Light-Inducible Recombinases for Bacterial Optogenetics

Abstract

Optogenetic tools can provide direct and programmable control of gene expression. Light-inducible recombinases, in particular, offer a powerful method for achieving precise spatiotemporal control of DNA modification. However, to-date this technology has been largely limited to eukaryotic systems. Here, we develop optogenetic recombinases for Escherichia coli that activate in response to blue light. Our approach uses a split recombinase coupled with photodimers, where blue light brings the split protein together to form a functional recombinase. We tested both Cre and Flp recombinases, Vivid and Magnet photodimers, and alternative protein split sites in our analysis. The optimal configuration, Opto-Cre-Vvd, exhibits strong blue light-responsive excision and low ambient light sensitivity. For this system we characterize the effect of light intensity and the temporal dynamics of light-induced recombination. These tools expand the microbial optogenetic toolbox, offering the potential for precise control of DNA excision with light-inducible recombinases in bacteria.

Keywords: Cre; inducible recombinase; optogenetics; photoactivation; recombinase.

Figure 1.

Light-inducible recombination in E. coli. (a) Split Cre fragments are linked to Vvd photodimers and expressed under the control of an IPTG-inducible promoter (P_lacUV5). When exposed to blue light, Vvd dimerizes, forming functional Cre protein. Cre can then act on the reporter plasmid, excising the loxP-flanked transcription terminator and allowing expression of RFP. RFP is under the control of a constitutive promoter (P_W4). (b) Gel electrophoresis images showing DNA excision. PCR of the reporter region containing loxP-flanked terminator shows a 500 bp band when full terminator is intact, and 300 bp band after recombination. Negative control contains cells with the reporter plasmid alone (terminator upstream of rfp); positive control contains cells with recombinase and a precut reporter plasmid (no terminator upstream of rfp). (c) Single-cell fluorescence microscopy showing RFP expression for cells with and without light exposure. Insets below show representative cell images (scale bar = 2 μm). Error bars show standard error around the mean (n ≈ 300 cells per sample). In addition, we tested for statistical significance between conditions with and without light exposure using a two-tailed Welch’s t-test by using individual microscopy images as replicates, **P < 0.005.

Figure 2.

Comparison of optogenetic recombinase protein variants. (a) RFP reporter output with and without light exposure for Cre and Flp recombinases, each with Vvd and Magnet photodimers. Split sites used are Cre with nCre length of 43 AA, Flp with nFlp length of 374 AA. (b) Assay of split sites tested for Cre-Vvd and (c) Cre-Mag. Numbers shown for split sites on the x-axis are the length of nCre. All figure data were obtained using fluorescence microscopy. In all cases, 100 μM IPTG was used for induction and samples were exposed to 1 h of light at 120 μW/cm². Error bars show standard error around the mean (n ≈ 750 cells per sample). Statistical significance comparing conditions with and without light use a two-tailed Welch’s t-test using microscopy images as replicates. *P < 0.05, **P < 0.005.

Figure 3.

Characterization of Opto-Cre-Vvd. (a) RFP reporter output for Opto-Cre-Vvd without light (−), with a 5 min exposure to ambient light (a), and with full exposure to blue light (+). (b) Effect of blue light intensity on Opto-Cre-Vvd activation when exposed for 1 or 4 h. (c) Effect of IPTG induction levels for Opto-Cre-Vvd. Statistical significance comparing conditions with and without light use a two-tailed Welch’s t-test using microscopy images as replicates. *P < 0.05, **P < 0.005. In all cases, error bars show standard error around the mean from microscopy data (n ≈ 350 cells per sample).

Figure 4.

Light exposure duration necessary to induce excision by Opto-Cre-Vvd. (a) DNA gel image showing reporter bands with and without transcription terminator excision, (b) single-cell fluorescence microscopy averages of RFP values, (c) representative microscopy images (scale bar = 2 μm), and (d) samples of culture spotted on agar plates of Opto-Cre-Vvd exposed to different durations of blue light. Error bars show standard error around the mean (n ≈ 750 cells per sample). (e) Opto-Cre-Vvd activation in real-time, with the blue bar indicating light exposure. Shaded error bars represent standard deviation around the mean from plate reader data (n = 3 wells).