Deep mutational scanning (DMS) makes it possible to perform massively parallel quantification of the relationship between genetic variants and phenotypes of interest. However, the difficulties in introducing large variant libraries into mammalian cells greatly hinder DMS under physiological states. Here we developed two novel strategies for DMS library construction in mammalian cells, namely ‘piggyBac- in-vitro ligation’ and ‘piggyBac- in-vitro ligation-PCR’. For the first strategy, we took the ‘ in-vitro ligation’ approach to prepare high-diversity linear dsDNAs, and integrate them into the mammalian genome with a piggyBac transposon system. For the second strategy, we further added a PCR step using the in-vitro ligation dsDNAs as templates, for the construction of high-content genome-integrated libraries via large-scale transfection. Both strategies could successfully establish genome-integrated EGFP-chromophore randomized libraries in HEK293T cells and enrich the green fluorescence-chromophore amino acid sequences. And we further identified a novel transcriptional activator peptide with the ‘piggyBac- in-vitro ligation-PCR’ strategy. Our novel strategies greatly facilitate the construction of large variant DMS library in mammalian cells, and may have great application potential in the future.

As central players in cellular structure and function, proteins have long been central themes in life science research. Analyzing the impact of protein sequence variation on its structure and function is one of the important means to study proteins. In recent years, a technology called deep mutational scanning (DMS) has been widely used in the field of protein research. It introduces thousands of mutations in parallel in specific regions of proteins through high-abundance DNA libraries. After screening, high-throughput sequencing is employed to score each mutation, revealing sequence-function correlations. Due to its high-throughput, fast and easy, and labor-saving features, DMS has become an important method for protein function research and protein engineering. This review briefly summarizes the principle of DMS technology, highlighting its applications in mammalian cells. Moreover, this review analyzes the current technical bottlenecks, aiming to facilitate relevant research.

Objective:A linear-double-stranded DNA (ldsDNA) based AND-gate strategy was developed to construct nanobody library in mammalian cells.Measures:We employed the ldsDNA-based AND-gate genetic circuit to introduce nanobody library into cultured mammalian cells. The sequence complexity of the complementary determining regions (CDRs) was introduced into the up- and down-stream ldsDNA by PCR amplification, respectively. After input ldsDNAs being co-transfected into HEK293T cells for 48h, RNA was extracted then cDNA was synthesized. PCR was employed to amplify the library nanobody sequences. High-throughput sequencing (HTS) was employed to analyze the library nanobody sequences.Results:We combined the clean merged paired-end reads from three biological repeats and got 4, 173, 356 reads. About 88.18% of the merged reads contain both upstream- and downstream-ldsDNA sequences. The most abundant read length is 264-bp, which corresponds to the intact sequence length. A total of 22, 172 unique nanobody sequences were identified by high-throughput sequencing. Moreover, the library CDR sequences followed the NNK degeneracy.Conclusion:We developed a novel ldsDNA-based AND gate genetic circuit to construct nanobody library in mammalian cells.

DNA-encoded peptide/protein libraries are the startingpoint forprotein evolutionary modification and functional peptide/antibodyselection. Different display technologies, protein directed evolution,and deep mutational scanning (DMS) experiments employ DNA-encodedlibraries to provide sequence variations for downstream affinity-or function-based selections. Mammalian cells promise the inherentpost-translational modification and near-to-natural conformation ofexogenously expressed mammalian proteins and thus are the best platformfor studying transmembrane proteins or human disease-related proteins.However, due to the current technical bottlenecks of constructingmammalian cell-based large size DNA-encoded libraries, the advantagesof mammalian cells as screening platforms have not been fully exploited.In this review, we summarize the current efforts in constructing DNA-encodedlibraries in mammalian cells and the existing applications of theselibraries in different fields.

While different display technologies, represented by phage display, have been widely used in drug discovery, they still can hardly achieve function-based peptide screening, which in most cases is performed in mammalian cells. And most attempts to screen functional peptides with mammalian platforms utilized plasmids to store coding information. Our previous work established double-stranded DNAs (dsDNAs) as innovative biological parts to implement AND-gate genetic circuits in mammalian cells. In the current study, we employ dsDNAs with terminal NNK degenerate codons to implement AND-gate genetic circuits and generate peptide libraries in mammalian cells. This dsDNA-based AND-gate (DBAG) peptide library construction strategy is easy to perform, requiring only PCR reaction and cell transfection. High-throughput sequencing (HTS) and single-cell sequencing results revealed both peptide length and amino acid sequence diversity of DBAG peptide libraries. Moreover, as a feasibility test of this strategy, we identified an MDM2-interacting peptide by applying the DBAG peptide library to a mammalian cell-based two-hybrid system. Our work establishes dsDNAs with terminal degenerate codons as biological parts to build peptide libraries in mammalian cells, which may have great application potential in the future.