications of generative AI on transformer architecture in this field. In conclusion, object detection algorithms have huge potential for the identification of HB and could make the diagnosis of MAFLD more accurate and efficient in the near future.

Synthesis of 4-(trichloromethyl)pyrido[2',1':3,4]pyrazino[2,1-b]quinazolinones through a cyclized dearomatization and trichloromethylation cascade strategy.

A variety of 4-(trichloromethyl)pyrido[2',1':3,4]pyrazino[2,1-b]quinazolinones were prepared in moderate to good yields with high regioselectivity through intramolecular 6-endo-dig cyclization and trichloromethylation of N3-alkynyl-2-pyridinyl-tethered quinazolinones in chloroform. Mechanistic studies revealed that chloroform might serve as a trichloromethyl anion precursor. Furthermore, the reaction could be easily performed on gram scales and an estrone-derived 4-(trichloromethyl)pyrido[2',1':3,4]pyrazino[2,1-b]quinazolinone was prepared over five steps. The present method features broad substrate scope, good functional group tolerance, new dearomatization of pyridine rings, and chloroform as the trichloromethylation reagent.

Design and synthesis of luotonin A-derived topoisomerase targeting scaffold with potent antitumor effect and low genotoxicity.

Conventional topoisomerase (Topo) inhibitors typically usually exert their cytotoxicity by damaging the DNAs, which exhibit high toxicity and tend to result in secondary carcinogenesis risk. Molecules that have potent topoisomerase inhibitory activity but involve less DNA damage provide more desirable scaffolds for developing novel chemotherapeutic agents. In this work, we broke the rigid pentacyclic system of luotonin A and synthesized thirty-three compounds as potential Topo inhibitors based on the devised molecular motif. Further investigation disclose that two compounds with the highest antiproliferation activity against cancer cells, 5aA and 5dD, had a distinct Topo I inhibitory mechanism different from those of the classic Topo I inhibitors CPT or luteolin, and were able to obviate the obvious cellular DNA damage typically associated with clinically available Topo inhibitors. The animal model experiments demonstrated that even in mice treated with a high dosage of 50 mg/kg 5aA, there were no obvious signs of toxicity or loss of body weight. The tumor growth inhibition (TGI) rate was 54.3 % when 20 mg/kg 5aA was given to the T24 xenograft mouse model, and 5aA targeted the cancer tissue precisely without causing damage to the liver and other major organs.

Osthole Activates the Cholinergic Anti-Inflammatory Pathway via α7nAChR Upregulation to Alleviate Inflammatory Responses.

Osthole (also known as Osthol) is the main anti-inflammatory coumarin found in Cnidium monnieri and severs as the exclusive quality-controlled component according the Chinese Pharmacopoeia. However, its underlying anti-inflammatory mechanism remains unknown. In this study, we demonstrated that Osthole treatment significantly inhibited the generation of TNF-α, but not IL-6 in the classical LPS-stimulated RAW264.7 macrophage model. In addition, LPS induced the activation of both MAPK and NF-κB signalling pathways, of which the former was dose-dependently restrained by Osthole via suppressing the phosphorylation of JNK and P38 proteins, while the phosphorylation of IκB and P65 proteins remained unaffected. Interestingly, Osthole dose-dependently up-regulated the expression of the key cholinergic anti-inflammatory pathway regulator α7nAChR, and the TNF-α inhibition effect of Osthole was also significantly alleviated by the treatment of α7nAChR antagonist methylbetaine. These results demonstrate that Osthole may regulate TNF-α by promoting the expression of α7nAChR, thereby activate the vagus nerve-dependent cholinergic anti-inflammatory pathway.

Depletion of tilmicosin residue in Gushi chickens following oral administration via drinking water.

