Glial growth factor 2 treatment alleviates ischemia and reperfusion-damaged integrity of the blood-brain barrier through decreasing Mfsd2a/caveolin-1-mediated transcellular and Pdlim5/YAP/TAZ-mediated paracellular permeability.

The impairment of blood-brain barrier (BBB) integrity is the pathological basis of hemorrhage transformation and vasogenic edema following thrombolysis and endovascular therapy. There is no approved drug in the clinic to reduce BBB damage after acute ischemic stroke (AIS). Glial growth factor 2 (GGF2), a recombinant version of neuregulin-1ß that can stimulates glial cell proliferation and differentiation, has been shown to alleviate free radical release from activated microglial cells. We previously found that activated microglia and proinflammatory factors could disrupt BBB after AIS. In this study we investigated the effects of GGF2 on AIS-induced BBB damage as well as the underlying mechanisms. Mouse middle cerebral artery occlusion model was established: mice received a 90-min ischemia and 22.5 h reperfusion (I/R), and were treated with GGF2 (2.5, 12.5, 50 ng/kg, i.v.) before the reperfusion. We showed that GGF2 treatment dose-dependently decreased I/R-induced BBB damage detected by Evans blue (EB) and immunoglobulin G (IgG) leakage, and tight junction protein occludin degradation. In addition, we found that GGF2 dose-dependently reversed AIS-induced upregulation of vesicular transcytosis increase, caveolin-1 (Cav-1) as well as downregulation of major facilitator superfamily domain containing 2a (Mfsd2a). Moreover, GGF2 decreased I/R-induced upregulation of PDZ and LIM domain protein 5 (Pdlim5), an adaptor protein that played an important role in BBB damage after AIS. In addition, GGF2 significantly alleviated I/R-induced reduction of YAP and TAZ, microglial cell activation and upregulation of inflammatory factors. Together, these results demonstrate that GGF2 treatment alleviates the I/R-compromised integrity of BBB by inhibiting Mfsd2a/Cav-1-mediated transcellular permeability and Pdlim5/YAP/TAZ-mediated paracellular permeability.

Expanding the clinical spectrum of anti-DPPX encephalitis: a multicenter retrospective study.

Objective: Anti-dipeptidyl-peptidase-like protein-6 (DPPX) encephalitis is a rare autoimmune encephalitis, and clinical and experimental information regarding this disease is limited. We conducted this study to comprehensively describe the clinical characteristics, ancillary test results, neuroimaging results, and treatment response in a group of Chinese patients with anti-DPPX encephalitis for better understanding this disease. Methods: We recruited 14 patients who tested positive for anti-DPPX antibodies in the serum and/or cerebrospinal fluid from 11 medical centers between March 2021 and June 2023. This retrospective study evaluated data on symptoms, autoantibody test, auxiliary examinations, treatments, and outcomes. Results: The average age at diagnosis was 45.93 ± 4.62 years (range: 11-72 years), and 9 of the 14 patients were males. The main symptoms included cognitive impairment (50.0%, 7/14), central nervous system hyperexcitability (42.9%, 6/14), gastrointestinal dysfunction (35.7%, 5/14), and psychiatric disorders (35.7%, 5/14). Notably, we discovered specific findings on 18F-fluorodeoxyglucose positron-emission tomography (PET)/magnetic resonance imaging in two patients. Co-existing autoantibodies were identified in two patients. Parainfection was identified in four patients. One patient had other autoimmune diseases, and one had tumor. Eleven patients received immunotherapy and most patients improved at discharge. Surprisingly, three male patients but no female patients relapsed during the 6 months of follow-up. Conclusion: The development and outcome of anti-DPPX encephalitis are variable. Male patients were predominant in our cohort. The most common symptoms were the classical triad of prodromal gastrointestinal dysfunction, cognitive and mental disorders, and central nervous system hyperexcitability. Infections, immune dysregulation, and tumors may be important etiologies. Long-term monitoring of disease development should be done in male patients. Overall, our results highlight novel clinical characteristics of anti-DPPX encephalitis.

