Precise, Orthogonal Remote-Control of Cell-Free Systems Using Photocaged Nucleic Acids

Cell-free expression of a gene to protein has become a vital tool in nanotechnology and synthetic biology. Remote-control of cell-free systems with multiple, orthogonal wavelengths of light would enable precise, noninvasive modulation, opening many new applications in biology and medicine. While there has been success in developing ON switches, the development of OFF switches has been lacking. Here, we have developed orthogonally light-controlled cell-free expression OFF switches by attaching nitrobenzyl and coumarin photocages to antisense oligonucleotides. These light-controlled OFF switches can be made from commercially available oligonucleotides and show a tight control of cell-free expression. Using this technology, we have demonstrated orthogonal degradation of two different mRNAs, depending on the wavelength used. By combining with our previously generated blue-light-activated DNA template ON switch, we were able to start transcription with one wavelength of light and then halt the translation of the corresponding mRNA to protein with a different wavelength, at multiple timepoints. This precise, orthogonal ON and OFF remote-control of cell-free expression will be an important tool for the future of cell-free biology, especially for use with biological logic gates and synthetic cells.


Oligonucleotide Sequences
Unmodified oligonucleotides were purchased from IDT, and amino-modified oligonucleotides were purchased from ATDBio. All oligonucleotides were made up to 100 µM in 10 mM tris-EDTA buffer, pH 8.

S7
pairing with mRNA (counting from mRNA start codon, see Supplementary Table 2), the oligonucleotide length in number of bases, the number of amino modifications on the ASO and the sequence. The names assigned to specific sequences used throughout the text is indicated on the right. X = internal C6-hexylamino-dT modification, 2 = 5'-C6-hexylamino-phosphate-dT modification.

General
Experiments containing blue light-activatable biotin groups were performed under reduced laboratory lighting conditions (direct overhead lights turned off). Data was plotted using python's matplotlib and seaborn libraries. Error bars and confidence intervals were computed using seaborn and show a 95% confidence interval. T-tests were performed using python's scipy library, using the stats.ttest_ind() method and a one-tailed t-test was applied. Activation percentages were calculated using the following equation: × 100, where µ = mean, N = background sample, S = sample of interest, R = reference sample. Gel electrophoresis data was analysed using ImageLab software.
Reactions were incubated in the dark at room temperature for 1 hour, with frequent mixing and centrifugation, then left at 4 °C overnight. Reactions were quenched with 350 μL of 20 mM tris pH 7. The mixtures were then partially desalted by centrifugation with Amicon Ultra-0.5 mL centrifugal filters (Merck) according to the manufacturer's protocol, then purified further by HPLC.

Preparation of blue light-activatable antisense oligonucleotides
The amine-modified oligonucleotide (Entry 19, Supplementary

HPLC purification of modified antisense oligonucleotides
The modified oligonucleotides were purified by HPLC on an Agilent Polaris C-18 column (150 x 4.5 mm), heated to 50 °C, using a gradient of 5-28% MeCN over 36 minutes with 10 mM TEAB pH 8.5 throughout. The resulting oligonucleotides were lyophilised, resuspended in H2O and analysed by LCMS for purity (details on LCMS below).

Binding of streptavidin
The biotinylated antisense oligonucleotides were incubated with monovalent streptavidin (kindly provided by Howarth's laboratory, Department of Biochemistry, University of Oxford) or tetravalent streptavidin (NEB) in a 4-fold excess to the number of photocleavable biotin groups. The reactions were incubated in the dark for 2-3 hours at room temperature, then left at 4 °C overnight.

Preparation of linear template DNA
PCR reactions were carried out using DreamTaq DNA polymerase MasterMix (2X, ThermoFisher), forward and reverse primers (Supplementary

Transcription of mVenus and mCherry mRNA
In vitro transcription was performed using the linear DNA templates prepared, with the HiScribe™ T7 High Yield RNA Synthesis Kit (NEB), following the manufacturer's protocol. The reactions were diluted by 10-fold with water and incubated in the presence of DNase I (ThermoFisher) for 15 minutes at 37 °C, to remove the DNA template. The resulting RNA was then purified using the GeneJet RNA Clean-up and Concentration Kit (ThermoFisher) following the manufacturer's protocol and eluted in H2O. Prior to use, mRNA was tested for size and purity against a commercial ssRNA size marker.

