Synthesis and biological evaluation of C-4 substituted phenoxazine-bearing hydroxamic acids with potent class II histone deacetylase inhibitory activities.

Class II histone deacetylases (HDACs) are considered as potential targets to treat Alzheimer's disease (AD). Previously, C-3 substituted phenothiazine-containing compounds with class II HDAC-inhibiting activities was found to promote neurite outgrowth. This study replaced phenothiazine moiety with phenoxazine that contains many C-3 and C-4 substituents. Some resulting compounds bearing the C-4 substituent on a phenoxazine ring displayed potent class II HDAC inhibitory activities. Structure-activity relationship (SAR) of these compounds that inhibited HDAC isoenzymes was disclosed. Molecular modelling analysis demonstrates that the potent activities of C-4 substituted compounds probably arise from π-π stacked interactions between these compounds and class IIa HDAC enzymes. One of these, compound 7d exhibited the most potent class II HDAC inhibition (IC50= 3-870 nM). Notably, it protected neuron cells from H2O2-induced neuron damage at sub-µM concentrations, but with no significant cytotoxicity. These findings show that compound 7d is a lead compound for further development of anti-neurodegenerative agents.

Identification of a dual FLT3 and MNK2 inhibitor for acute myeloid leukemia treatment using a structure-based virtual screening approach.

Fms-like tyrosine kinase 3 (FLT3) is considered a promising therapeutic target for acute myeloid leukemia (AML) in the clinical. However, monotherapy with FLT3 inhibitor is usually accompanied by drug resistance. Dual inhibitors might be therapeutically beneficial to patients with AML due to their ability to overcome drug resistance. Mitogen-activated protein kinase (MAPK)-interacting kinases (MNKs) phosphorylate eukaryotic translation initiation factor 4E (eIF4E), which brings together the RAS/RAF/ERK and PI3K/AKT/mTOR oncogenic pathways. Therefore, dual inhibition of FLT3 and MNK2 might have an additive effect against AML. Herein, a structure-based virtual screening approach was performed to identify dual inhibitors of FLT3 and MNK2 from the ChemDiv database. Compound K783-0308 was identified as a dual inhibitor of FLT3 and MNK2 with IC50 values of 680 and 406 nM, respectively. In addition, the compound showed selectivity for both FLT3 and MNK2 in a panel of 82 kinases. The structure-activity relationship analysis and common interactions revealed interactions between K783-0308 analogs and FLT3 and MNK2. Furthermore, K783-0308 inhibited MV-4-11 and MOLM-13 AML cell growth and induced G0/G1 cell cycle arrest. Taken together, the dual inhibitor K783-0308 showed promising results and can be potentially optimized as a lead compound for AML treatment.

Synthesis of Yakuchinone B-Inspired Inhibitors against Islet Amyloid Polypeptide Aggregation.

Type 2 diabetes mellitus (T2DM) is associated with pancreatic ß-cell dysfunction and insulin resistance. Islet amyloid polypeptide (IAPP) aggregation is found to induce islet ß-cell death in T2DM patients. Recently, we demonstrated that yakuchinone B derivative 1 exhibited inhibitory activity against IAPP aggregation. Thus, in this study, a series of synthesized yakuchinone B-inspired compounds were tested for their anti-IAPP aggregation activity. Four of these compounds, 4e-h, showed greater activity than the lead compound 1, in the sub-µM range (IC50 = 0.7-0.8 µM). The molecular docking analysis revealed crucial hydrogen bonds between the compounds and residues S19 and N22 and important hydrophobic interactions with residue I26. Notably, compounds 4g and 4h significantly protected ß-cells against IAPP-induced toxicity with EC50 values of 0.1 and 0.2 µM, respectively. In contrast, the protective activities of compounds 4e and 4f were weak. Overall, these results suggest that the compounds exhibiting IAPP aggregation-inhibiting activity have the potential to treat T2DM.