Fibroblast growth factor-21 alleviates proteasome injury via activation of autophagy flux in Parkinson's disease.

Parkinson's disease (PD) is one of the most common and complex Neurodegeneration, with an inherited metabolic disorder. Fibroblast growth factor 21 (FGF21), an endocrine hormone that belongs to the fibroblast growth factor superfamily, plays an extensive role in metabolic regulation. However, our understandings of the specific function and mechanisms of FGF21 on PD are still quite limited. Here, we aimed to elucidate the actions and the underlying mechanisms of FGF21 on dopaminergic neurodegeneration using cellular models of parkinsonism. To investigate the effects of FGF21 on dopaminergic neurodegeneration in vitro, proteasome impairment models of PD were utilized. Human dopaminergic neuroblastoma SH-SY5Y cells were treated with the proteasome inhibitor lactacystin (5 µmol/L) for 12 h, then with 50 ng/ml FGF-21 with or without 5 mmol/L of 3-methyladenine.The cells were dissected to assess alterations in autophagy using immunofluorescence, immunoblotting and electron microscopy assays. Our data indicate that FGF21 prevents dopaminergic neuron loss and shows beneficial effects against proteasome impairment induced PD syndrome, indicating it might be a potent candidate for developing novel drugs to deal with PD.

The clinical characteristics of neuronal intranuclear inclusion disease and its relation with inflammation.

BACKGROUND: Neuronal intranuclear inclusion disease (NIID) is a great imitator with a broad spectrum of clinical manifestations that include dementia, parkinsonism, paroxysmal symptoms, peripheral neuropathy, and autonomic dysfunction. Hence, it may also masquerade as other diseases such as Alzheimer's disease, Parkinson's disease, and Charcot-Marie-Tooth disease. Recent breakthroughs on neuroimaging, skin biopsy, and genetic testing have facilitated the diagnosis. However, early identification and effective treatment are still difficult in cases of NIID. OBJECTIVE: To further study the clinical characteristics of NIID and investigate the relationship between NIID and inflammation. METHODS: We systematically evaluated the clinical symptoms, signs, MRI and electromyographical findings, and pathological characteristics of 20 NIID patients with abnormal GGC repeats in the NOTCH2NLC gene. Some inflammatory factors in the patients were also studied. RESULTS: Paroxysmal symptoms such as paroxysmal encephalopathy, stroke-like episodes, and mitochondrial encephalomyopathy lactic acidosis and stroke (MELAS)-like episode were the most common phenotypes. Other symptoms such as cognitive dysfunction, neurogenic bladder, tremor, and vision disorders were also suggestive of NIID. Interestingly, not all patients showed apparent diffusion-weighted imaging (DWI) abnormality or intranuclear inclusions, while abnormal GGC repeats of NOTCH2NLC were seen in all patients. And fevers were noticed in some patients during encephalitic episodes, usually with increasing leukocyte counts and neutrophil ratios. Both IL-6 (p = 0.019) and TNF-α (p = 0.027) levels were significantly higher in the NIID group than in normal controls. CONCLUSION: Genetic testing of NOTCH2NLC may be the best choice in the diagnosis of NIID. Inflammation might be involved in the pathogenesis of NIID.

Protective Properties of the Extract of Chrysanthemum on Patients with Ischemic Stroke.

