Synthetic cells with self-activating optogenetic proteins communicate with natural cells

Abstract

Development of regulated cellular processes and signaling methods in synthetic cells is essential for their integration with living materials. Light is an attractive tool to achieve this, but the limited penetration depth into tissue of visible light restricts its usability for in-vivo applications. Here, we describe the design and implementation of bioluminescent intercellular and intracellular signaling mechanisms in synthetic cells, dismissing the need for an external light source. First, we engineer light generating SCs with an optimized lipid membrane and internal composition, to maximize luciferase expression levels and enable high-intensity emission. Next, we show these cells' capacity to trigger bioprocesses in natural cells by initiating asexual sporulation of dark-grown mycelial cells of the fungus Trichoderma atroviride. Finally, we demonstrate regulated transcription and membrane recruitment in synthetic cells using bioluminescent intracellular signaling with self-activating fusion proteins. These functionalities pave the way for deploying synthetic cells as embeddable microscale light sources that are capable of controlling engineered processes inside tissues.

Fig. 1. Optimizing the lipid composition for light-interacting synthetic cells.

a Illustration of a protein producing synthetic cell. b The effect of the lipid membrane composition on blue light absorbance in liposomes. Data is expressed as the mean ± standard deviation (n = 3 independent samples). Nested two-tailed t-test P values; **p = 0.0079, **p = 0.0058, **p = 0.0055 for comparisons of POPC vs. DOPC, DPPC and HSPC, respectively. c Size distribution of synthetic cells with 1:1 (w/w) POPC:cholesterol membrane composition. Data is expressed as mean of n = 3 independent samples. d Morphology of 1:1 (w/w) POPC:cholesterol synthetic cells imaged with cryogenic scanning electron microscopy (cryo-SEM). e The percentage of active synthetic cells measured using imaging flow cytometry of GFP-expressing synthetic cells (n = 331 synthetic cells without DNA and n = 308 synthetic cells with GFP DNA). Frequency is normalized to the total number of cells in each sample. f Illustration of a linear DNA oligonucleotide, encapsulated in a synthetic cell or free in solution, exposed to UV radiation. Formation of pyrimidine dimers is detected using PCR amplification. g Gel electrophoresis of the amplified PCR product of linear DNA after exposure to UV radiation with and without encapsulation in synthetic cells. Lane 1: free DNA exposed to UV. Lane 2: encapsulated DNA exposed to UV. Lane 3: no DNA control. Lane 4: untreated DNA control. This experiment was reproduced n = 3 times. Source data are provided as a Source Data file.

Fig. 2. Engineering light-generating synthetic cells.

a Comparison of light emission in cell-free protein synthesis reactions expressing Renilla Luciferase (Rluc) or Gaussia luciferase (Gluc) using unmodified lysate (with reducing conditions). Nested two-tailed t-test P value; ****p = 3.82 × 10^-6. b Light emission by Gluc-expressing synthetic cells containing lysate modified with glutathione and disulfide bond isomerase C (DsbC) to produce an oxidizing environment, compared to light emission by Rluc-expressing synthetic cells. Nested one-way ANOVA with multiple comparisons test adjusted P value; ****p = 6.6×10^-14. c Light emission from a Gluc-expressing synthetic cell solution. d Western blot quantification of Gluc production in synthetic cells. Gaussia luciferase concentration in synthetic cells is calculated based on n = 2 independent experiments. e Gluc production kinetics in synthetic cells at 37 °C. f Kinetics of the Gluc enzymatic reaction in synthetic cells after one addition of 100 µM coelenterazine. g Light emission from Gluc-expressing synthetic cells diluted to different concentrations after incubation. Nested one-way ANOVA with multiple comparisons test adjusted P-values; ****p = 8.3×10^-7, ***p = 0.00062, ****p = 3.7×10^-5, ****p = 1.3×10^-8, for comparisons of 4.2×10⁵ synthetic cells ml^-1 to 0.8×10⁵, 1.7×10⁵, 8.4×10⁵, and 16.8×10⁵ synthetic cells ml^-1, respectively. h Light emission in ranging coelenterazine concentrations by Gluc-expressing synthetic cells. Data is expressed as a mean ± s.e.m. (n = 3 independent samples). i Temporal control over light emission in Gluc-expressing synthetic cells with two timed additions of 0.125 nmol coelenterazine (second addition marked with a green arrow). for a, b, e, f, g, i. Data is expressed as a mean ± standard deviation (n = 3 independent samples). Source data are provided as a Source Data file.