Investigation of Selected Flavonoid Derivatives as Potent FLT3 Inhibitors for the Potential Treatment of Acute Myeloid Leukemia.

Acute myeloid leukemia (AML) is an aggressive disease with a poor prognosis and a high degree of relapse seen in patients. Overexpression of FMS-like tyrosine kinase 3 (FLT3) is associated with up to 70% of AML patients. Wild-type FLT3 induces proliferation and inhibits apoptosis in AML cells, while uncontrolled proliferation of FLT3 kinase activity is also associated with FLT3 mutations. Therefore, inhibiting FLT3 activity is a promising AML therapy. Flavonoids are a group of phytochemicals that can target protein kinases, suggesting their potential antitumor activities. In this study, several plant-derived flavonoids have been identified with FLT3 inhibitory activity. Among these compounds, compound 40 (5,7,4'-trihydroxy-6-methoxyflavone) exhibited the most potent inhibition against not only FLT3 (IC50 = 0.44 µM) but also FLT3-D835Y and FLT3-ITD mutants (IC50 = 0.23 and 0.39 µM, respectively). The critical interactions between the FLT3 binding site and the compounds were identified by performing a structure-activity relationship analysis. Furthermore, the results of cellular assays revealed that compounds 28, 31, 32, and 40 exhibited significant cytotoxicity against two human AML cell lines (MOLM-13 and MV-4-11), and compounds 31, 32, and 40 resulted in cell apoptosis and G0/G1 cell cycle arrest. Collectively, these flavonoids have the potential to be further optimized as FLT3 inhibitors and provide valuable chemical information for the development of new AML drugs.

Identification of a dual TAOK1 and MAP4K5 inhibitor using a structure-based virtual screening approach.

The STE20 kinase family is a complex signalling cascade that regulates cytoskeletal organisation and modulates the stress response. This signalling cascade includes various kinase mediators, such as TAOK1 and MAP4K5. The dysregulation of the STE20 kinase pathway is linked with cancer malignancy. A small-molecule inhibitor targeting the STE20 kinase pathway has therapeutic potential. In this study, a structure-based virtual screening (SBVS) approach was used to identify potential dual TAOK1 and MAP4K5 inhibitors. Enzymatic assays confirmed three potential dual inhibitors (>50% inhibition) from our virtual screening, and analysis of the TAOK1 and MAP4K5 binding sites indicated common interactions for dual inhibition. Compound 1 revealed potent inhibition of colorectal and lung cancer cell lines. Furthermore, compound 1 arrested cancer cells in the G0/G1 phase, which suggests the induction of apoptosis. Altogether, we show that the STE20 signalling mediators TAOK1 and MAP4K5 are promising targets for drug research.

Biological Evaluation of Selected Flavonoids as Inhibitors of MNKs Targeting Acute Myeloid Leukemia.

Excessive eIF4E phosphorylation by mitogen-activated protein kinase (MAPK)-interacting kinases 1 and 2 (MNK1 and MNK2; collectively, MNKs) has been associated with oncogenesis. The overexpression of eIF4E in acute myeloid leukemia (AML) is related to cancer cell growth and survival. Thus, the inhibition of MNKs and eIF4E phosphorylation are potential therapeutic strategies for AML. Herein, a structure-based virtual screening approach was performed to identify potential MNK inhibitors from natural products. Three flavonoids, apigenin, hispidulin, and luteolin, showed MNK2 inhibitory activity with IC50 values of 308, 252, and 579 nM, respectively. A structure-activity relationship analysis was performed to disclose the molecular interactions. Furthermore, luteolin exhibited substantial inhibitory efficacy against MNK1 (IC50 = 179 nM). Experimental results from cellular assays showed that hispidulin and luteolin inhibited the growth of MOLM-13 and MV4-11 AML cells by downregulating eIF4E phosphorylation and arresting the cell cycle at the G0/G1 phase. Therefore, hispidulin and luteolin showed promising results as lead compounds for the potential treatment for AML.

