Next Generation Sequencing As Assisting Tool In Protein Engineering And The Development Of A Hierarchical Dual-Index Barcoding Strategy For Screening Binder Proteins By Next Generation Sequencing

Abstract

Combinatorial gene libraries are collections of randomized DNA sequences which code, for e.g., antigen binding fragments of antibodies (Fabs). These libraries are typically enriched with molecular display methods, e.g. phage display. After panning, binders are cloned into bacteria for expression and screening of individual binders. Clones that provide highest signals in binding assays are sequenced with Sanger sequencing, revealing connections between genotype and function to develop improved binders. To deepen the knowledge in designing better binders, more data is needed from a vast variety of clones. In this study, a high-throughput screening method utilizing Illumina Next Generation Sequencing (NGS) was developed. Unlike Sanger, NGS offers a feasible, cost-efficient way to combine functional and sequencing data from almost 10 000 binders in parallel.

At first, custom primers were designed containing the hybridizing, indexing and Illumina TruSeq compatible overhang sequences for barcoding well coordinates. The indexing scheme was first tested with purified plasmid DNA containing known Fab sequence and later directly from transformed bacteria colonies using electrophoresis and Sanger sequencing. In addition, DNA gel analysis was done from four PCR plates used in actual screening. The production and screening of binders was done in 96-well format. Fab libraries were cloned into pAK400 vector, transformed into XL1-Blue E. coli by electroporation and grown on agar plates. Single colonies were picked for PCR and expression. After indexing, the amplified DNA was pooled from the wells and amplified with second set of primers to barcode the plate, allowing final pooling of samples.

At the time of writing this thesis, a total of 3 840 Fabs were screened using the developed method. The results from both DNA gel analysis and Sanger sequencing indicated that the indexing works as intended. Well barcoding PCR was highly efficient, as a DNA band with correct size was observed in 96.4% of screened wells. Potent anti-SpyCatcher Fabs were discovered from the screened libraries, two of them with observed KD of 1.9 and 30.3 nM. Skatole proved to be a difficult target, as the screened Fabs were found binding only to biotinylated skatole, in conjugation with streptavidin. More screening will be done to reach the goal of 7 680 indexed and functionally analyzed Fab clones.