Fig. 3. Activation of fungal cells using light-producing synthetic cells.

a Illustration of the experimental setup for photo-activation of conidiation in Trichoderma atroviride with Gaussia luciferase (Gluc)-expressing synthetic cells. b (i) Representative images of Trichoderma atroviride plates after exposure to Gluc-expressing synthetic cells or synthetic cells without DNA. (ii) A magnified image and a (iii) black and white thresholded image of the plate area marked with a white rectangle in which the synthetic cells were localized. c Quantitative analysis of the sporulated area out of the total area exposed to synthetic cells. Horizontal lines indicate median values; boxes indicate quartiles 1 and 3; whiskers indicate the min and max values (n = 5 independent samples). Welch’s two-tailed t-test P value; *p=0.0103. d The effect of varying synthetic cell concentrations on photo-activation of conidiation in Trichoderma atroviride. Data is expressed as a mean ± s.e.m. (n = 5 independent samples for 420,000 synthetic cells ml^-1, n = 3 independent samples for all other synthetic cell concentrations). Welch’s two-tailed t-test P-values; *p = 0.01, ***p = 0.0009, **p = 0.0098, **p = 0.0026, **p = 0.0095, **p = 0.0045, for comparisons of 420,000 and 84,000 synthetic cells ml^-1 vs. 42,000, 8400 and 4200 synthetic cells ml^-1, respectively. e A light dose-response calibration curve was generated in Trichoderma atroviride using a blue LED (black circles). A Michaelis-Menten curve was fitted to the data (dotted pink line, R²=0.9582) and the total light dose of the synthetic cell treatment was calculated according to the fitted model (pink square). Data is expressed as mean ± standard deviation (n = 2 independent samples for light doses of 20, 1000, 2000 μE m^-2 and the synthetic cells samples, n = 3 independent samples for all other light doses). Source data are provided as a Source Data file.

Fig. 4. Bioluminescent signaling self-activates transcription in synthetic cells.

a, Illustration of the light-dependent transcription mechanism mediated by the transcription factor EL222. b, EL222 concentration affected the production of Renilla luciferase (Rluc) in cell-free protein synthesis (CFPS) reactions under light and dark conditions. Data is expressed as a mean ± standard deviation (n = 5 independent samples for 10 μM EL222 under light conditions, n=6 independent samples for all other experimental groups). Two-way ANOVA with multiple comparisons test adjusted P-values; ****p = 3.26×10^-8, ****p = 2.51×10^-8, ***p = 0.0005, for comparison of light vs. dark conditions of 2.5, 5 and 10 μM, respectively. c RFP production kinetics in CFPS reactions supplemented with EL222 under dark and light conditions, with or without RFP DNA. Data is expressed as a mean ± s.e.m. (n = 4 independent samples for the “no DNA” groups, n = 6 independent samples for other groups). d Light-to-dark ratio of Rluc expression in synthetic cells containing EL222 and a DNA plasmid expressing Rluc under different promoters. Data is expressed as a mean ± s.e.m. (n = 4 independent samples for the pBLind promoter group, n = 3 for other experimental groups). One-way ANOVA with multiple comparisons test adjusted P value; **p = 0.0016, **p = 0.005, for comparisons of the pBLind group vs. the no DNA and pT7 groups, respectively. e A block diagram of the Gluc-EL222 fusion protein elements. Below, a schematic representation of a synthetic cells containing the Gluc-EL222 fusion protein for bioluminescent activation of transcription. f RFP production in synthetic cells containing the Gluc-EL222 protein with or without addition of coelenterazine. Horizontal lines indicate median values; boxes indicate quartiles 1 and 3; whiskers indicate the min and max values (n = 5 independent samples). Student’s two-tailed t-test P value; *p = 0.0134. Source data are provided as a Source Data file.

Fig. 5. Bioluminescence-activated membrane recruitment in synthetic cells.

a A block diagram of the Gaussia luciferase (Gluc)-iLID fusion protein elements. Below, a schematic representation of protein recruitment to the synthetic cell membrane by hetero-dimerization of the fusion protein Gluc-iLID with RFP-sspB. b Microscopy image of luciferase light emission from synthetic cells with membrane-bound Gluc-EL222 after coelenterazine addition. This experiment was reproduced n = 5 times. c (i) Membrane recruitment of RFP-sspB with iLID or Gluc-iLID using 488 nm laser illumination or by addition of coelenterazine to activate the bioluminescent reaction. RFP intensity is normalized to the average intensity of each cell in the dark conditions. Data is expressed as a mean ± s.e.m. Welch’s two-tailed t-test P-values; **p = 0.0092, ***p = 0.0001, for comparison of Gluc-iLID vs. iLID activated by laser and iLID supplemented with coelenterazine at t = 6 min, respectively. (ii) Representative single-cell images of RFP-sspB recruitment to a synthetic cell with membrane-bound Gluc-iLID in the dark and after two and four doses of 0.2 nmol coelenterazine (Top row). Bottom row displays the fluorescence of the synthetic cell’s DOPE-Cy5 lipid that composed 0.5 mol% of the membrane (n = 3 for iLID + coelenterazine, n = 11 for iLID + laser, n = 14 for Gluc-iLID + coelenterazine). Source data are provided as a Source Data file.