Total Synthesis and Metabolic Stability of Hispidulin and Its d-Labelled Derivative.

Hispidulin is a naturally occurring flavone known to have various Central nervous system (CNS) activities. Proposed synthetic approaches to synthesizing hispidulin have proven unsatisfactory due to their low feasibility and poor overall yields. To solve these problems, this study developed a novel scheme for synthesizing hispidulin, which had an improved overall yield as well as more concise reaction steps compared to previous methods reported. Additionally, using the same synthetic strategy, d-labelled hispidulin was synthesized to investigate its metabolic stability against human liver microsome. This work may produce new chemical entities for enriching the library of hispidulin-derived compounds.

Total Synthesis of Hispidulin and the Structural Basis for Its Inhibition of Proto-oncogene Kinase Pim-1.

A new method is applied to synthesize hispidulin, a natural flavone with a broad spectrum of biological activities. Hispidulin exhibits inhibitory activity against the oncogenic protein kinase Pim-1. Crystallographic analysis of Pim-1 bound to hispidulin reveals a binding mode distinct from that of quercetin, suggesting that the binding potency of flavonoids is determined by their hydrogen-bonding interactions with the hinge region of the kinase. Overall, this work may facilitate construction of a library of hispidulin-derived compounds for investigating the structure-activity relationship of flavone-based Pim-1 inhibitors.

Synthesis of site-specific Fab-drug conjugates using ADP-ribosyl cyclases.

Targeted delivery of small-molecule drugs via covalent attachments to monoclonal antibodies has proved successful in clinic. For this purpose, full-length antibodies are mainly used as drug-carrying vehicles. Despite their flexible conjugation sites and versatile biological activities, intact immunoglobulins with conjugated drugs, which feature relatively large molecular weights, tend to have restricted tissue distribution and penetration and low fractions of payloads. Linking small-molecule therapeutics to other formats of antibody may lead to conjugates with optimal properties. Here, we designed and synthesized ADP-ribosyl cyclase-enabled fragment antigen-binding (Fab) drug conjugates (ARC-FDCs) by utilizing CD38 catalytic activity. Through rapidly forming a stable covalent bond with a nicotinamide adenine dinucleotide (NAD+ )-based drug linker at its active site, CD38 genetically fused with Fab mediates robust site-specific drug conjugations via enzymatic reactions. Generated ARC-FDCs with defined drug-to-Fab ratios display potent and antigen-dependent cytotoxicity against breast cancer cells. This work demonstrates a new strategy for developing site-specific FDCs. It may be applicable to different antibody scaffolds for therapeutic conjugations, leading to novel targeted agents.

Inhibition of Acetylcholinesterase and Amyloid-ß Aggregation by Piceatannol and Analogs: Assessing In Vitro and In Vivo Impact on a Murine Model of Scopolamine-Induced Memory Impairment.

Currently, no drug is effective in delaying the cognitive impairment of Alzheimer's disease, which ranks as one of the top 10 causes of death worldwide. Hydroxylated stilbenes are active compounds that exist in fruit and herbal plants. Piceatannol (PIC) and gnetol (GNT), which have one extra hydroxyl group in comparison to resveratrol (RSV), and rhapontigenin (RHA) and isorhapontigenin (isoRHA), which were metabolized from PIC in vivo and contain the same number of hydroxyl groups as RSV, were evaluated for their effects on Alzheimer's disease-associated factors in vitro and in animal experiments. Among the five hydroxylated stilbenes, PIC was shown to be the most active in DPPH radical scavenging and in inhibitory activities against acetylcholinesterase and amyloid-ß peptide aggregations, with concentrations for half-maximal inhibitions of 40.2, 271.74, and 0.48 µM. The different interactions of the five hydroxylated stilbenes with acetylcholinesterase or amyloid-ß were obtained by molecular docking. The scopolamine-induced ICR mice fed with PIC (50 mg/kg) showed an improved learning behavior in the passive avoidance tests and had significant differences (p < 0.05) compared with those in the control group. The RHA and isoRHA at 10 µM were proven to stimulate neurite outgrowths in the SH-SY5Y cell models. These results reveal that nutraceuticals or functional foods containing PIC have the potential for use in the treatment of neurodegenerative disorders.