This study aimed to examine the depletion of tilmicosin residues in Gushi chickens following the administration at a concentration of 75 mg/L in their drinking water for three consecutive days. Plasma, liver, kidney, lung, muscle, and skin + fat samples were collected from 6 chickens at 6 h, 1, 3, 5, and 7 days after the treatment. Tilmicosin concentrations in the samples were determined using a high-performance liquid chromatography (HPLC) method. The findings revealed that the highest tilmicosin residues were detected in the liver, followed by the kidney, lung, skin + fat, muscle, and plasma. Notably, at 7 days post-treatment, no drug residue was detected in all samples except for the liver and kidney. The non-compartmental model was employed to calculate relevant pharmacokinetic parameters. The elimination half-lives (t1/2λz ) of tilmicosin were as follows, ranked from long to short: skin + fat (45.42 h), liver (44.17 h), kidney (40.06 h), plasma (37.64 h), lung (31.39 h), and muscle (30.05 h). Considering the current residue depletion and the maximum residue limits (MRLs) set by Chinese regulatory authorities, the withdrawal times for tilmicosin were estimated as 18.91, 10.81, and 8.58 days in the kidney, liver, and skin + fat, respectively. A rounded-up value of 19 days was selected as the conclusive withdrawal time. Furthermore, based on the observed tilmicosin concentrations in plasma and lung, combined with previously reported minimum inhibitory concentration (MIC) values against Mycoplasma gallisepticum, the current dosing regimen was deemed adequate for treating Mycoplasma gallisepticum infections in Gushi chickens.

Psychromarinibacter sediminicola sp. nov., a novel moderately halophilic, metabolically diverse bacterium isolated from a solar saltern sediment, and comparison between members of family Roseobacteraceae.

Known for its species abundance and evolutionary status complexity, family Roseobacteraceae is an important subject of many studies on the discovery, identification, taxonomic status, and ecological properties of marine bacteria. This study compared and analyzed the phylogenetic, genomic, biochemical, and chemo taxonomical properties of seven species from three genera (Psychromarinibacter, Lutimaribacter, and Maritimibacter) of the family Roseobacteraceae. Moreover, a novel strain, named C21-152T was isolated from solar saltern sediment in Weihai, China. The values of 16S rRNA gene sequence similarity, the average nucleotide identity (ANI), the average amino acid identity (AAI), and the digital DNA-DNA hybridization (dDDH) between genomes of the novel strain and Psychromarinibacter halotolerans MCCC 1K03203T were 97.19, 78.49, 73.45, and 21.90%, respectively. Genome sequencing of strain C21-152T revealed a complete Sox enzyme system related to thiosulfate oxidization as well as a complete pathway for the final conversion of hydroxyproline to α-ketoglutarate. In addition, strain C21-152T was resistant to many antibiotics and had the ability to survive below 13% salinity. This strain had versatile survival strategies in saline environments including salt-in, compatible solute production and compatible solute transport. Some of its physiological features enriched and complemented the knowledge of the characteristics of the genus Psychromarinibacter. Optimum growth of strain C21-152T occurred at 37 â„ƒ, with 5-6% (w/v) NaCl and at pH 7.5. According to the results of the phenotypic, chemotaxonomic characterization, phylogenetic properties and genome analysis, strain C21-152T should represent a novel specie of the genus Psychromarinibacter, for which the name Psychromarinibacter sediminicola sp. nov. is proposed. The type strain is C21-152T (= MCCC 1H00808T = KCTC 92746T = SDUM1063002T).

Novel urinary protein panels for the non-invasive diagnosis of non-alcoholic fatty liver disease and fibrosis stages.

BACKGROUND & AIMS: There is an unmet clinical need for non-invasive tests to diagnose non-alcoholic fatty liver disease (NAFLD) and individual fibrosis stages. We aimed to test whether urine protein panels could be used to identify NAFLD, NAFLD with fibrosis (stage F ≥ 1) and NAFLD with significant fibrosis (stage F ≥ 2). METHODS: We collected urine samples from 100 patients with biopsy-confirmed NAFLD and 40 healthy volunteers, and proteomics and bioinformatics analyses were performed in this derivation cohort. Diagnostic models were developed for detecting NAFLD (UPNAFLD model), NAFLD with fibrosis (UPfibrosis model), or NAFLD with significant fibrosis (UPsignificant fibrosis model). Subsequently, the derivation cohort was divided into training and testing sets to evaluate the efficacy of these diagnostic models. Finally, in a separate independent validation cohort of 100 patients with biopsy-confirmed NAFLD and 45 healthy controls, urinary enzyme-linked immunosorbent assay analyses were undertaken to validate the accuracy of these new diagnostic models. RESULTS: The UPfibrosis model and the UPsignificant fibrosis model showed an AUROC of .863 (95% CI: .725-1.000) and 0.858 (95% CI: .712-1.000) in the training set; and .837 (95% CI: .711-.963) and .916 (95% CI: .825-1.000) in the testing set respectively. The UPNAFLD model showed an excellent diagnostic performance and the area under the receiver operator characteristic curve (AUROC) exceeded .90 in the derivation cohort. In the independent validation cohort, the AUROC for all three of the above diagnostic models exceeded .80. CONCLUSIONS: Our newly developed models constructed from urine protein biomarkers have good accuracy for non-invasively diagnosing liver fibrosis in NAFLD.