Investigation of the protective effect of chrysanthemum extract in ischemic strokes patients is among the challenging issues with the traditional hospital system in general and smart technology-based hospitals in particular. In this study, we have evaluated the protective effect of chrysanthemum extract on patients with ischemic stroke by detecting the severity of stroke, neuronal indexes, and oxidative stress biomarkers. For this purpose, forty-six patients with ischemic stroke were randomly divided into the control group (n = 30) and chrysanthemum group (n = 30). The control group received standard stroke treatment, and the chrysanthemum group was treated with chrysanthemum extract 400 mg/day (200 mg/day, twice/day) on the basis of standard treatment. The groups were compared the effect of saffron capsules using the National Institute of Health Stoke Scale (NIHSS), serum neuron specific enolase (NSE), S100, brain-derived neurotrophic factor (BDNF), malondialdehyde (MDA), Su-peroxide dismutase (SOD), and total antioxidant capacity (TAC ) levels, at the time of first day and fourth day after treatment. On the first day after treatment, there was no significant difference in the NIHSS score, serum NSE, S100, BDNF, MDA, SOD, and TAC levels between the chrysanthemum group and the control group (P > 0.05). On the fourth day after treatment, the NIHSS, serum NSE, S100, and MDA levels were significantly reduced in the chrysanthemum group compared to the control group, while the BDNF, SOD, and TAC levels were higher (P < 0.05). In addition, compared to the levels on the first day, the NIHSS, serum NSE, S100, and MDA levels were significantly reduced, and the BDNF, SOD, and TAC levels were increased in the chrysanthemum group on the fourth day (P < 0.05). Chrysanthemum extract has the effects of scavenging oxygen free radicals and antioxidation and has a neuroprotective effect on ischemic stroke patients.

Role of NADPH Oxidase-Induced Hypoxia-Induced Factor-1α Increase in Blood-Brain Barrier Disruption after 2-Hour Focal Ischemic Stroke in Rat.

We recently showed that inhibition of hypoxia-induced factor-1α (HIF-1α) decreased acute ischemic stroke-induced blood-brain barrier (BBB) damage. However, factors that induce the upregulation of HIF-1α expression remain unclear. Nicotinamide adenine dinucleotide phosphate (NADPH) oxidase played a critical role in reperfusion-induced BBB damage after stroke. However, the role of NADPH oxidase in BBB injury during the acute ischemia stage remains unclear. This study is aimed at investigating the role of NADPH oxidase in BBB injury and the expression of HIF-1α after acute ischemic stroke. A sutured middle cerebral artery occlusion (MCAO) model was used to mimic ischemic stroke in rats. Our results show that the inhibition of NADPH oxidase by apocynin can significantly reduce the BBB damage caused by 2 h ischemic stroke accompanied by reducing the degradation of tight junction protein occludin. In addition, treatment with apocynin significantly decreased the upregulation of HIF-1α induced by 2 h MCAO. More importantly, apocynin could also inhibit the MMP-2 upregulation. Of note, HIF-1α was not colocalized with a bigger blood vessel. Taken together, our results showed that inhibition of NADPH oxidase-mediated HIF-1α upregulation reduced BBB damage accompanied by downregulating MMP-2 expression and occludin degradation after 2 h ischemia stroke. These results explored the mechanism of BBB damage after acute ischemic stroke and may help reduce the associated cerebral hemorrhage transformation after thrombolysis and endovascular treatment after ischemic stroke.

Genetic testing of FUS, HTRA2, and TENM4 genes in Chinese patients with essential tremor.

INTRODUCTION: Essential tremor (ET) is one of the most prevalent movement disorders. The genetic etiology of ET has not been well defined although a significant proportion (≥50%) are familial cases. Linkage analysis and genome-wide association studies (GWASs) have identified several risk variants. In recent years, whole-exome sequencing of ET has revealed several specific causal variants in FUS (p.Q290X), HTRA2 (p.G399S), and TENM4 (c.4324 G>A, c.4100C>A, and c.3412G>A) genes. OBJECTIVE: To investigate the genetic contribution of these three genes to ET, the protein-coding sequences of FUS, HTRA2, and TENM4 were analyzed in a total of 238 ET patients and 272 controls from eastern China using direct Sanger sequencing. RESULTS: We identified two synonymous coding single nucleotide polymorphisms (SNPs), rs741810 and rs1052352 in FUS, and three previously reported synonymous SNPs, rs11237621, rs689369, and rs2277277 in TENM4. No nonsynonymous exonic variants were identified in these subjects. We found that the frequency of the rs1052352C allele was significantly higher (P = .001) in the ET group than in the control group. CONCLUSION: Overall, our findings suggest that rs1052352 of FUS might contribute to ET risk in Chinese population.

miR-30b protects nigrostriatal dopaminergic neurons from MPP(+)-induced neurotoxicity via SNCA.