UV irradiation conditions
Samples in an open 200 µL PCR tube were held in a PCR tube rack (StarLabs) over aluminium foil and irradiated top-down. Irradiation was performed with a ThorLabs 365 nm LED (M365L3) equipped with a collimator (COP5-A) from a distance of 34 cm at an irradiance of 2.12 mW·cm -2 , controlled by a ThorLabs driver (LEDD1B) set at 1 A maximum drive current. The power was measured at the position of the sample using the ThorLabs analog handheld laser power meter console PM100A, using a photodiode power sensor (Thorlabs S120VC), which has a measuring surface area of 0.7088 cm 2 (Ø of aperture: 0.95 cm).

Blue light irradiation conditions
Samples in an open 200 µL PCR tube were held in a PCR tube rack (StarLabs) over aluminium foil and irradiated top-down. Irradiation was performed with a ThorLabs 455 nm LED (M455L4) equipped with a collimator (COP4-A) from a distance of 30 cm at an irradiance of 64 mW·cm -2 for 1 minute. The power was measured at the position of the sample using the ThorLabs analog handheld laser power meter console PM100A, using a photodiode power sensor (Thorlabs S120VC), which has a measuring surface area of 0.7088 cm 2 (Ø of aperture: 0.95 cm).

In vitro transcription control using uvLA-V3
Reagents from the HiScribe™ T7 High Yield RNA Synthesis kit (NEB) were combined in a PCR tube as a master mix considering 3 μL final volume per reaction condition. Each 3 μL reaction contained the following: 0.05 μL T7 RNA Polymerase, 1X T7 RNA Polymerase buffer, and 10 mM NTPs, to which 0.001 pmol of mVenus linear DNA template, 3 U of RNase H (recombinant E. coli, Takara) and 0.05 pmol of caged ASO uvLA-V3 were also added. The reactions were incubated in the dark at 37 °C for 3 hours. At this point, a 3 μL aliquot was transferred to a different tube and irradiated with UV light (conditions above) for 5 minutes, while 0.05 pmol of amino-ASO V3 were added to a separate 3 μL aliquot as control, before all tubes were put back at 37 °C and incubated for 1 more hour. Each reaction was finally quenched with 0.5 μL of 100 mM EDTA, RNA loading dye (NEB, B0363S) was added, the samples heated to 70 °C for 10 minutes and analysed by agarose gel electrophoresis.

Two-wavelength control of in vitro transcription
Reagents from the HiScribe™ T7 High Yield RNA Synthesis kit (NEB) were combined in a PCR tube as a master mix considering 3 μL final volume per reaction condition. Each 3 μL reaction contained the following: 0.05 μL T7 RNA Polymerase, 1X T7 RNA Polymerase buffer, 10 mM NTPs, to which 0.001 pmol of bLA-mVenus DNA, 3 U of RNase H (recombinant E. Coli, Takara) and 0.05 pmol of caged ASO uvLA-V3 were also added. The reaction was started by irradiating the tube with blue light (conditions above) for 1 minute, followed by incubation in the dark at 37 °C. Every hour 3 μL were removed from the master mix reaction tube and quenched with 0.5 μL of 100 mM EDTA. After 3 hours, one 3 μL was separately irradiated with UV light (conditions above) for 5 minutes, then put back at 37 °C. All reactions were quenched with 0.5 μL of 100 mM EDTA after 1 more hour (4 hours total). The reaction controls included one reaction where no light was applied, and one reaction where only UV light was applied at the start. In parallel, an equivalent set of reactions were run with noncaged nucleic acids as control. A linear mV DNA template was added to start the reaction and the amino-ASO V3 was added to initiate RNase H-mediated degradation. All reactions were finally mixed with RNA loading dye (NEB, B0363S), heated to 70 °C for 10 minutes and analysed by agarose gel electrophoresis.

Light-controlled gene knockdown in cell-free protein synthesis with uvLA-V3
In a 200 µL PCR tube, 5 ng/µl linear mV template DNA was added to PURExpress® (NEB, E6800) with 0.67 U/µL RNase H (Takara) and 0.2 ng/µL NH2-V3 or uvLA-V3 ASO. The resulting solutions were kept at room temperature in the dark and illuminated with UV as required (5 minutes), before placing them in a thermocycler and incubated at 37 °C for 4 hours. 2 µL of each solution was then placed into 39 µL of H2O and mixed by pipetting. 40 µL of the resulting solutions were then transferred into a 384 well plate and placed into a plate reader (Tecan Infinity M1000) and fluorescence measurements were taken (λEx/Em: 515/527 nm, Gain 173).

