Developing a SARS-CoV-2 main protease binding prediction random forest model for drug repurposing for COVID-19 treatment

The coronavirus disease 2019 (COVID-19) global pandemic resulted in millions of people becoming infected with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus and close to seven million deaths worldwide. It is essential to further explore and design effective COVID-19 treatment drugs that target the main protease of SARS-CoV-2, a major target for COVID-19 drugs. In this study, machine learning was applied for predicting the SARS-CoV-2 main protease binding of Food and Drug Administration (FDA)-approved drugs to assist in the identification of potential repurposing candidates for COVID-19 treatment. Ligands bound to the SARS-CoV-2 main protease in the Protein Data Bank and compounds experimentally tested in SARS-CoV-2 main protease binding assays in the literature were curated. These chemicals were divided into training (516 chemicals) and testing (360 chemicals) data sets. To identify SARS-CoV-2 main protease binders as potential candidates for repurposing to treat COVID-19, 1188 FDA-approved drugs from the Liver Toxicity Knowledge Base were obtained. A random forest algorithm was used for constructing predictive models based on molecular descriptors calculated using Mold2 software. Model performance was evaluated using 100 iterations of fivefold cross-validations which resulted in 78.8% balanced accuracy. The random forest model that was constructed from the whole training dataset was used to predict SARS-CoV-2 main protease binding on the testing set and the FDA-approved drugs. Model applicability domain and prediction confidence on drugs predicted as the main protease binders discovered 10 FDA-approved drugs as potential candidates for repurposing to treat COVID-19. Our results demonstrate that machine learning is an efficient method for drug repurposing and, thus, may accelerate drug development targeting SARS-CoV-2.

Serum Derived Extracellular Vesicles Mediated Delivery of Synthetic miRNAs in Human Endothelial Cells

Extracellular vesicles (EVs) have emerged in the last decades as a cell-to-cell communication mechanism. One of their mechanism of action is the direct delivery of their cargo, composed of bioactive molecules to target cells. Different methods (direct electroporation, cell transfection, chemical transfection) were developed to vehicle therapeutic molecules through EVs. However, most of these techniques presented some limitations such as EV disruption and aggregation. In the present study, we demonstrated that a direct temperature-controlled co-incubation of EVs with defined miRNAs is a stable method to deliver information to target cells without affecting EV constitutive content. We chose serum as an easy and abundant source of EVs applicable to autologous treatment after EV modification. Exogenous cel-miR-39 loaded on serum EVs (SEVs) was taken up by human endothelial cells, demonstrating an adequate miRNA loading efficacy based on the co-incubation method. Moreover, SEVs co-incubation with the angiomiRNA-126 (miR-126) enhanced their angiogenic properties in vitro and in vivo by increasing the capacity to induce capillary-like structure formation of human endothelial cells. MiR-126 loaded EVs were also shown to stimulate mouse endothelial cells to invade Matrigel plugs and create more vessels with respect to the EV naive counterpart. When SEVs were loaded with miR-19b, an anti-angiogenic miRNA, they were able to reduce Vascular endothelial growth factors (VEGF) pro-angiogenic capacity, supporting the selective biological effect mediated by the carried miRNA. Lastly, we identified Annexin A2 (ANXA2) as one of the molecules involved in the exogenous RNA binding to serum EV surface, favoring miRNA delivery to target endothelial cells for potential therapeutic application.

Improved anti-cancer effect of epidermal growth factor-gold nanoparticle conjugates by protein orientation through site-specific mutagenesis