OBJECTIVE: To explore the function of miR-30b in pathogenesis of Parkinson's disease (PD) and its underlying molecular mechanism. MATERIALS AND METHODS: We used 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPP(+)) as a tool for constructing the PD cell model, using miR-30b mimics or inhibitors to manipulate miR-30b level for an experimental model of acquisition. The cell viability of SH-SY5Y was detected by CCK, and luciferase was used to screen the binding of target genes. The protein levels of SNCA were measured by Western blot. Then, we investigate the changes in pro- and anti-apoptotic markers with or without miR-30b treatment. RESULTS: There was a significant low expression of MiR-30b in MPP(+)-induced cells. SH-SY5Y cell viability was rescued by MiR-30b overexpression. Luciferase experiments showed that MiR-30b may bind to the 3'-UTR side of SNCA and inhibited its expression. By Western blot, the SNCA level was markedly decreased by miR-30b. miR-30b attenuated the upregulation of Bax and the depletion of Bcl-2 induced by MPP(+).

Hydrazone-Linked Heptazine Polymeric Carbon Nitrides for Synergistic Visible-Light-Driven Catalysis.

Heptazine-based conjugated polymeric carbon nitrides (PCNs) are promising metal-free photocatalysts, yet their synthesis is challenging due to the electron-deficiency and insolubility of heptazine units. Indeed, heptazine-containing polymers have only been prepared through nucleophilic substitution with amines by using toxic cyameluric chloride as the starting material. Herein, we report the novel and environmentally friendly method for preparing heptazine-based mesoporous PCNs with hydrazone links formed through a simple Schiff base condensation of melem-NH2 and aldehydes. Unlike cyameluric chloride, melem-NH2 is non-toxic, stable, and can be readily obtained from melem and hydrazine in solution. We demonstrate that the hydrazone linkages and the heptazine units synergistically enhance the photocatalytic activity of PCNs in visible-light-driven aerobic oxidation of benzyl alcohol to benzaldehyde. In particular, the polymer constructed from melem-NH2 and p-phthalaldehyde shows 17 times more activity than graphitic carbon nitride (g-C3 N4 ).

A ketone-functionalized carbazolic porous organic framework for sensitive fluorometric determination of p-nitroaniline.

A novel ketone-functionalized carbazolic porous framework named PBPMCz is presented for fluorometric determination of p-nitroaniline (PNA). PBPMCz was prepared by FeCl3-promoted oxidative coupling polymerization of 1,3,5-tris((4-(9H-carbazol-9-yl)phenyl)methanone-1-yl)benzene. The polymer possesses a BET surface area of above 907 m2∙g-1 with a pore volume of 0.72 cm3∙g-1. Compared to the ketone-free framework, the green fluorescence of the probe PBPMCz is more strongly quenched by PNA. Figures of merit include (a) excitation/emission wavelengths of 366/540 nm; (b) a Stern-Volmer constant (Ksv) of 2.2 × 104 M-1, and (c) a detection limit of 1.1 µM. Furthermore, PBPMCz shows different quenching behaviors of PNA compared with o-nitroaniline and m-nitroaniline. The excellent performance of the fluorescent probe is ascribed to the abundant carbazole sites and ketone groups in PBPMCz. These facilitate the electron transfer and hydrogen-bonding interactions between PNA and the polymer. Graphical abstract Schematic presentation of a luminescent carbazolic porous organic framework (CzPOF) modified with keto groups. It shows ultra-sensitivity to quenching by PNA over other nitroaniline isomers.

Dl-NBP (Dl-3-N-Butylphthalide) Treatment Promotes Neurological Functional Recovery Accompanied by the Upregulation of White Matter Integrity and HIF-1α/VEGF/Notch/Dll4 Expression.