Control of transcription and gene knockdown with two wavelengths in cell-free protein synthesis
In a 200 µL PCR tube, 5 ng/ul of blue light-activatable mV template DNA (as prepared previously) 1 were added to PURExpress® (NEB, E6800) with 0.67 U/µL RNase H (Takara) and 0.33 ng/µL uvLA-V3 and 0.2 ng/µL NH2-V3 or uvLA-V3 ASO. The resulting solutions were kept at room temperature in the dark and illuminated as required (1 minute with blue and 3 mins with UV), before placing them in a thermocycler and incubated at 37 °C. To illuminate at different timepoints, tubes were cooled to room temperature in the dark and illuminated as required, before placing them back into the thermocycler for a total of 4 hours incubation time. 2 µL of each solution were then placed into 39 µL of H2O and mixed by pipetting. 40 µL of the resulting solutions were then transferred into a 384 well plate and placed into a plate reader S11 (Tecan Infinity M1000) and fluorescence measurements were taken (λEx/Em: 515/527 nm, Gain 173).

Two-wavelength control of RNase H-mediated mRNA degradation
1 pmol of either mV or mC mRNA was incubated with 6 U of RNase H (recombinant E. coli, Takara) and 1.2 ng of each bLA-C1 and uvLA-V3 in a buffer system containing 30 mM HEPES pH 7, 100 mM KCl, 20 mM MgCl2 and 2 mM DTT. The samples were illuminated as required (5 minutes for UV irradiation or 1 minute for blue irradiation), incubated at 37 °C for 1 hour. After incubation, RNA loading dye (NEB, B0363S) was added, the samples heated to 70 °C for 10 minutes and analysed by agarose gel electrophoresis.

Two wavelength-controlled gene knockdown in cell-free protein synthesis
In a 200 µL PCR tube, 5 ng/µl each of linear mV and mC DNA (as prepared were added to PURExpress® (NEB, E6800) with 0.67 U/µL RNase H (Takara), 0.5 ng/µL bLA-C1 ASO and 0.33 ng/µL uvLA-V2 ASO. The resulting solutions were kept at room temperature in the dark and illuminated as required (3 minutes for UV irradiation or 1 minute for blue irradiation), before placing them in a thermocycler at 37 °C for 4 hours. 2 µL of each solution were then placed into 39 µL of H2O and mixed by pipetting. 40 µL of the resulting solutions were then transferred into a 384 well plate and placed into a plate reader (Tecan Infinity M1000) and fluorescence measurements were taken (λEx/Em: 515/527 nm for mV, gain 183, and λEx/Em: 587/610 nm for mCherry, gain 234).

Agarose gel electrophoresis
Agarose gels were prepared at 1.5% agarose in 1X TBE buffer and 1X Gel-Red® nucleic acid stain (Biotium), then run at 100 V in 1X TBE buffer. Samples were prepared using RNA loading dye (NEB, B0363S), heated to 70 °C for 10 minutes then cooled on ice before loading to denature the RNA. Samples were run against a low-range ssRNA ladder (NEB, N0364S) or ssRNA ladder (NEB, N0362S).

RNase H-mediated degradation of mRNA experiments
RNase H-mediated mVenus-mRNA degradation controlled by UV Supplementary Figure 1. Agarose gel showing RNase H-mediated mRNA degradation to compare the activity of V1 after modification in different positions. The amino-modified antisense oligonucleotide (ASO) V1 is the most active compared to V1b and V1c. The activity before and after UV irradiation of the three ASOs modified with uvLA-biotin and binding of monovalent (mSA) or tetravalent streptavidin (tetSA) is also compared. uvLA-V1 and uvLA-V1c show a much better on/off ratio compared to uvLA-V1b, with uvLA-V1 showing better mRNA degradation after UV. Before UV irradiation, binding of mSA or tetSA completely prevents RNase H activity, whereas after applying UV, the photocleavable group containing mSA cleaves off more easily than with tetSA.