Figures and data

A. Illustrations of types of pi interactions in which Phenylalanine can participate. B. Electrostatic views of benzene (as a structural surrogate for Phenylalanine) and fluorinated analogs. Coloring is to convention of Red = negative, Blue = positive.

Strategy and identification of synthetases. A. Chemical structure of para-methyl tetra-fluoro Phe, which was used for screening in a PylRs library in E. coli. B. Heatmap of synthetases hits that resulted from positive and negative screening in terms of their relative GFP N150TAG rescue (fluorescence) in the presence of para-methyl tetrafluoro Phenylalanine. C. Unnatural Protein 50 (UP50) curves for the top performing synthetases from the screens.

Permissivity of the identified synthetases. A. Chemical structures of the fluorinated phenylalanine residues used in the screen. B. Green fluorescence from rescue of sfGFP-N150TAG in E.coli using 1 mM of each ncAA after 24-hour incubation.

Effects of fluorination on interaction energies with cations

ESI mass spectrometry confirmation of sitespecific encoding of fluorinated Phe residues within superfolder GFP in E coli. sfGFP was purified via His tag from 50 mL growths. Table at top right: mass differences of sfGFP N150TAG from WT GFP conform to prediction for the molecular mass of each fluoro Phe species. In some conditions for B5 synthetase, a small fraction of mass consistent with Phe incorporation was observed (asterisks). Mass spectrometry was not pursued for the combination of 2F Phe and D6 due to low expression in the screen.

Function of Phex synthetases in HEKT cells. A. Micrographs of HEKT cells co-transfected with GFP_N150TAG and Phex.D6in the presence or absence of the given amino acid. GFP (top) and brightfield (bottom). In all cases, images taken approximately 24 hours after transfection and unnatural amino acid used at 2 mM concentration. B. Lysate GFP fluorescence for a range of fluorinated phenylalanine residues.

ESI mass spec confirmation of encoding of fluorinated Phe analogs within sfGFP expressed in HEKT cells. ncAAs were added at 2 mM concentration and cells were harvested approximately 36 hours post transfection. sfGFP-WT-V5-HIS or N150_TAG-V5-HIS was purified via HIS tag. Table (at top right) reports mass differences of sfGFP N150TAG from WT GFP, which conform to prediction for the molecular mass of each fluoro Phe species.

Site-specific encoding of tri-fluoro-Phe within two large, human ion channels. A. Western blot of CFTR using C terminal antibody (ab596) of WT CFTR or F508TAG-CFTR cotransfected with PheX-D6 in the presence or absence of 2 mM 2,3,6 F Phe. WT lysates are loaded at 1:10 of TAG conditions B. Western blot of Nav 1.5 using C terminal antibody (D9J7S) of WT hNav 1.5 or F1486TAG-hNav 1.5 co-transfected with Phex.D6 in the presence or absence of 2 mM 2,3,6 F Phe. C. Quantification (densitometry) of signal from 60 jig of loaded lysate. D. Mass spectra of tryptic fragment from expressed and purified hNav 1.5 F1486(2,3,6F Phe) via D6. E. Example current families for whole (HEK) cells expressing WT or mutant channels. Cells held at -140 mV and activated via depolarization in 5 mV increments. F. Activation (GA/) curves for WT and mutant channels. G. Steady state inactivation curves for WT and mutant channels.

Large scale prep and yield of sfGFP bearing penta-F phenylalanine at position N150 using Phex.D6-Purification was started from 3.26 g of HEK cells (31 10 cm dishes) A. SDS page gel imaged for GFP before staining L= Ladder, IN= input, i.e., diluted lysate, FT= flowthrough, W1, W2, W3 = washes, X= empty, E = elution (diluted 1:1 in wash buffer). 30 µL of sample loaded per lane from 100 mL (IN/FT), 10 mL (Washes), 1.5 mL (E). Thus, ∼1/50 of the elution was run in lane E. B. Coomassie stain of the same gel. Buffer exchange and concentration of elution resulted in 240 µL of 1.041 abs. Factoring the extinction coefficient of sfGFP_V5_HIS and a purity estimate (68%) from densitometry of the bands in Coomassie of E yields a total sfGFP estimate of 110.4 µg, or 34 µg/g of cells (wet pellet weight).

Deconvoluted ESI mass spectra for sfGFP-N150TAG expressed with 2-mono fluoro phenylalanine and PheX-B5. Both 2 mM and 0.2 mM concentration of amino acid resulted in extensive incorporation of the amino acid at natural Phe codons, as indicated by masses at multiples of +18 Da. Signal at expected mass is absent in the 2 mM condition and a very small fraction (∼1%) in 0.2 mM condition (black arrowheads).

GFP lysate values from HEKT cells co transfected with sfGFP150TAG and either Phex.D6 (left) or Phex.B5 (right) and harvested approximately 24 hours post transfection. Error bars are ± STD.

UV absorbance of Liquid Chromatography of purified sfGFP samples. Both WT (left) and mutants (in this case D6 encoding 2,3,5,6 tetra-F Phe) eluted as 3 typical peaks for sfGFP, a dominant peak, as well as one at ∼ -42 Da and a third at ∼ + 532 Da. The deconvoluted mass spectra for the dominant (center) peaks are shown in the main figures.

Protein sequences of sfGFP variants used for expression and purification in this study.

Gating Parameters for hNav 1.5 variants. Bolded = p value of 0.0014 compared to WT. P>0.24 for all other comparisons.