Protein evolution with an expanded genetic code

Abstract

We have devised a phage display system in which an expanded genetic code is available for directed evolution. This system allows selection to yield proteins containing unnatural amino acids should such sequences functionally outperform ones containing only the 20 canonical amino acids. We have optimized this system for use with several unnatural amino acids and provide a demonstration of its utility through the selection of anti-gp120 antibodies. One such phage-displayed antibody, selected from a naïve germline scFv antibody library in which six residues in V(H) CDR3 were randomized, contains sulfotyrosine and binds gp120 more effectively than a similarly displayed known sulfated antibody isolated from human serum. These experiments suggest that an expanded "synthetic" genetic code can confer a selective advantage in the directed evolution of proteins with specific properties.

Fig. 1.

Phagemid-display of unnatural amino acids. (A) Structure of sulfotyrosine. (B) Structure of para-acetyl-phenylalanine. (C) Structure of bipyridyl-alanine. (D) Structure of 4-borono-phenylalanine. (E) Yield of phage under optimized conditions where +UAA corresponds to phage produced with the corresponding unnatural amino acid (UAA) supplemented in the media at optimized concentrations and −UAA corresponds to phage produced in the absence of the corresponding unnatural amino acid. Titers were determined in triplicate and error bars correspond to plus standard deviations. (F) Detection of pIII-scFv fusion from whole phage produced using pSEX-GermTAG in the presence of unnatural amino acids. Approximaely 10⁸ phage particles (corresponding to ≈200-μl phage culture precipitated and concentrated to ≈10 μl) per sample were run on a denaturing PAGE gel under reducing conditions and subsequently transferred to a membrane for Western blotting with an anti-pIII antibody. When the same volume of phage similarly produced and prepared from pSEX-GermTAG in the absence of unnatural amino acid was run and Western blotted for pIII, no bands were detected (data not shown). Hyperphage corresponds to phage without displayed scFv. Control corresponds to phage displaying the pIII-scFv fusion from pSEX-GermTAT with only common amino acids; the 51 kDa band is a result of nonspecific proteolysis of the 62-kDa band (pIII-scFv fusion).

Fig. 2.

Phage ELISA for gp120-binding with 412d-2SY selected from a doped 412d library compared with 412d-Y, where sulfotyrosines were replaced by tyrosines. For each sample, 0.33-μg gp120 was coated onto a Maxisorp plate, blocked with 2% milk, and bound with the respective phage. Phage was detected with an anti-M13 antibody. BSA-binding was used as a control to show specific gp120-binding. “Mock” refers to signal generated without any phage added.

Fig. 3.

Phage ELISA for gp120-binding with 66CC8-SY, 66CC8-Y, and 66CC14. For each sample, 0.3-μg gp120 was coated onto a microtiter plate well, blocked with 2% milk, and bound with the respective phage. Phage was detected with an anti-M13 antibody. BSA-binding was used as a control to show specific gp120-binding. (A) ELISA for a representative gp120-binding experiment done with two phage concentrations. (B) Average ELISA signals representing four separate experiments done with two separate phage preparations. Each experiment used the same amount of phage across all samples. Averages were calculated from signals normalized to 412d-2SYs binding within the same experiment. Error bars represent plus standard deviations. We note that this consolidated graph exaggerates variation because the separate ELISA experiments use different phage concentrations, and represent different incubation, washing, development, and detection times.