Development of novel detection system for sweet potato leaf curl virus using recombinant scFv

Abstract

Sweet potato leaf curl virus (SpLcV) causes yield losses in sweet potato cultivation. Diagnostic techniques such as serological detection have been developed because these plant viruses are difficult to treat. Serological assays have been used extensively with recombinant antibodies such as whole immunoglobulin or single-chain variable fragments (scfv). An scfv consists of variable heavy (V H) and variable light (V L) chains joined with a short, flexible peptide linker. An scFv can serve as a diagnostic application using various combinations of variable chains. Two SPLCV-specific scFv clones, F7 and G7, were screened by bio-panning process with a yeast cell which expressed coat protein (cp) of SpLcV. the scfv genes were subcloned and expressed in Escherichia coli. The binding affinity and characteristics of the expressed proteins were confirmed by enzyme-linked immunosorbent assay using SPLCV-infected plant leaves. Virus-specific scFv selection by a combination of yeast-surface display and scFv-phage display can be applied to detection of any virus.

Enhancing neutralizing activity against influenza H1N1/PR8 by engineering a single-domain VL-M2 specific into a bivalent form

“…The DNA fragment was amplified using PCR and cloned between the NheI and BamHI (NEB, USA) sites of a pCTCON plasmid (ampicillin-resistant) for YSD (M2::pCTCON). The M2::pCTCON plasmid was transformed into Saccharomyces cerevisiae EBY100 competent cells, prepared according to the Clontech manual (Clontech, Japan) as previously described [ 36 ]. Briefly, EBY100 yeast was freshly prepared in yeast-peptone-glucose media at 0.4 OD 600 and made competent by chemical solution (1 M Sorbitol/1 mM CaCl 2 , 0.1 M LiAc/10 mM DTT).…”

“…Phage titers were determined using kanamycin-resistant colony forming units (CFUs) in XL-1 blue. Bio-panning methods were applied to screen M2-specific phage scFvs (M2::scFv), as previously described [ 36 , 37 ]. Briefly, positive yeast (M2::YSD), blank (PBS), and negative samples (empty yeast, EBY100) diluted in PBS at an OD 600 of 0.6 were coated onto a 96-well Maxibinding immunoplate (SPL Life Sciences, Republic of Korea) and incubated overnight in a humid box at 4°C.…”

Structure‐Based Drug Discovery with Deep Learning**

“…[144] 3D grid-based approaches have also been used for lead optimization by predicting relative binding free energies linked to small modifications of ligand structures. [145,146] More recent approaches have replaced gridbased representations with graphs, [49,97,98,147] allowing to explicitly represent atom neighborhoods and connectivity, and apply roto-translational invariant graph neural networks for binding affinity prediction.…”

“…Predict the interaction between one or more proteins and one or more ligands. Amino-acid sequence [89][90][91][92][93][94] 3D structure [47,[95][96][97][98]…”

Rotational Variance‐Based Data Augmentation in 3D Graph Convolutional Network

“…Although the three invariances are innately captured by visual processing, such as pattern recognition, [9,10] the requirement of rotational invariance in DL models is not obvious in the chemistry domain. For example, the binding affinity of a ligand to a target protein is not rotation‐invariant but critically rotation‐variant, requiring the prediction of optimized ligand orientation to the protein in the virtual screening [11–17] . In this respect, we have proposed the use of the coordinate of a node in for incorporating the 3D bond information to a spectral GNN model of 3D graph convolutional network (3DGCN), and 3DGCN shows the capability for recognizing the ligand orientations to a target protein by rotational variance to some extent (Figure S1) [18,19] .…”