Extracts and Scirpusin B from Recycled Seeds and Rinds of Passion Fruits (Passiflora edulis var. Tainung No. 1) Exhibit Improved Functions in Scopolamine-Induced Impaired-Memory ICR Mice.

In this paper, the seeds and rinds of passion fruit, which are the agricultural waste of juice processing, were recycled to investigate their biological activities for sustainable use. De-oiled seed powders (S) were successively extracted by refluxing 95% ethanol (95E), 50E, and hot water (HW), respectively, to obtain S-95EE, S-50EE, and S-HWE. Dried rind powders were successively extracted by refluxing HW and 95E to obtain rind-HWE and rind-95EE, respectively. S-50EE and S-95EE showed the most potent extracts, such as anti-amyloid-ß1-42 aggregations and anti-acetylcholinesterase inhibitors, and they exhibited neuroprotective activities against amyloid-ß25-35-treated or H2O2-treated SH-SY5Y cells. Scirpusin B and piceatannol were identified in S-95EE, S-50EE, and rind-HWE, and they showed anti-acetylcholinesterase activity at 50% inhibitory concentrations of 62.9 and 258.9 µM, respectively. Daily pretreatments of de-oiled seed powders and rind-HWE (600 mg/kg), S-95EE, and S-50EE (250 mg/kg) or scirpusin B (40 mg/kg) for 7 days resulted in improved learning behavior in passive avoidance tests and had significant differences (p < 0.05) compared with those of the control in scopolamine-induced ICR mice. The seeds and rinds of passion fruit will be recycled as materials for the development of functional foods, promoting neuroprotection and delaying the onset of cognitive dysfunctions.

O-methylated flavonol as a multi-kinase inhibitor of leukemogenic kinases exhibits a potential treatment for acute myeloid leukemia.

BACKGROUND: Acute myeloid leukemia (AML) is a heterogeneous disease with poor overall survival characterized by various genetic changes. The continuous activation of oncogenic pathways leads to the development of drug resistance and limits current therapeutic efficacy. Therefore, a multi-targeting inhibitor may overcome drug resistance observed in AML treatment. Recently, groups of flavonoids, such as flavones and flavonols, have been shown to inhibit a variety of kinase activities, which provides potential opportunities for further anticancer applications. PURPOSE: In this study, we evaluated the anticancer effects of flavonoid compounds collected from our in-house library and investigated their potential anticancer mechanisms by targeting multiple kinases for inhibition in AML cells. METHODS: The cytotoxic effect of the compounds was detected by cell viability assays. The kinase inhibitory activity of the selected compound was detected by kinase-based and cell-based assays. The binding conformation and interactions were investigated by molecular docking analysis. Flow cytometry was used to evaluate the cell cycle distribution and cell apoptosis. The protein and gene expression were estimated by western blotting and qPCR, respectively. RESULTS: In this study, an O-methylated flavonol (compound 11) was found to possess remarkable cytotoxic activity against AML cells compared to treatment in other cancer cell lines. The compound was demonstrated to act against multiple kinases, which play critical roles in survival signaling in AML, including FLT3, MNK2, RSK, DYRK2 and JAK2 with IC50 values of 1 - 2 µM. Compared to our previous flavonoid compounds, which only showed inhibitions against MNKs or FLT3, compound 11 exhibited multiple kinase inhibitory abilities. Moreover, compound 11 showed effectiveness in inhibiting internal tandem duplications of FLT3 (FLT3-ITDs), which accounts for 25% of AML cases. The interactions between compound 11 and targeted kinases were investigated by molecular docking analysis. Mechanically, compound 11 caused dose-dependent accumulation of leukemic cells at the G0/G1 phase and followed by the cells undergoing apoptosis. CONCLUSION: O-methylated flavonol, compound 11, can target multiple kinases, which may provide potential opportunities for the development of novel therapeutics for drug-resistant AMLs. This work provides a good starting point for further compound optimization.

