ABSTRACT
HYPOTHESIS: Metal-organic frameworks (MOFs) are highly suitable precursors for supercapacitor electrode materials owing to their high porosity and stable backbone structures that offer several advantages for redox reactions and rapid ion transport. EXPERIMENTS: In this study, a carbon-coated Ni9S8 composite (Ni9S8@C-5) was prepared via sulfuration at 500 â using a spherical Ni-MOF as the sacrificial template. FINDING: The stable carbon skeleton derived from Ni-MOF and positive structure-activity relationship due to the multinuclear Ni9S8 components resulted in a specific capacity of 278.06 mAh·g-1 at 1 A·g-1. Additionally, the hybrid supercapacitor (HSC) constructed using Ni9S8@C-5 as the positive electrode and the laboratory-prepared coal pitch-based activated carbon (CTP-AC) as the negative electrode achieved an energy density of 69.32 Wh·kg-1 at a power density of 800.06 W·kg-1, and capacity retention of 83.06 % after 5000 cycles of charging and discharging at 5 A·g-1. The Ni-MOF sacrificial template method proposed in this study effectively addresses the challenges associated with structural collapse and agglomeration of Ni9S8 during electrochemical reactions, thus improving its electrochemical performance. Hence, a simple preparation method is demonstrated, with broad application prospects in supercapacitor electrodes.
ABSTRACT
Cardinality constraint, namely, constraining the number of nonzero outputs of models, has been widely used in structural learning. It can be used for modeling the dependencies between multidimensional labels. In hashing, the final outputs are also binary codes, which are similar to multidimensional labels. It has been validated that estimating how many 1's in a multidimensional label vector is easier than directly predicting which elements are 1 and estimating cardinality as a prior step will improve the classification performance. Hence, in this article, we incorporate cardinality constraint into the unsupervised image hashing problem. The proposed model is divided into two steps: 1) estimating the cardinalities of hashing codes and 2) then estimating which bits are 1. Unlike multidimensional labels that are known and fixed in the training phase, the hashing codes are generally learned through an iterative method and, therefore, their cardinalities are unknown and not fixed during the learning procedure. We use a neural network as a cardinality predictor and its parameters are jointly learned with the hashing code generator, which is an autoencoder in our model. The experiments demonstrate the efficiency of our proposed method.
ABSTRACT
Mumps virus (MuV) is highly neurotropic and neurovirulent, hence, the neurovirulence of virus seeds used in the production of mumps vaccines must be tested. The previous neurovirulence evaluation method involves measuring the area of the cavity in the Lewis neonatal rat brain caused by MuV through paraffin sectioning and hematoxylin-eosin (HE) staining. However, the processes of paraffin sectioning and HE staining are time consuming and complicated. To solve this problem, in this study, a vibratome sectioning system was first deployed to evaluate MuV neurovirulence in the rat brain instead of paraffin sectioning and HE staining. The results showed that the vibratome sectioning method could assess the neurovirulence potential of MuV more objectively and efficiently. In addition, the effects of different MuV doses and the ages of the rats in days on this evaluation method were explored. The results indicate that MuV at no less than 10 50 % cell culture infective dose (CCID50) could cause obvious cavity formation in 1-day-old rat brains. The neonatal rat model developed in this study could evaluate the neurovirulence of different MuV strains with high sensitivity and good repeatability.