Hepatocytic ballooning in non-alcoholic steatohepatitis: Dilemmas and future directions.

Hepatocytic ballooning is a key histological feature in the diagnosis of non-alcoholic steatohepatitis (NASH) and is an essential component of the two most widely used histological scoring systems for diagnosing and staging non-alcoholic fatty liver disease (NAFLD) [namely, the NAFLD activity score (NAS), and the steatosis, activity and fibrosis (SAF) scoring system]. As a result of the increasing incidence of NASH globally, the diagnostic challenges of hepatocytic ballooning are unprecedented. Despite the clear pathological concept of hepatocytic ballooning, there are still challenges in assessing hepatocytic ballooning in 'real life' situations. Hepatocytic ballooning can be confused with cellular oedema and microvesicular steatosis. Significant inter-observer variability does exist in assessing the presence and severity of hepatocytic ballooning. In this review article, we describe the underlying mechanisms associated with hepatocytic ballooning. Specifically, we discuss the increased endoplasmic reticulum stress and the unfolded protein response, as well as the rearrangement of the intermediate filament cytoskeleton, the appearance of Mallory-Denk bodies and activation of the sonic Hedgehog pathway. We also discuss the use of artificial intelligence in the detection and interpretation of hepatocytic ballooning, which may provide new possibilities for future diagnosis and treatment.

Vsig4+ resident single-Kupffer cells improve hepatic inflammation and fibrosis in NASH.

BACKGROUND: The role of macrophages in the pathogenesis of nonalcoholic steatohepatitis (NASH) is complex and unclear. METHODS: Single-cell RNA sequencing was performed on nonparenchymal cells isolated from NASH and control mice. The expression of Vsig4+ macrophages was verified by qPCR, flow cytometry and immunohistochemistry. Primary hepatic macrophages were cocultured with primary hepatocytes or hepatic stellate cells (LX2) cells by Transwell to detect immunofluorescence and oil red O staining. RESULTS: Two main single macrophage subsets were identified that exhibited a significant change in cell percentage when NASH occurred: resident Kupffer cells (KCs; Cluster 2) and lipid-associated macrophages (LAMs; Cluster 13). Nearly 82% of resident single KCs in Cluster 2 specifically expressed Cd163, and an inhibited subgroup of Cd163+ resident single-KCs was suggested to be protective against NASH. Similar to Cd163, Vsig4 was both enriched in and specific to Cluster 2. The percentage of Vsig4+-KCs was significantly decreased in NASH in vivo and in vitro. Hepatocytes and hepatic stellate cells produced less lipid droplet accumulation, proinflammatory protein (TNF-α) and profibrotic protein (α-SMA) in response to coculture with Vsig4+-KCs than in those cocultured with lipotoxic KCs. CONCLUSIONS: A subgroup of Vsig4+ resident single-KCs was shown to improve hepatic inflammation and fibrosis in NASH.

Diterpenoids from Euphorbia fischeriana with Kv1.3 Inhibitory Activity.

Euphorbia diterpenoids possess inhibitory effects of Kv1.3 ion channel, but most of this research has focused on diterpenoids with jatrophane-related or ingenane-related skeletons. In the present study, nine undescribed (1-9) and 16 known (10-25) diterpenoids, based on jatrophane, lathyrane, ingenane, abietane, and atisane skeletons, were identified from the methanol extract of the aerial parts of Euphorbia fischeriana. The structures were established by analysis of the spectroscopic data as well as by single-crystal X-ray diffraction analysis. Among the isolated diterpenoids, macrocyclic jatrophanes and lathyranes exerted Kv1.3 blocking activity. Compound 8 exhibited good selectivity on the inhibition of the Kv 1.3 channel rather than hERG channel, with a selectivity index over 7.0. The selective activity of lathyrane diterpenoids indicates that macrocyclic diterpenoids have the potential to be further investigated as therapeutic agents for the treatment of autoimmune diseases.