A novel dual HDAC and HSP90 inhibitor, MPT0G449, downregulates oncogenic pathways in human acute leukemia in vitro and in vivo.

Acute leukemia is a highly heterogeneous disease; therefore, combination therapy is commonly used for patient treatment. Drug-drug interaction is a major concern of combined therapy; hence, dual/multi-target inhibitors have become a dominant approach for cancer drug development. HDACs and HSP90 are involved in the activation of various oncogenic signaling pathways, including PI3K/AKT/mTOR, JAK/STAT, and RAF/MEK/ERK, which are also highly enriched in acute leukemia gene expression profiles. Therefore, we suggest that dual HDAC and HSP90 inhibitors could represent a novel therapeutic approach for acute leukemia. MPT0G449 is a dual effect inhibitor, and it showed cytotoxic effectiveness in acute leukemia cells. Molecular docking analysis indicated that MPT0G449 possessed dual HDAC and HSP90 inhibitory abilities. Furthermore, MPT0G449 induced G2 arrest and caspase-mediated cell apoptosis in acute leukemia cells. The oncogenic signaling molecules AKT, mTOR, STAT3, STAT5, MEK, and ERK were significantly downregulated after MPT0G449 treatment in HL-60 and MOLT-4 cells. In vivo xenograft models confirmed the antitumor activity and showed the upregulation of acetyl-histone H3 and HSP70, biomarkers of pan-HDAC and HSP90 inhibition, with MPT0G449 treatment. These findings suggest that the dual inhibition of HDAC and HSP90 can suppress the expression of oncogenic pathways in acute leukemia, and MPT0G449 represents a novel therapeutic for anticancer treatment.

Acetylcholinesterase inhibitory activity and neuroprotection in vitro, molecular docking, and improved learning and memory functions of demethylcurcumin in scopolamine-induced amnesia ICR mice.

In this study, demethylcurcumin (DC), a minor constituent in curcuminoids, showed better anti-acetylcholinesterase (anti-AChE) activities, anti-amyloid ß peptide aggregation, neuroprotective activities in 6-hydroxydopamine-treated SH-SY5Y cell models, and anti-nitric oxide production in lipopolysaccharide-treated RAW 264.7 macrophages than those of curcumin. Based on molecular docking analyses with AChE, the meta-hydroxyl group in DC, nonexistent in curcumin, showed the formation of hydrogen bonds with Ser293 and Tyr341 in the binding sites of AChE. For animal experiments, scopolamine-induced amnesia ICR mice were used to analyze the learning and memory functions of DC in comparison with the positive control donepezil. Mice fed with DC (50 mg kg-1) or donepezil (5 mg kg-1) showed improvement and a significant difference compared to those in the control group (P < 0.05, 0.01, or 0.001) in a passive avoidance test and in a water maze probe test. The brain extracts of the mice in the DC or donepezil group showed reduced AChE activities and higher ORAC activities and also showed a significant difference compared to those in the control group (P < 0.05, 0.01, or 0.001). DC might be beneficial for developing functional foods or as a lead compound for the treatment of degenerative disorders.

Synthesis and biological evaluation of acridine-based histone deacetylase inhibitors as multitarget agents against Alzheimer's disease.