Subject(s)
Mumps virus , Mumps , Animals , Rats , Rats, Wistar , Paraffin , Eosine Yellowish-(YS) , Hematoxylin , Rats, Inbred Lew , Virulence , Mumps VaccineABSTRACT
Infection with laboratory-attenuated rabies virus (RABV), but not wild-type (wt) RABV, can enhance the permeability of the blood-brain barrier (BBB), which is considered a key determinant for RABV pathogenicity. A previous study showed that the enhancement of BBB permeability is directly due not to RABV infection but to virus-induced inflammatory molecules. In this study, the effect of the matrix metallopeptidase (MMP) family on the permeability of the BBB during RABV infection was evaluated. We found that the expression level of MMP8 was upregulated in mice infected with lab-attenuated RABV but not with wt RABV. Lab-attenuated RABV rather than wt RABV activates inflammatory signaling pathways mediated by the nuclear factor κB (NF-κB) and mitogen-activated protein kinase (MAPK) pathways. Activated NF-κB (p65) and AP-1 (c-Fos) bind to the MMP8 promoter, resulting in upregulation of its transcription. Analysis of mouse brains infected with the recombinant RABV expressing MMP8 indicated that MMP8 enhanced BBB permeability, leading to infiltration of inflammatory cells into the central nervous system (CNS). In brain-derived endothelial cells, treatment with MMP8 recombinant protein caused the degradation of tight junction (TJ) proteins, and the application of an MMP8 inhibitor inhibited the degradation of TJ proteins after RABV infection. Furthermore, an in vivo experiment using an MMP8 inhibitor during RABV infection demonstrated that BBB opening was diminished. In summary, our data suggest that the infection of lab-attenuated RABV enhances the BBB opening by upregulating MMP8. IMPORTANCE The ability to change BBB permeability was associated with the pathogenicity of RABV. BBB permeability was enhanced by infection with lab-attenuated RABV instead of wt RABV, allowing immune cells to infiltrate into the CNS. We found that MMP8 plays an important role in enhancing BBB permeability by degradation of TJ proteins during RABV infection. Using an MMP8 selective inhibitor restores the reduction of TJ proteins. We reveal that MMP8 is upregulated via the MAPK and NF-κB inflammatory pathways, activated by lab-attenuated RABV infection but not wt RABV. Our findings suggest that MMP8 has a critical role in modulating the opening of the BBB during RABV infection, which provides fresh insight into developing effective therapeutics for rabies and infection with other neurotropic viruses.
Subject(s)
Blood-Brain Barrier/metabolism , Matrix Metalloproteinase 8/metabolism , Rabies virus , Rabies/virology , Animals , Brain , Endothelial Cells/metabolism , Matrix Metalloproteinase 8/genetics , Mice , NF-kappa B/metabolismABSTRACT
BACKGROUND: Neurotropic virus infection can cause serious damage to the central nervous system (CNS) in both humans and animals. The complexity of the CNS poses unique challenges to investigate the infection of these viruses in the brain using traditional techniques. METHODS: In this study, we explore the use of fluorescence micro-optical sectioning tomography (fMOST) and single-cell RNA sequencing (scRNA-seq) to map the spatial and cellular distribution of a representative neurotropic virus, rabies virus (RABV), in the whole brain. Mice were inoculated with a lethal dose of a recombinant RABV encoding enhanced green fluorescent protein (EGFP) under different infection routes, and a three-dimensional (3D) view of RABV distribution in the whole mouse brain was obtained using fMOST. Meanwhile, we pinpointed the cellular distribution of RABV by utilizing scRNA-seq. RESULTS: Our fMOST data provided the 3D view of a neurotropic virus in the whole mouse brain, which indicated that the spatial distribution of RABV in the brain was influenced by the infection route. Interestingly, we provided evidence that RABV could infect multiple nuclei related to fear independent of different infection routes. More surprisingly, our scRNA-seq data revealed that besides neurons RABV could infect macrophages and the infiltrating macrophages played at least three different antiviral roles during RABV infection. CONCLUSION: This study draws a comprehensively spatial and cellular map of typical neurotropic virus infection in the mouse brain, providing a novel and insightful strategy to investigate the pathogenesis of RABV and other neurotropic viruses.
