Fig. 1: PhastID is fast and efficient for proximity labeling.

a Domain organization of naturally found (in red) and engineered BPLs examined in this study. BirA*, 35.2 kDa; BASU,¹²⁸ 28.8 kDa; CrBPL, 25.1 kDa; BioID2, 26.6 kDa; MfBPL, 27.4 kDa; PhBPL, 26.5 kDa; PkBPL, 26.3 kDa. TurboID, evolved from EcBirA with multiple mutations. MiniTurbo,⁹² truncated TurboID with mutations. AirID,¹²⁹ de novo designed. b, c HEK293T cells transiently expressing the indicated HA-Flag-tagged biotin ligases were cultured in the presence of 50 μM biotin for 16 h, and collected for western blot analysis (b). SA, Streptavidin Protein DyLight™ 680. GAPDH served as a loading control. Signals from each lane were quantified by ImageJ and presented as mean ± SEM (n = 3) (c). Total streptavidin signals from each lane were divided by the corresponding Flag signals and then normalized to the BioID2 group. Statistical significance was determined using the two-tailed Student’s t-test. *P < 0.05. d HeLa cells stably expressing HA-tagged TRF1 fused to different biotin ligases were cultured with 50 μM biotin for 16 h, and then subjected to immunofluorescence analysis with an anti-HA antibody (purple), a telomere probe (TTAGGGTTAGGGTTAGGG) (red), and FITC-conjugated streptavidin (green). Vector alone served as a negative control. DAPI was used to mark the nuclei. Scale bars, 5 μm. Boxed regions are enlarged in zoom-in panels. e, f Kinetics of different BPLs-TRF1 were compared. The stable cell lines were cultured in dialyzed serum for 3 days, and then added with 50 μM biotin for the indicated times. Cells were harvested for western blot analysis (Supplementary information, Fig. S1c, d). Total streptavidin signals from each lane were divided by the corresponding HA signals and then normalized to the 16-h BioID2 sample as appropriate. Data from all the time points were quantified and plotted (mean ± SEM, n = 3). Comparison of new biotin ligases and BioID2 were shown in e, and comparison of BioID2-/PhBPL-/Turbo-TRF1 were shown in f. Statistical significance was determined using the two-way ANOVA followed by Bonferroni post hoc test. PhBPL-TRF1 showed significantly stronger labeling than BioID2-TRF1 at the indicated time points (**P < 0.01, ***P < 0.001) (e). PhBPL-TRF1 showed significantly stronger labeling than BioID2-TRF1 at 60 min (**P < 0.01); TurboID-TRF1 showed significantly stronger labeling than BioID2-TRF1 at 60 min and 16 h (*P < 0.05, ***P < 0.001) (f).