Multitarget agents simultaneously trigger molecules in functionally complementary pathways, and are therefore considered to have potential in effectively treating Alzheimer's disease (AD), which has a complex pathogenetic mechanism. In this study, the HDAC inhibitor core is incorporated into the acetylcholine esterase (ACE) inhibitor acridine-derived moiety and resulted in compounds that exhibited higher class IIa HDAC (4, 5, 7, and 9)- and class IIb HDAC6-inhibiting activity when compared to the pan-HDAC inhibitor SAHA in clinical practice. One of these compounds, 11b, displayed greater selectivity toward HDAC6 than other isoform enzymes. In contrast, the activity of compound 6a was selective toward class IIa HDAC and HDAC6. These two compounds exhibited strong activity against Aß-aggregation as well as significantly disrupted Aß-oligomer. Additionally, 11b and 6a strongly inhibited AChE. These experimental findings demonstrate that compounds 11b and 6a are HDAC-Aß-aggregation-AChE inhibitors. Notably, they can enhance neurite outgrowth, but with no significant neurotoxicity. Further biological evaluation revealed the various cellular effects of multitarget compounds 11b and 6a, which have the potential to treat AD.

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs.

In this work we address the task of semantic image segmentation with Deep Learning and make three main contributions that are experimentally shown to have substantial practical merit. First, we highlight convolution with upsampled filters, or 'atrous convolution', as a powerful tool in dense prediction tasks. Atrous convolution allows us to explicitly control the resolution at which feature responses are computed within Deep Convolutional Neural Networks. It also allows us to effectively enlarge the field of view of filters to incorporate larger context without increasing the number of parameters or the amount of computation. Second, we propose atrous spatial pyramid pooling (ASPP) to robustly segment objects at multiple scales. ASPP probes an incoming convolutional feature layer with filters at multiple sampling rates and effective fields-of-views, thus capturing objects as well as image context at multiple scales. Third, we improve the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. The commonly deployed combination of max-pooling and downsampling in DCNNs achieves invariance but has a toll on localization accuracy. We overcome this by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF), which is shown both qualitatively and quantitatively to improve localization performance. Our proposed "DeepLab" system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching 79.7 percent mIOU in the test set, and advances the results on three other datasets: PASCAL-Context, PASCAL-Person-Part, and Cityscapes. All of our code is made publicly available online.

Discovery of aliphatic-chain hydroxamates containing indole derivatives with potent class I histone deacetylase inhibitory activities.

Histone deacetylase (HDAC) is a validated drug target for various diseases. This study combined indole recognition cap with SAHA, an FDA-approved HDAC inhibitor used to treat cutaneous T-cell lymphoma (CTCL). The structure activity relationship of the resulting compounds that inhibited HDAC was disclosed as well. Some compounds exhibited much stronger inhibitory activities than SAHA. We identified two meta-series compounds 6j and 6k with a two-carbon linker had IC50 values of 3.9 and 4.5 nM for HDAC1, respectively. In contrast, the same oriented compounds with longer carbon chain linkers showed weaker inhibition. The result suggests that the linker chain length greatly contributed to enzyme inhibitory potency. In addition, comparison of enzyme-inhibiting activity between the compounds and SAHA showed that compounds 6j and 6k displayed higher inhibiting activity for class I (HDAC1, -2, -3 and -8). The molecular docking and structure analysis revealed structural differences with the inhibitor cap and metal-binding regions between the HDAC isozymes that affect interactions with the inhibitors and play a key role for selectivity. Further biological evaluation showed multiple cellular effects associated with compounds 6j- and 6k-induced HDAC inhibitory activity.

Design of Diarylheptanoid Derivatives as Dual Inhibitors Against Class IIa Histone Deacetylase and ß-amyloid Aggregation.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with multiple etiologies. Beta-amyloid (Aß) self-aggregation and overexpression of class IIa histone deacetylases (HDACs) are strongly implicated with AD pathogenesis. In this study, a series of novel diarylheptanoid derivatives were designed, synthesized and evaluated for use as dual Aß self-aggregation and class IIa HDAC inhibitors. Among these compounds, 4j, 5c, and 5e displayed effective inhibitions for Aß self-aggregation, HDAC5 activity and HDAC7 activity with IC50 values of <10 µM. The compounds contain three common features: (1) a catechol or pyrogallol moiety, (2) a carbonyl linker and (3) an aromatic ring that can function as an HDAC cap and create hydrophobic interactions with Aß1-42. Furthermore, compounds 4j, 5c, and 5e showed no significant cytotoxicity to human neuroblastoma SH-SY5Y cells and also exhibited neuroprotective effect against H2O2-induced toxicity. Overall, these promising in vitro data highlighted compounds 4j, 5c, and 5e as lead compounds that are worthy for further investigation.