Subject(s)
Brain/cytology , Rabies virus/pathogenicity , Rabies/complications , Animals , Brain/abnormalities , Disease Models, Animal , Mice , Rabies/physiopathology , Rabies virus/metabolism , Single-Cell Analysis/methods , Single-Cell Analysis/statistics & numerical data , Tomography, Optical/methods , Tomography, Optical/statistics & numerical dataABSTRACT
Rabies, caused by rabies virus (RABV), is a widespread zoonosis that is nearly 100% fatal. Alteration of the metabolic environment affects viral replication and the immune response during viral infection. In this study, glucose uptake was increased in mouse brains at the late stage of infection with different RABV strains (lab-attenuated CVS strain and wild-type DRV strain). To illustrate the mechanism underlying glucose metabolism alteration, comprehensive analysis of lysine acetylation and target analysis of energy metabolites in mouse brains infected with CVS and DRV strains were performed. A total of 156 acetylated sites and 115 acetylated proteins were identified as significantly different during RABV infection. Compared to CVS- and mock-infected mice, the lysine acetylation levels of glycolysis and tricarboxylic acid (TCA) cycle enzymes were decreased, and enzyme activity was upregulated in DRV-infected mouse brains. Metabolomic analysis revealed high levels of oxaloacetate (OAA) in RABV-infected mouse brains. Specifically, the OAA level in CVS-infected mouse brains was higher than that in DRV-infected mouse brains, which contributed to the enhancement of the metabolic rate at the substrate level. Finally, we confirmed that OAA could reduce excessive neuroinflammation in CVS-infected mouse brains by inhibiting JNK and P38 phosphorylation. Taken together, this study provides fresh insight into the different strategies the host adapts to regulate glucose metabolism for energy requirements after different RABV strain infections and suggests that OAA treatment is a strategy to prevent neural damage during RABV infection. IMPORTANCE Both viral replication and the host immune response are highly energy dependent. It is important to understand how the rabies virus affects energy metabolism in the brain. Glucose is the direct energy source for cell metabolism. Previous studies have revealed that there is some association between acetylation and metabolic processes. In this study, comprehensive protein acetylation and glucose metabolism analysis were conducted to compare glucose metabolism in mouse brains infected with different RABV strains. Our study demonstrates that the regulation of enzyme activity by acetylation and OAA accumulation at the substrate level are two strategies for the host to respond to energy requirements after RABV infection. Our study also indicates the role OAA could play in neuronal protection by suppressing excessive neuroinflammation.
Subject(s)
Brain/metabolism , Glucose/metabolism , Rabies virus/pathogenicity , Rabies/metabolism , Acetylation , Animals , Brain/drug effects , Brain/immunology , Brain/virology , Energy Metabolism , Inflammation , Mice , Neuroprotective Agents/metabolism , Neuroprotective Agents/therapeutic use , Oxaloacetic Acid/metabolism , Oxaloacetic Acid/therapeutic use , Proteome/metabolism , Rabies/drug therapy , Rabies/virologyABSTRACT
The development of science and technology has deepened people's understanding of cancer, changing the management of malignant tumors in the medical field. Given the common precancerous characteristics of colorectal cancer (CRC), researchers studied early CRC screening. The complexity of traditional diagnostics forced medical staff to speed up CRC innovation early screening methods. Here, we prepared nano-colloidal gold raw materials with different particle sizes (15 and 30 nm) and observed the morphological characteristics and properties of the materials. Simultaneously, the nanocolloidal gold double antibody sandwich kit was designed through the optimum pH value and protein content screening experiment. The results of clinical enteroscopy confirmed the important guiding significance of the equipment in early CRC screening.
Subject(s)
Colorectal Neoplasms , Occult Blood , Chromatography, Affinity , Colorectal Neoplasms/diagnosis , Early Detection of Cancer , Gold Colloid , HumansABSTRACT
Rabies virus (RABV) infection can initiate the host immune defence response and induce an antiviral state characterized by the expression of interferon (IFN)-stimulated genes (ISGs), among which the family of genes of IFN-induced protein with tetratricopeptide repeats (Ifits) are prominent representatives. Herein, we demonstrated that the mRNA and protein levels of Ifit1, Ifit2 and Ifit3 were highly increased in cultured cells and mouse brains after RABV infection. Recombinant RABV expressing Ifit3, designated rRABV-Ifit3, displayed a lower pathogenicity than the parent RABV in C57BL/6 mice after intramuscular administration, and Ifit3-deficient mice exhibited higher susceptibility to RABV infection and higher mortality during RABV infection. Moreover, compared with their individual expressions, co-expression of Ifit2 and Ifit3 could more effectively inhibit RABV replication in vitro. These results indicate that murine Ifit3 plays an essential role in restricting the replication and reducing the pathogenicity of RABV. Ifit3 acts synergistically with Ifit2 to inhibit RABV replication, providing further insight into the function and complexity of the Ifit family.
