Background

Various antibody-like proteins (ALPs) have been obtained using an in vitro selection approach and employed in the biological and medical fields (Schumacher et al., 2018; Simeon and Chen 2018). During this in vitro selection, initial library construction is an important step in obtaining high-affinity binders. For example, to develop a nanobody, the VHH domains of camel antibodies obtained after immunization were used for the construction of an initial library (Arbabi Ghahroudi et al., 1997). Alternatively, the VHH domains from naïve genes or a semi-synthetic or synthetic library could be used as the initial library (Goldman et al., 2006; Monegal et al., 2009; Yan et al., 2014).

For other non-immunoglobulin scaffolds, synthetic libraries were used exclusively as the initial library (Nord et al., 1997; Koide et al., 1998; Beste et al., 1999; Binz et al., 2003; Grabulovski et al., 2007; Olson et al., 2008). Oligonucleotides containing random codons were assembled by extension PCR and ligated to other DNA fragments for synthesis of the DNA library. As randomized codons, NNK codons (with N representing A, C, G, or T; and K representing T or G) were generally used because they cover all 20 amino acids and can reduce the redundancy of codons and the frequency of stop codons. Alternatively, direct codon synthesis using trinucleotide phosphoramidites (Virnekäs et al., 1994) was used to mimic the amino acid frequency in the CDR3 sequences of antibodies (Zemlin et al., 2003; Wojcik et al., 2010; Koide et al., 2012).

The construction of medium-sized libraries (10¹⁰ diversity) is relatively simple, whereas that of large-sized libraries (10¹³ to 10¹⁴ diversity) requires special care to maintain the diversity (Olson and Roberts 2007). For example, when a 50 nM DNA product is obtained after 10 cycles of PCR, the concentration of the original template can be calculated as 50 pM. If we use only 100 µl reaction mixture, the maximum diversity that can be achieved using the conditions described above is limited to 3 × 10⁹.

Recently, by modifying the traditional mRNA display (Nemoto et al., 1997; Roberts and Szostak 1997) and the original version of the TRAP display (Ishizawa et al., 2013; Kawakami et al., 2013), we developed an improved TRAP display for the quick in vitro selection of ALPs from a large library (>10¹³ theoretical diversity), from which we selected high-affinity monobodies (sub-nM K_D) against the SARS-CoV-2 spike protein (Kondo et al., 2020). In the present article, we describe the preparation of a monobody mRNA library with >10¹³ diversity (Figures 1 and 2). We first divided the monobody gene into two fragments, one containing 8 or 10 random codons in the BC loop sequence (A-fragment) and the other containing 10 or 12 random codons in the FG loop sequence (B-fragment). As a random codon, we used a codon mix with the following ratios: 20% Tyr, 10% Ser, 15% Gly, 10% Trp, and 3% each of all remaining amino acids, with the exception of Cys, which was similar to the original cocktail (30% Tyr, 15% Ser, 10% Gly, 5% Phe, 5% Trp, and 2.5% each of all remaining amino acids, with the exception of Cys) (Wojcik et al., 2010; Koide et al., 2012). To prepare a template DNA for each of the two fragments, we ligated 2 to 3 oligonucleotides assembled by guide DNAs using the T4 DNA ligase. We added a modification at the 3′ end of the guide DNAs to prevent them from becoming a primer in the following amplification step. After amplifying the product DNAs by large-scale PCR with minimum amplification cycles (15 ml, 7 cycles), these DNA fragments were digested with BsaI and ligated using T4 DNA ligase. After amplification of the DNA product by large-scale PCR (60 ml, 4 cycles), mRNAs were synthesized by T7 RNA polymerase. This procedure would be useful for the preparation of high-diversity mRNA libraries for various ALPs in the future.

Figure 1. Monobody library sequences. A. DNA sequence of the monobody library [T7 promoter, green; codon mix (20% Tyr, 10% Ser, 15% Gly, 10% Trp, and 3% each of all remaining amino acids, with the exception of Cys), yellow; an21 sequence (the 21-mer sequence for annealing to HEX-PuL), cyan]. B. Protein sequence of the monobody library (randomized residues X, yellow).

Figure 2. Scheme used for preparation of the monobody mRNA library.