Cerebellar α6 -subunit-containing GABAA receptors: a novel therapeutic target for disrupted prepulse inhibition in neuropsychiatric disorders.

BACKGROUND AND PURPOSE: The pathophysiological role of α6 -subunit-containing GABAA receptors, which are mainly expressed in cerebellar granule cells, remains unclear. Recently, we demonstrated that hispidulin, a flavonoid isolated from a local herb that remitted a patient's intractable motor tics, attenuated methamphetamine-induced hyperlocomotion in mice as a positive allosteric modulator (PAM) of cerebellar α6 GABAA receptors. Here, using hispidulin and a selective α6 GABAA receptor PAM, the pyrazoloquinolinone Compound 6, we revealed an unprecedented role of cerebellar α6 GABAA receptors in disrupted prepulse inhibition of the startle response (PPI), which reflects sensorimotor gating deficits manifested in several neuropsychiatric disorders. EXPERIMENTAL APPROACH: PPI disruptions were induced by methamphetamine and NMDA receptor antagonists in mice. Effects of the tested compounds were measured in Xenopus oocytes expressing recombinant α6 ß3 γ2S GABAA receptors. KEY RESULTS: Hispidulin given i.p. or by bilateral intracerebellar (i.cb.) injection rescued PPI disruptions induced by methamphetamine, ketamine, MK-801 and phencyclidine. Intracerebellar effects of hispidulin were mimicked by Ro15-4513 and loreclezole (two α6 GABAA receptor PAMs), but not by diazepam (an α6 GABAA receptor-inactive benzodiazepine) and were antagonized by furosemide (i.cb.), an α6 GABAA receptor antagonist. Importantly, Compound 6 (i.p.) also rescued methamphetamine-induced PPI disruption, an effect prevented by furosemide (i.cb.). Both hispidulin and Compound 6 potentiated α6 ß3 γ2S GABAA receptor-mediated GABA currents. CONCLUSIONS AND IMPLICATIONS: Positive allosteric modulation of cerebellar α6 GABAA receptors rescued disrupted PPI by attenuating granule cell activity. α6 GABAA receptor-selective PAMs are potential medicines for treating sensorimotor gating deficits in neuropsychiatric disorders. A mechanistic hypothesis is based on evidence for cerebellar contributions to cognitive functioning including sensorimotor gating.

Modeling Image Patches with a Generic Dictionary of Mini-Epitomes.

The goal of this paper is to question the necessity of features like SIFT in categorical visual recognition tasks. As an alternative, we develop a generative model for the raw intensity of image patches and show that it can support image classification performance on par with optimized SIFT-based techniques in a bag-of-visual-words setting. Key ingredient of the proposed model is a compact dictionary of mini-epitomes, learned in an unsupervised fashion on a large collection of images. The use of epitomes allows us to explicitly account for photometric and position variability in image appearance. We show that this flexibility considerably increases the capacity of the dictionary to accurately approximate the appearance of image patches and support recognition tasks. For image classification, we develop histogram-based image encoding methods tailored to the epitomic representation, as well as an "epitomic footprint" encoding which is easy to visualize and highlights the generative nature of our model. We discuss in detail computational aspects and develop efficient algorithms to make the model scalable to large tasks. The proposed techniques are evaluated with experiments on the challenging PASCAL VOC 2007 image classification benchmark.