Dl-3-n-butylphthalide (dl-NBP) was approved by the FDA of China for the treatment of acute ischemic stroke. Dl-NBP has been shown to promote neurological functional recovery and enhance white matter integrity using an endothelin-1-induced focal permanent cerebral ischemia model, which could mimic those patients who have no opportunity to receive either tissue plasminogen activator (tPA) thrombolysis or endovascular therapy. However, it is not clear whether dl-NBP could promote neurological functional recovery in a focal transient cerebral ischemia model, which could mimic those patients who have the opportunity to receive either tPA thrombolysis or endovascular therapy. In this study, using a model of middle cerebral artery occlusion in mice, we aim to explore the effect of two-week dl-NBP treatment on neurological functional recovery after ischemic stroke as well as its underlying mechanism. Our results showed that dl-NBP treatment promoted functional recovery assessed by neurological scores and an adhesive remove test, and this improved the integrity of white matter after 60-min ischemia and 14-day reperfusion. In addition, dl-NBP increased the number of RECA-1 positive vessels and enhanced the expression of the tight junction protein occludin. More importantly, dl-NBP also promoted the expression of hypoxia-induced factor-1α, the vascular endothelial growth factor, Notch, and delta-like ligand 4. In conclusion, our study provides evidence that dl-NBP treatment could also promote functional recovery after focal transient ischemia stroke, and this recovery is associated with upregulated white matter integrity, microvessels, and the tight junction protein occludin. Our results suggested that, in future, dl-NBP may also be applied in clinic to promote functional recovery during the later phase of focal transient ischemic stroke.

Inhibition of HIF-1α Reduced Blood Brain Barrier Damage by Regulating MMP-2 and VEGF During Acute Cerebral Ischemia.

Increase of blood brain barrier (BBB) permeability after acute ischemia stroke is a predictor to intracerebral hemorrhage transformation (HT) for tissue plasminogen activator (tPA) thrombolysis and post-endovascular treatment. Previous studies showed that 2-h ischemia induced damage of BBB integrity and matrix metalloproteinase-2 (MMP-2) made major contribution to this disruption. A recent study showed that blocking ß2-adrenergic receptor (ß2-AR) alleviated ischemia-induced BBB injury by reducing hypoxia-inducible factor-1 alpha (HIF-1α) level. In this study, we sought to investigate the interaction of HIF-1α with MMP-2 and vascular endothelial growth factor (VEGF) in BBB injury after acute ischemia stroke. Rat suture middle cerebral artery occlusion (MCAO) model was used to mimic ischemia condition. Our results showed that ischemia produced BBB damage and MMP-2/9 upregulation was colocalized with Rhodamine-dextran leakage. Pretreatment with YC-1, a HIF-1α inhibitor, alleviated 2-h ischemia-induced BBB injury significantly accompanied by decrease of MMP-2 upregulation. In addition, YC-1 also prevented VEGF-induced BBB damage. Of note, VEGF was shown to be colocalized with neurons but not astrocytes. Taken together, BBB damage was reduced by inhibition of interaction of HIF-1α with MMP-2 and VEGF during acute cerebral ischemia. These findings provide mechanisms underlying BBB damage after acute ischemia stroke and may help reduce thrombolysis- and post-endovascular treatment-related cerebral hemorrhage.

Study of GCH1 and TH genes in Chinese patients with Parkinson's disease.

Whole-exome sequencing of Parkinson's disease (PD) patients has revealed that the frequency of GTP-cyclohydrolase I (GCH1) variants was significantly higher in patients than in controls. GCH1 rs11158026 was also found to increase the risk of PD. To investigate genetic contribution of dopa-responsive dystonia-related genes to PD, GCH1, and tyrosine hydroxylase (TH) were tested in PD patients. A total of 859 study subjects comprising 421 patients with PD and 438 controls were recruited. For GCH1 gene, one known variant (c.239G > A, p.S80N) was detected in a patient who was diagnosed with PD clinically. In TH, 3 heterozygous variants, c.1495G > A (p. V499M, rs1800033), c.334 A > G (p.V112M, rs6356), and c.813 G > A (p. K271K, rs6357), were identified. After stratification by age, the frequency of rs6356G allele was significantly lower (p = 0.041) for the late-onset PD group than controls. Our results indicate that to analyze the relationship between dopa-responsive dystonia-related genes and PD, it is important to screen GCH1 and test rs6356 of TH in a larger sample.