Epidermal growth factor (EGF)-nanoparticle conjugates have the potential for cancer therapeutics due to the unique cytotoxic activity in cancer cells with EGF receptor (EGFR) overexpression. To gain its maximum activity, the EGF molecule should be immobilized on the nanoparticle surface in a defined orientation so as the bulky nanoparticle will not interfere EGF-EGFR interaction. Herein, we demonstrate successful enhancement of the anti-cancer activity of EGF-gold nanoparticle conjugates (EGF-GNPs) by controlling the EGF orientation on the surface of the nanoparticle through site-specific mutagenesis. Three lysine-free EGF variants (RR, RS, and SR) were designed, where two endogenous lysine residues were replaced with either arginine (R) or serine (S). The EGF mutants can be conjugated to the GNPs in a controlled orientation through the single amino group at the N-terminus. The ability of the mutants to induce extracellular signal-regulated kinase (ERK) phosphorylation was no different from wild type EGF (WT) in soluble form, rather lowered for one mutant (RR). However, after conjugated to GNPs, the SR mutants exhibited an enhanced biological activity than WT, in terms of ERK phosphorylation and growth inhibition of cancer cells. Further analysis of the binding constant of each mutant indicated the emergent enhanced activity of the GNP conjugates of the SR mutant was not solely contributed to the orientation, but to its higher binding activity to EGFR. These results validate the present genetic recombination strategy to improve the anticancer efficiency of EGF-GNPs.

The Influence of Moisture Content and Temperature on the Long-Term Storage Stability of Freeze-Dried High Concentration Immunoglobulin G (IgG)

High protein concentration products for targeted therapeutic use are often freeze-dried to enhance stability. The long-term storage stability of freeze-dried (FD) plasma-derived Immunoglobulin G (IgG) from moderate to high concentrations (10-200 mg/mL) was assessed. Monomer content, binding activity and reconstitution times were evaluated over a 12-month period under accelerated and real-term storage conditions. In the first case study it was shown that FD IgG from 10 to 200 mg/mL had minimal monomer/activity losses at up to ambient temperature after 12 months of storage. However, at 45 °C the sucrose-to-protein ratio played a significant impact on IgG stability above 50 mg/mL. All IgG concentrations witnessed moisture ingress over a 12-month period. The impact of moisture ingress from environmental exposure (between 0.1% and 5% w/w moisture) for IgG 50 mg/mL was assessed, being generated by exposing low moisture batches to an atmospheric environment for fixed time periods. Results showed that at -20 °C and 20 °C there was no significant difference in terms of monomer or antigen-binding activity losses over 6 months. However, at 45 °C, there were losses in monomer content, seemingly worse for higher moisture content samples although model binding activity indicated no losses. Finally, the difference between a low moisture product (0.1-1% w/w) and a moderately high moisture (3% w/w) product generated by alternative freeze-drying cycles, both stoppered under low oxygen headspace conditions, was evaluated. Results showed that at -20 °C and 20 °C there was no difference in terms of binding activity or monomer content. However, at 45 °C, the low moisture samples had greater monomer and binding activity losses than samples from the highest moisture cycle batch, indicating that over-drying can be an issue.

A Comprehensive Cheminformatics Analysis of Structural Features Affecting the Binding Activity of Fullerene Derivatives

Nanostructures like fullerene derivatives (FDs) belong to a new family of nano-sized organic compounds. Fullerenes have found a widespread application in material science, pharmaceutical, biomedical, and medical fields. This fact caused the importance of the study of pharmacological as well as toxicological properties of this relatively new family of chemicals. In this work, a large set of 169 FDs and their binding activity to 1117 proteins was investigated. The structure-based descriptors widely used in drug design (so-called drug-like descriptors) were applied to understand cheminformatics characteristics related to the binding activity of fullerene nanostructures. Investigation of applied descriptors demonstrated that polarizability, topological diameter, and rotatable bonds play the most significant role in the binding activity of FDs. Various cheminformatics methods, including the counter propagation artificial neural network (CPANN) and Kohonen network as visualization tool, were applied. The results of this study can be applied to compose the priority list for testing in risk assessment related to the toxicological properties of FDs. The pharmacologist can filter the data from the heat map to view all possible side effects for selected FDs.