Subject(s)
Intracellular Signaling Peptides and Proteins/metabolism , Rabies virus/physiology , Rabies/virology , Adaptor Proteins, Signal Transducing/genetics , Adaptor Proteins, Signal Transducing/metabolism , Animals , Apoptosis Regulatory Proteins/genetics , Apoptosis Regulatory Proteins/metabolism , Brain/metabolism , Brain/virology , Cell Line , Female , Humans , Immunity, Innate , Intracellular Signaling Peptides and Proteins/genetics , Mice , Mice, Inbred C57BL , RNA, Messenger/genetics , RNA, Messenger/metabolism , RNA-Binding Proteins/genetics , RNA-Binding Proteins/metabolism , Rabies/immunology , Rabies virus/pathogenicity , Transcriptome , Viral Load , Virus ReplicationABSTRACT
Perfluorooctanoic acid (PFOA) is an emerging organic pollutant that has become ubiquitous in waterways and is difficult to be removed from wastewater using traditional treatment methods. In this study, amino-functionalized graphene oxide (AGO) aerogels were prepared as a potential remediation tool for water contaminated by PFOA. The structure of the prepared absorbent material was characterized by Fourier transform infrared spectroscopy, Raman spectroscopy, scanning electron microscope, and X-ray diffraction. The use of various adsorption times, temperatures, solution pH, and absorbent amount were investigated to determine optimum conditions for PFOA adsorption. Adsorption kinetics and thermodynamics of the absorbent were analyzed as well. AGO aerogels exhibited a high adsorption capacity of PFOA (1575 mgâg-1) and high removal efficiency (99.95%) in a solution containing 10 mg PFOA L-1, likely due to the interconnected porous microstructures and amino groups of the AGO aerogels. The adsorption kinetics and isotherm of PFOA were well-fitted using pseudo-second-order and the Freundlich modelling. The adsorption mechanism of PFOA onto AGO aerogels followed spontaneous, exothermic, and physical processes. This study shows the potential of this material to remove PFOA from PFOA-contaminated waters effectively by providing insight into the understanding of the adsorption mechanisms of PFOA onto AGO aerogels.
ABSTRACT
Retrieving nearest neighbors across correlated data in multiple modalities, such as image-text pairs on Facebook and video-tag pairs on YouTube, has become a challenging task due to the huge amount of data. Multimodal hashing methods that embed data into binary codes can boost the retrieving speed and reduce storage requirements. Since unsupervised multimodal hashing methods are usually inferior to supervised ones and the supervised ones require too much manually labeled data, the proposed method in this paper utilizes partially available labels to design a semisupervised multimodal hashing method. The labels for unlabeled data are treated as an interval type-2 fuzzy set and estimated by the available labels. By defuzzifying the estimated labels using hard partitioning, a supervised multimodal hashing method is used to generate binary codes. Experiments show that the proposed semisupervised method with 50% labels can get a medium performance among the compared supervised ones and achieve approximate performance to the best supervised method with 90% labels. With only 10% labels, the proposed method can still compete with the worst compared supervised one. Furthermore, the proposed label estimation method has been experimentally proven to be more feasible for a multilabeled MIRFlickr data set in a hash lookup task.
Subject(s)
Mumps virus , Viral Proteins , Animals , Chlorocebus aethiops , Genotype , Mumps virus/genetics , Vero Cells , Viral Proteins/geneticsABSTRACT
Halogenated amino acids and peptides are an emerging class of disinfection byproducts (DBPs), having been detected in drinking water and in washed food products. However, the toxicological significance of these emerging DBPs remains unclear. In this study, the cytotoxicity of eight halogenated tyrosyl compounds was investigated in Chinese hamster ovary (CHO) cells using real-time cell analysis (RTCA). Dihalogenated tyrosyl compounds are more cytotoxic than their monohalogenated analogues. The cytotoxicity of the dihalogenated compounds is associated with their ability to induce intracellular reactive oxygen species (ROS), suggesting that oxidative stress is an important toxicity pathway of these compounds. Pearson correlation analysis of the cytotoxicity (IC50 values) of these compounds with eight physicochemical parameters showed strong associations with their lipophilicity (logP) and reactivity (polarizability, ELUMO). Finally, cytotoxicity testing of the concentrated extracts of a chloraminated mixture of eight dipeptides with bromide or iodide showed the cytotoxicity of these mixtures in the order: iodinated peptides > brominated peptides ≥ chlorinated peptides. These results demonstrate that halogenated peptide DBPs are toxicologically relevant, and further research is needed to understand the implications of long-term exposure for human health.
