Nat Methods 2015 Yu

NATURE METHODS | ADVANCE ONLINE PUBLICATION | 1 is unlikely to misfold or aggregate in cells12. We hypothesized that some of the many BphP sequences in protein sequence databases might be monomeric in the BphP form, which is likely to occur if the protein lacks strong hydrophobic interactions at the putative dimer interface. In contrast, in IFP1.4’s parent, DrBphP13, the dimer interface includes several residues (Fig. 1a): Leu311 appears to play a critical role, as the mutation L311K disrupts the dimer interface3. Analysis of ~40 BphP sequences from the NCBI database revealed BrBphP (from Bradyrhizobium) as a potential candidate, as the residue corresponding to Leu311 in DrBphP is a polar threonine (Fig. 1b). Indeed, size-exclusion chromatography indicated that BrBphP eluted later than dimeric DrBphP and at a time similar to that of the monomeric form of IFP1.4 (Supplementary Fig. 2), suggesting that BrBphP is a monomer. We engineered the nonfluorescent BrBphP into a fluorescent mutant. In brief, we selected several residues (Asp199, Tyr168, Val178 and Asn258) surrounding BV for saturation mutagenesis, which was followed by DNA shuffling14 and random mutagenesis. The final fluorescent mutant mIFP absorbed maximally at 683 nm (Supplementary Fig. 3), with excitation and emission maxima of 683 and 704 nm, respectively (Fig. 1c), a quantum yield of 8% and an extinction coefficient of 82,000 M−1 cm−1 (Supplementary Table 1). We confirmed that mIFP was monomeric at high concentrations (17 and 34 M) (Fig. 1d and Supplementary Fig. 4). It contains 19 mutations (Supplementary Figs. 5 and 6), including 5 near the D-ring of BV that likely limit its rotation, contributing to the engineered fluorescence by increasing radiative decay of the excited state (Fig. 1e). Mutated residues in mIFP, IFP1.4 and iRFP do not overlap and thus might be targeted for further engineering (Supplementary Fig. 7). mIFP was stable in pH 4–10 (Fig. 1f). Its molecular brightness was similar to that of IFP1.4 and iRFP (Supplementary Table 1), and its cellular brightness was similar to that of iRFP and tenfold greater than that of IFP1.4 in live HeLa cells (Fig. 1g,h and Supplementary Fig. 8). mIFP was 6.3 times more photostable than IFP1.4 (Supplementary Fig. 9) but approximately one-fifth as photostable as iRFP in HEK293 cells (Supplementary Table 1). Photobleaching of mIFP was irreversible (Supplementary Fig. 10), suggesting no residual photoisomerization. mIFP was similar to IFP2.0 and iRFP in terms of maturation rate, BV binding kinetics and affinity (Supplementary Figs. 11–13). To demonstrate mIFP as a protein tag for use in live-cell imaging, we constructed ~30 mIFP fusion proteins, targeting both the N and C termini with an appropriate-length linker (Online Methods). We successfully expressed and imaged the mIFP fusions in cultured cells without addition of the cofactor, which suggests that mIFP A naturally monomeric infrared fluorescent protein for protein labeling in vivo