4.8. Femtosecond Transient Absorption (fs-TA) Spectroscopy

Our fs-TA setup was built on the basis of a 1 kHz-repetition-rate Ti:sapphire regenerative amplifier (Legend Elite-USP-1K-HE, Coherent, Inc.) which produces the ~800 nm fundamental pulses with ~35 fs duration and ~3.7 W power (56, 57). In brief, the fs 490 nm photoexcitation pulse train was generated from a home-built two-stage fs noncollinear optical parametric amplifier (NOPA) system and then compressed by a chirped mirror pair (DCM-12, 400–700 nm, Laser Quantum, Inc.). The probe pulse as supercontinuum white light was produced by focusing a portion of the fundamental laser output onto a 2-mm-pathlength deionized-water-filled quartz cell (Spectrosil 1-Q-2, Starna Cells, Inc.), then passed through a chirped mirror pair (DCM-9, 450–950 nm, Laser Quantum, Inc.) for temporal compression (57). The actinic pump power was set at ~0.25 mW by using a circular variable metallic neutral density (ND) filter (50FS04DV.4, Newport, Inc.). The pump and probe pulses were both focused by a reflective parabolic mirror on the 1-mm-pathlength quartz flow cell (Spectrosil 48-Q-1, Starna Cells, Inc.) filled with sample solution at room temperature. We note that tetrazine is not photostable over a long period of time, so we stored samples (both free tetrazines and GFP-Tet variants) in the dark inside a 4 °C fridge and only took them out before each fs-TA experiment and then immediately put them back after each measurement. Our flow cell and low actinic pump power did not cause any notable damage to sample solution during the experiments (Figure S8). Past the sample in our optical setup, the probe beam was collimated and focused into a spectrograph (IsoPlane SCT-320, Princeton Instruments, Inc.), dispersed by a reflective grating (300 grooves/mm, 300 nm blaze wavelength), and then imaged on a CCD array camera (PIXIS:100F, Princeton Instruments, Inc.) for spectral data collection and processing.

The sample solution linear flow rate was measured at 0.12 mL/s in our flow cell with a dimension of 0.8 × 0.1 × 4 cm³ (L × W × H) and a total volume of 0.3 mL (48-Q-1, see above). The diameter of actinic pump spot size was ~150 μm, hence it takes ~10 ms for the sample to flow out of the pump laser spot. In our experiments, each spectral data point was collected with 3000 laser shots at 1 kHz repetition rate, with a phase-stable optical chopper synchronized at 500 Hz, so there are five laser pulses acting as the fs pump (with the corresponding fs probe) before the sample is completely replenished, then repeated for 300 times consecutively to increase the signal-to-noise ratio. The data were collected from −2 to 900 ps with 50 fs steps from −100 fs to 2 ps, 0.2 ps steps from 2 to 4 ps, 0.5 ps steps from 4 to 10 ps, 5 ps steps from 10 to 100 ps, and 50 ps steps from 100 to 900 ps. Given the typical ns lifetimes of protein chromophores in this work, our measurement scheme is sufficient in recording high-quality ultrafast electronic spectra that accurately track the chromophore dynamics and enable a systematic comparison between related GFP-Tet protein systems.