Subject(s)
Disinfection , Halogenation , Tyrosine/chemistry , Tyrosine/toxicity , Animals , CHO Cells , Cell Survival/drug effects , Cells, Cultured , Cricetulus , Dose-Response Relationship, Drug , Molecular Structure , Reactive Oxygen Species/analysis , Reactive Oxygen Species/metabolism , Tyrosine/analogs & derivativesABSTRACT
Rabies is one of the most dangerous and widespread zoonosis and is characterized by severe neurological signs and a high case-mortality rate of nearly 100%. Vaccination is the most effective way to prevent rabies in humans and animals. In this study, the relationship between exhaustive exercise and the humoral immune response after immunization with inactivated rabies vaccine was investigated in a mouse model with one-time exhaustive exercise. It was found that compared with the mice with no exercise after vaccination, no significant differences were found in those with exhaustive exercise after vaccination on body-weight changes, virus-neutralizing antibody (VNA) titers, antibody subtypes and survivor ratio after lethal rabies virus (RABV) challenge. This study indicated that exhaustive exercise does not reduce the effects of the rabies inactivated vaccine.
Subject(s)
Rabies Vaccines/therapeutic use , Rabies virus/immunology , Vaccines, Inactivated/therapeutic use , Animals , Antibodies, Viral/immunology , Cell Line , Cricetinae , Disease Models, Animal , Female , Immunity, Humoral/immunology , Immunization/methods , Mice , Rabies/immunology , Rabies/prevention & control , Vaccination/methodsABSTRACT
Iodinated disinfection byproducts (I-DBPs) are highly toxic, but few precursors of I-DBPs have been investigated. Tyrosine-containing biomolecules are ubiquitous in surface water. Here we investigated the formation of I-DBPs from the chloramination of seven tyrosyl dipeptides (tyrosylglycine, tyrosylalanine, tyrosylvaline, tyrosylhistidine, tyrosylglutamine, tyrosylglutamic acid, and tyrosylphenylalanine) in the presence of potassium iodide. High resolution mass spectrometry and tandem mass spectrometry (MS/MS) analyses of the benchtop reaction solutions found that all seven precursors formed both I- and Cl-substituted tyrosyl dipeptide products. Iodine substitutions occurred on the 3- and 3,5-positions of the tyrosyl-phenol ring while chlorine substituted on the free amino group. To reach the needed sensitivity to detect iodinated tyrosyl dipeptides in authentic waters, we developed a high performance liquid chromatography (HPLC)-MS/MS method with multiple reaction monitoring mode and solid phase extraction. HPLC-MS/MS analysis of tap and corresponding raw water samples, collected from three cities, identified four iodinated peptides, 3-I-/3,5-di-I-Tyr-Ala and 3-I-/3,5-di-I-Tyr-Gly, in the tap waters but not in the raw waters. The corresponding precursors, Tyr-Ala and Tyr-Gly, were also detected in the same tap and raw water samples. This study demonstrates that iodinated dipeptides exist as DBPs in drinking water.
Subject(s)
Disinfectants , Drinking Water , Water Pollutants, Chemical , Water Purification , Cities , Dipeptides , Disinfection , Halogenation , Tandem Mass SpectrometryABSTRACT
Ecological risk assessments for mixtures have attracted considerable attention. In this study, 38 pesticides in the real environment were taken as objects and their toxicities to different organisms from three trophic levels were employed to assess the ecological risk of the mixture. The first tier assessment was based on the CA effect and the obtained sum of risk quotients (SRQspecies-CA) were 3.06-9.22. The second tier assessment was based on non-CA effects and the calculated SRQspecies-TU are 5.37-9.29 using joint effects (TUsum) as modified coefficients, which is higher than SRQspecies-CA and indicates that ignoring joint effects might run the risk of underestimating the actual impact of pesticide mixtures. Due to the influences of synergistic and antagonistic effects, risk contribution of components to mixture risks based on non-CA effects are different from those based on the CA effect. Moreover, it was found that the top 8 dominating components explained 95.5%-99.8% of mixture risks in this study. The dominating components are similar in the two tiers for a given species. Accordingly, a novel two-tiered approach was proposed to assess the ecological risks of mixtures based on joint effects. This study provides new insights for ecological risk assessments with the consideration of joint effects of components in the pesticide mixtures.