Construction and Characterization of an Anti-Hepatitis B Virus preS1 Humanized Antibody that Binds to the Essential Receptor Binding Site

Hepatitis B virus (HBV) is a major cause of liver cirrhosis and hepatocellular carcinoma. With recent identification of HBV receptor, inhibition of virus entry has become a promising concept in the development of new antiviral drugs. To date, 10 HBV genotypes (A-J) have been defined. We previously generated two murine anti-preS1 monoclonal antibodies (mAbs), KR359 and KR127, that recognize amino acids (aa) 19-26 and 37-45, respectively, in the receptor binding site (aa 13-58, genotype C). Each mAb exhibited virus neutralizing activity in vitro, and a humanized version of KR127 effectively neutralized HBV infection in chimpanzees. In the present study, we constructed a humanized version (HzKR359-1) of KR359 whose antigen binding activity is 4.4-fold higher than that of KR359, as assessed by competitive ELISA, and produced recombinant preS1 antigens (aa 1-60) of different genotypes to investigate the binding capacities of HzKR359-1 and a humanized version (HzKR127-3.2) of KR127 to the 10 HBV genotypes. The results indicate that HzKR359-1 can bind to five genotypes (A, B, C, H, and J), and HzKR127-3.2 can also bind to five genotypes (A, C, D, G, and I). The combination of these two antibodies can bind to eight genotypes (A-D, G-J), and to genotype C additively. Considering that genotypes A-D are common, whereas genotypes E and F are occasionally represented in small patient population, the combination of these two antibodies might block the entry of most virus genotypes and thus broadly neutralize HBV infection.

The Divergence of Flowering Time Modulated by FT/TFL1 Is Independent to Their Interaction and Binding Activities

FLOWERING LOCUS T (FT) and TERMINAL FLOWER1 (TFL1) proteins share highly conserved amino acid residues but they play opposite regulatory roles in promoting and repressing the flowering response, respectively. Previous substitution models and functional analysis have identified several key amino acid residues which are critical for the promotion of flowering. However, the precise relationship between naturally occurring FT/TFL1 homologs and the mechanism of their role in flowering is still unclear. In this study, FT/TFL1 homologs from eight Rosaceae species, namely, Spiraea cantoniensis, Pyracantha fortuneana, Photinia serrulata, Fragaria ananassa, Rosa hybrida, Prunus mume, Prunus persica and Prunus yedoensis, were isolated. Three of these homologs were further characterized by functional analyses involving site-directed mutagenesis. The results showed that these FT/TFL1 homologs might have diverse functions despite sharing a high similarity of sequences or crystal structures. Functional analyses were conducted for the key FT amino acids, Tyr-85 and Gln-140. It revealed that TFL1 homologs cannot promote flowering simply by substitution with key FT amino acid residues. Mutations of the IYN triplet motif within segment C of exon 4 can prevent the FT homolog from promoting the flowering. Furthermore, physical interaction of FT homologous or mutated proteins with the transcription factor FD, together with their lipid-binding properties analysis, showed that it was not sufficient to trigger flowering. Thus, our findings revealed that the divergence of flowering time modulating by FT/TFL1 homologs is independent to interaction and binding activities.

Production and evaluation of a recombinant subunit vaccine against botulinum neurotoxin serotype B using a 293E expression system

Although Escherichia coli and yeast were commonly used to express recombinant Hc of botulinum neurotoxins, as an alternative, in current study, a 293E expression system was used to express the Hc of botulinum neurotoxin serotype B (BHc) as soluble recombinant protein for experimental vaccine evaluation. Our results demonstrated that the 293E expression system could produce high level of recombinant secreted BHc protein, which was immunorecognized specifically by anti-botulinum neurotoxin serotype B (BoNT/B) sera and showed ganglioside binding activities. The serological response and efficacy of recombinant BHc formulated with aluminum hydroxide adjuvant were evaluated in mice. Immunization with Alhydrogel-formulated BHc subunit vaccine afforded the effective protection against BoNT/B challenge. A frequency- and dose-dependent effect to immunization with BHc subunit vaccine was observed and the ELISA antibody titers correlated well with neutralizing antibody titers and protection. And a solid-phase assay showed that the neutralizing antibodies from the BHc-immunized mice inhibited the binding of BHc to the ganglioside GT1b. Our results also show that the plasmid pABE293SBHc derived of the 293E expression system as DNA vaccine is capable of inducing stronger humoral response and protective efficacy against BoNT/B than the pVAX1SBHc. In summary, immunization with the 293E-expressed BHc protein generates effective immune protection against BoNT/B as E. coli or yeast-expressed BHc, so the efficient expression of botulinum Hc protein for experimental vaccine can be prepared using the 293E expression system.