Design and synthesis of pseudo-rutaecarpines as potent anti-inflammatory agents via regulating MAPK/NF-κB pathways to relieve inflammation-induced acute liver injury in mice.

Pseudo-natural products (PNPs) design strategy provides a great valuable entrance to effectively identify of novel bioactive scaffolds. In this report, novel pseudo-rutaecarpines were designed via the combination of several privileged structure units and 46 target compounds were synthesized. Most of them display moderate to potent inhibitory effect on LPS-induced NO production and low cytotoxicity in RAW264.7 macrophage. The results of the anti-inflammatory efficacy and action mechanism of compounds 7l and 8c indicated that they significantly reduced the release of IL-6, IL-1ß and TNF-α. Further studies revealed that they can strongly inhibit the activation of NF-κB and MAPK signal pathways. The LPS-induced acute liver injury mice model studies not only confirmed their anti-inflammatory efficacy in vivo but also could effectively relieve the liver injury in mice. The results suggest that compounds 7l and 8c might serve as lead compounds to develop therapeutic drugs for treatment of inflammation.

Myocardial Work for Dynamic Monitoring of Myocardial Injury in Neonatal Asphyxia.

To assess the value of parameters of myocardial work for dynamic monitoring of myocardial injury after neonatal asphyxia. Fifty-three neonates with asphyxia admitted within 24 h after delivery were divided into a mild asphyxia group (n = 40) and severe asphyxia group (n = 13). Echocardiography was performed within 24 h post-birth, within 72 h post-birth (48 h after first echo), and during recovery. The left ventricular ejection fraction on M-mode echocardiography and by Simpson's biplane method (LVEF and Bi-EF, respectively), stroke volume (SV), cardiac output (CO), cardiac index (CI), global longitudinal strain (GLS), global work index (GWI), global constructive work (GCW), and other parameters were measured. Echocardiographic indicators were compared between groups and over time. GWI was significantly increased at 72 h in the mild asphyxia group (P < 0.05) but showed no significant change over time in the severe asphyxia group (P > 0.05). While GCW increased significantly over time in both groups (P < 0.05), it increased earlier in the mild asphyxia group. Time and grouping factors had independent effects on GWI and GCW (P > 0.05). The characteristics of differences in GWI and GCW between the two groups were different from those for LVEF, Bi-EF, SV, CO, CI, and GLS and their change characteristics with improvement from treatment. GWI and GCW changed significantly during recovery from neonatal asphyxia, and their change characteristics differed between mild and severe asphyxia cases. Myocardial work parameters can be used as valuable supplements to traditional indicators of left ventricular function to dynamically monitor the recovery from myocardial injury after neonatal asphyxia.

Towards Adversarial Robustness for Multi-Mode Data through Metric Learning.

Adversarial attacks have become one of the most serious security issues in widely used deep neural networks. Even though real-world datasets usually have large intra-variations or multiple modes, most adversarial defense methods, such as adversarial training, which is currently one of the most effective defense methods, mainly focus on the single-mode setting and thus fail to capture the full data representation to defend against adversarial attacks. To confront this challenge, we propose a novel multi-prototype metric learning regularization for adversarial training which can effectively enhance the defense capability of adversarial training by preventing the latent representation of the adversarial example changing a lot from its clean one. With extensive experiments on CIFAR10, CIFAR100, MNIST, and Tiny ImageNet, the evaluation results show the proposed method improves the performance of different state-of-the-art adversarial training methods without additional computational cost. Furthermore, besides Tiny ImageNet, in the multi-prototype CIFAR10 and CIFAR100 where we reorganize the whole datasets of CIFAR10 and CIFAR100 into two and ten classes, respectively, the proposed method outperforms the state-of-the-art approach by 2.22% and 1.65%, respectively. Furthermore, the proposed multi-prototype method also outperforms its single-prototype version and other commonly used deep metric learning approaches as regularization for adversarial training and thus further demonstrates its effectiveness.

Pharmacokinetics of danofloxacin after single oral and intravenous administration in non-laying hens.