Spatial-temporal signature of resting-state BOLD signals in classic trigeminal neuralgia.

Resting-state functional magnetic resonance imaging (R-fMRI) signals are spatiotemporally organized. R-fMRI studies in patients with classic trigeminal neuralgia (CTN) have suggested alterations in functional connectivity. However, far less attention has been given to investigations of the local oscillations and their frequency-specific changes in these patients. The objective of this study was to address this issue in patients with CTN. R-fMRI data from 17 patients with CTN and 19 age- and gender-matched healthy controls (HCs) were analyzed using amplitude of low-frequency fluctuation (ALFF). The ALFF was computed across different frequencies (slow-4: 0.027-0.073 Hz; slow-5: 0.01-0.027 Hz; and typical band: 0.01-0.08 Hz) in patients with CTN compared to HCs. In the typical band, patients with CTN showed increases of ALFF in bilateral temporal, occipital, and left middle frontal regions and in the left middle cingulate gyrus, as well as decreases of ALFF in the right inferior temporal region and in regions (medial prefrontal regions) of default mode network. These significant group differences were identified in different sub-bands, with greater brainstem findings in higher frequencies (slow-4) and extensive default mode network and right postparietal results in lower frequencies (slow-5). Furthermore, significant relationships were found between subjective pain ratings and both amplitudes of higher frequency (slow-4) blood oxygen level-dependent (BOLD) signals in pain localization brain regions and lower frequencies (slow-5) in pain signaling/modulating brain regions in the patients, and decreased ALFF within the prefrontal regions was significantly correlated with pain duration in the patients. This result supports our hypothesis that trigeminal pain has a characteristic spatiotemporal distribution of low-frequency BOLD signals. These findings might contribute to a better understanding of the impact of CTN on the brain's intrinsic architecture. Future studies should take the frequencies into account when measuring brain resting BOLD signals of patients with CTN.

Effects of the Novel Fungicide Benzothiostrobin on Sclerotinia sclerotiorum in the Laboratory and on Sclerotinia Stem Rot in Rape Fields.

We determined the effects and efficacy of benzothiostrobin, a new strobilurin-derived fungicide, against the plant-pathogenic fungus Sclerotinia sclerotiorum (the causal agent of Sclerotinia stem rot). Mycelial growth and sclerotial germination in vitro were strongly inhibited by benzothiostrobin in the presence of salicylhydroxamic acid. On detached rapeseed leaves, benzothiostrobin at 40 µg/ml reduced lesion development by 87%. No cross-resistance was detected between benzothiostrobin and carbendazim, iprodione, fludioxonil, or boscalid. A formulated mixture of benzothiostrobin and fluazinam at 1:1 had synergistic activity against S. sclerotiorum in vitro. In field trials, benzothiostrobin alone or formulated with fluazinam at 1:1 (150 g a.i. ha-1) was significantly (P < 0.05) superior to iprodione in controlling Sclerotinia stem rot of rapeseed. These results suggest that benzothiostrobin has substantial potential for the control of Sclerotinia stem rot.

Quantitative Regression Models for the Prediction of Chemical Properties by an Efficient Workflow.

Rapid safety assessment is more and more needed for the increasing chemicals both in chemical industries and regulators around the world. The traditional experimental methods couldn't meet the current demand any more. With the development of the information technology and the growth of experimental data, in silico modeling has become a practical and rapid alternative for the assessment of chemical properties, especially for the toxicity prediction of organic chemicals. In this study, a quantitative regression workflow was built by KNIME to predict chemical properties. With this regression workflow, quantitative values of chemical properties can be obtained, which is different from the binary-classification model or multi-classification models that can only give qualitative results. To illustrate the usage of the workflow, two predictive models were constructed based on datasets of Tetrahymena pyriformis toxicity and Aqueous solubility. The qcv (2) and qtest (2) of 5-fold cross validation and external validation for both types of models were greater than 0.7, which implies that our models are robust and reliable, and the workflow is very convenient and efficient in prediction of various chemical properties.