Subject(s)
Ecology/methods , Environmental Pollutants/pharmacology , Pesticides/pharmacology , Risk Assessment , Animals , Chlorophyta/drug effects , Drug Combinations , Fishes , Invertebrates/drug effects , Pesticides/analysisABSTRACT
Hashing methods have been widely investigated for fast approximate nearest neighbor searching in large data sets. Most existing methods use binary vectors in lower dimensional spaces to represent data points that are usually real vectors of higher dimensionality. We divide the hashing process into two steps. Data points are first embedded in a low-dimensional space, and the global positioning system method is subsequently introduced but modified for binary embedding. We devise dataindependent and data-dependent methods to distribute the satellites at appropriate locations. Our methods are based on finding the tradeoff between the information losses in these two steps. Experiments show that our data-dependent method outperforms other methods in different-sized data sets from 100k to 10M. By incorporating the orthogonality of the code matrix, both our data-independent and data-dependent methods are particularly impressive in experiments on longer bits.
ABSTRACT
Rabies is an ancient disease but remains endemic in most parts of the world and causes approximately 59,000 deaths annually. The mechanism through which the causative agent, rabies virus (RABV), evades the host immune response and infects the host central nervous system (CNS) has not been completely elucidated thus far. Our previous studies have shown that lab-attenuated, but not wild-type (wt), RABV activates the innate immune response in the mouse and dog models. In this present study, we demonstrate that lab-attenuated RABV causes abortive infection in astrocytes, the most abundant glial cells in the CNS. Furthermore, we found that lab-attenuated RABV produces more double-stranded RNA (dsRNA) than wt RABV, which is recognized by retinoic acid-inducible gene I (RIG-I) or melanoma differentiation-associated protein 5 (MDA5). Activation of mitochondrial antiviral-signaling protein (MAVS), the common adaptor molecule for RIG-I and MDA5, results in the production of type I interferon (IFN) and the expression of hundreds of IFN-stimulated genes, which suppress RABV replication and spread in astrocytes. Notably, lab-attenuated RABV replicates in a manner identical to that of wt RABV in MAVS-/- astrocytes. It was also found that lab-attenuated, but not wt, RABV induces the expression of inflammatory cytokines via the MAVS- p38/NF-κB signaling pathway. These inflammatory cytokines increase the blood-brain barrier permeability and thus enable immune cells and antibodies infiltrate the CNS parenchyma, resulting in RABV control and elimination. In contrast, wt RABV restricts dsRNA production and thus evades innate recognition by RIG-I/MDA5 in astrocytes, which could be one of the mechanisms by which wt RABV evades the host immune response in resident CNS cells. Our findings suggest that astrocytes play a critical role in limiting the replication of lab-attenuated RABV in the CNS.
ABSTRACT
In the study, rubber accelerator tetrabenzylthiuramdisulfide (TBzDT) was synthesized with two-step method with hydrogen peroxide as oxidant firstly. TBzDT was detected and characterized with XRD, FT-IR, TG-DTA. Its micro-structure was revealed. Chemical bond types into TBzDT molecule were revealed with FT-IR. TBzDT phase composition and structure were given by crystallographic data from XRD detecting such as cell parameters, crystal face index. The phase composition and qualitative identification of TBzDT structure were completed. Two kinds of information were detected with TG-DTA as to quality change and thermal effect. TBzDT phase transition and decomposition temperature were 142.5, 200.9 â respectively. The decomposition temperature of TBzDT was relevtively high. It could provided reference with research on rubber vulcanizing properties by TBzDT on rubber vulcanizing machine. A little SC2 contained into TBzDT was revealed by FTIR, TG-DTA from different sides.
ABSTRACT
Blur in facial images significantly impedes the efficiency of recognition approaches. However, most existing blind deconvolution methods cannot generate satisfactory results due to their dependence on strong edges, which are sufficient in natural images but not in facial images. In this paper, we represent point spread functions (PSFs) by the linear combination of a set of pre-defined orthogonal PSFs, and similarly, an estimated intrinsic (EI) sharp face image is represented by the linear combination of a set of pre-defined orthogonal face images. In doing so, PSF and EI estimation is simplified to discovering two sets of linear combination coefficients, which are simultaneously found by our proposed coupled learning algorithm. To make our method robust to different types of blurry face images, we generate several candidate PSFs and EIs for a test image, and then, a non-blind deconvolution method is adopted to generate more EIs by those candidate PSFs. Finally, we deploy a blind image quality assessment metric to automatically select the optimal EI. Thorough experiments on the facial recognition technology database, extended Yale face database B, CMU pose, illumination, and expression (PIE) database, and face recognition grand challenge database version 2.0 demonstrate that the proposed approach effectively restores intrinsic sharp face images and, consequently, improves the performance of face recognition.