The current study aimed to explore the pharmacokinetics of danofloxacin in non-laying hens after a single oral (PO) and intravenous (IV) dose, both at 5 mg/kg body weight (BW). Eighteen 13-week-old healthy hens were equally and randomly divided into two groups. After both doses, blood samples (approximately 1 ml) were collected at different time points. Danofloxacin concentrations were quantified by a validated high-performance liquid chromatography (HPLC) method followed by a non-compartmental analysis using the software of WinNonLin. The elimination half-lives (t1/2λz s) after PO and IV routes were determined as 8.15 ± 3.37 and 7.69 ± 3.40 h, respectively. After IV administration, danofloxacin had an initial concentration (C0 ) of 3.62 µg/ml, a volume of distribution at steady state (VSS ) of 3579.72 ± 454.29 ml/kg, and a total body clearance (Cl) of 0.49 ml/h/g. After PO administration, the absolute bioavailability and absorption half-life (t1/2ka ) were calculated as 100.99% ± 23.10% and 0.82 ± 0.58 h, respectively. Based on the calculated ratio values of AUC/MIC and Cmax /MIC, an oral dose of 5 mg/kg danofloxacin would be expected to successfully treat hens infected with strains with MIC values ≤0.1 µg/ml.

Chemical Constituents of Thesium chinense Turcz and Their In Vitro Antioxidant, Anti-Inflammatory and Cytotoxic Activities.

Three novel compounds (1-3) along with twenty-six known compounds, two known steroids (4-5) and twenty-four known phenylpropanoids (6-29) were isolated from the whole plant of Thesium chinense Turcz. The structures of the three new compounds were elucidated on the basis of ESI-MS, HR-ESIMS, 1D and 2D NMR, IR, UV spectroscopic data. The absolute stereochemistry of compound 1 was determined by the Gauge-Including Atomic Orbitals (GIAO) method. The in vitro antioxidant, anti-inflammatory and cytotoxic activities of the isolated compounds were evaluated by DPPH radical-scavenging assay, LPS-activated RAW 264.7 cells model and CCK-8 kit, respectively. Compound 11 showed high antioxidant activity with an SC50 value of 16.2 ± 1.6 µM. Compound 21 showed considerable anti-inflammatory activity with an IC50 value of 28.6 ± 3.0 µM. Compounds 4 and 5 displayed potent cytotoxic activity against human NCI-H292, SiHa, A549, and MKN45 cell lines, with the compound 4 having IC50 values of 17.4 ± 2.4, 22.2 ± 1.1, 9.7 ± 0.9, 9.5 ±0.7 µM, and the compound 5 having all IC50 values less than 0.1 µM in vitro.

Towards a Controllable and Reversible Privacy Protection System for Facial Images through Enhanced Multi-Factor Modifier Networks.

Privacy protection data processing has been critical in recent years when pervasively equipped mobile devices could easily capture high-resolution personal images and videos that may disclose personal information. We propose a new controllable and reversible privacy protection system to address the concern in this work. The proposed scheme can automatically and stably anonymize and de-anonymize face images with one neural network and provide strong security protection with multi-factor identification solutions. Furthermore, users can include other attributes as identification factors, such as passwords and specific facial attributes. Our solution lies in a modified conditional-GAN-based training framework, the Multi-factor Modifier (MfM), to simultaneously accomplish the function of multi-factor facial anonymization and de-anonymization. It can successfully anonymize face images while generating realistic faces satisfying the conditions specified by the multi-factor features, such as gender, hair colors, and facial appearance. Furthermore, MfM can also de-anonymize de-identified faces to their corresponding original ones. One crucial part of our work is design of physically meaningful information-theory-based loss functions, which include mutual information between authentic and de-identification images and mutual information between original and re-identification images. Moreover, extensive experiments and analyses show that, with the correct multi-factor feature information, the MfM can effectively achieve nearly perfect reconstruction and generate high-fidelity and diverse anonymized faces to defend attacks from hackers better than other methods with compatible functionalities. Finally, we justify the advantages of this work through perceptual quality comparison experiments. Our experiments show that the resulting LPIPS (with a value of 0.35), FID (with a value of 28), and SSIM (with a value of 0.95) of MfM demonstrate significantly better de-identification effects than state-of-the-art works. Additionally, the MfM we designed can achieve re-identification, which improves real-world practicability.