Activity of a novel strobilurin fungicide benzothiostrobin against Sclerotinia sclerotiorum.

Benzothiostrobin is a novel strobilurin fungicide. In this study, baseline sensitivity of Sclerotinia sclerotiorum (Lib.) de Bary to benzothiostrobin was determined using 100 strains collected during 2012 and 2013 from different geographical regions in Jiangsu Province of China, and the average EC50 value was 0.0218 (± 0.0111)µg/mL for mycelial growth. After benzothiostrobin treatment, hyphae were contorted with offshoot of top increasing and cell membrane permeability increased markedly, while sclerotial production and oxalic acid content significantly decreased. Benzothiostrobin strongly inhibited mycelial respiration within 12h and the oxygen consumption of the mycelia could not be inhibited after 24h. On detached rapeseed leaves, the protective and curative activity test of benzothiostrobin suggested that benzothiostrobin had good control efficiency against S. sclerotiorum, and protective activity was better than curative activity. These results will contribute to us evaluating the potential of the new strobilurin fungicide benzothiostrobin for management of diseases caused by S. sclerotiorum and understanding the mode of action of benzothiostrobin against S. sclerotiorum.

In silico prediction of chemical acute oral toxicity using multi-classification methods.

Chemical acute oral toxicity is an important end point in drug design and environmental risk assessment. However, it is difficult to determine by experiments, and in silico methods are hence developed as an alternative. In this study, a comprehensive data set containing 12, 204 diverse compounds with median lethal dose (LD50) was compiled. These chemicals were classified into four categories, namely categories I, II, III and IV, based on the criterion of the U.S. Environmental Protection Agency (EPA). Then several multiclassification models were developed using five machine learning methods, including support vector machine (SVM), C4.5 decision tree (C4.5), random forest (RF), κ-nearest neighbor (kNN), and naïve Bayes (NB) algorithms, along with MACCS and FP4 fingerprints. One-against-one (OAO) and binary tree (BT) strategies were employed for SVM multiclassification. Performances were measured by two external validation sets containing 1678 and 375 chemicals, separately. The overall accuracy of the MACCS-SVM(OAO) model was 83.0% and 89.9% for external validation sets I and II, respectively, which showed reliable predictive accuracy for each class. In addition, some representative substructures responsible for acute oral toxicity were identified using information gain and substructure frequency analysis methods, which might be very helpful for further study to avoid the toxicity.

In silico prediction of chemical Ames mutagenicity.

Mutagenicity is one of the most important end points of toxicity. Due to high cost and laboriousness in experimental tests, it is necessary to develop robust in silico methods to predict chemical mutagenicity. In this paper, a comprehensive database containing 7617 diverse compounds, including 4252 mutagens and 3365 nonmutagens, was constructed. On the basis of this data set, high predictive models were then built using five machine learning methods, namely support vector machine (SVM), C4.5 decision tree (C4.5 DT), artificial neural network (ANN), k-nearest neighbors (kNN), and naïve Bayes (NB), along with five fingerprints, namely CDK fingerprint (FP), Estate fingerprint (Estate), MACCS keys (MACCS), PubChem fingerprint (PubChem), and Substructure fingerprint (SubFP). Performances were measured by cross validation and an external test set containing 831 diverse chemicals. Information gain and substructure analysis were used to interpret the models. The accuracies of fivefold cross validation were from 0.808 to 0.841 for top five models. The range of accuracy for the external validation set was from 0.904 to 0.980, which outperformed that of Toxtree. Three models (PubChem-kNN, MACCS-kNN, and PubChem-SVM) showed high and reliable predictive accuracy for the mutagens and nonmutagens and, hence, could be used in prediction of chemical Ames mutagenicity.