Opto-APC: Engineering of cells that display phytochrome B on their surface for optogenetic studies of cell-cell interactions

Abstract

The kinetics of a ligand-receptor interaction determine the responses of the receptor-expressing cell. One approach to experimentally and reversibly change this kinetics on demand is optogenetics. We have previously developed a system in which the interaction of a modified receptor with an engineered ligand can be controlled by light. In this system the ligand is a soluble Phytochrome B (PhyB) tetramer and the receptor is fused to a mutated PhyB-interacting factor (PIF^S). However, often the natural ligand is not soluble, but expressed as a membrane protein on another cell. This allows ligand-receptor interactions in two dimensions. Here, we developed a strategy to generate cells that display PhyB as a membrane-bound protein by expressing the SpyCatcher fused to a transmembrane domain in HEK-293T cells and covalently coupling purified PhyB-SpyTag to these cells. As proof-of-principle, we use Jurkat T cells that express a GFP-PIF^S-T cell receptor and show that these cells can be stimulated by the PhyB-coupled HEK-293T cells in a light dependent manner. Thus, we call the PhyB-coupled cells opto-antigen presenting cells (opto-APCs). Our work expands the toolbox of optogenetic technologies, allowing two-dimensional ligand-receptor interactions to be controlled by light.

Keywords: SpyCatcher; T cell receptor; interaction; ligand; optogenetics; phytochrome B; receptor.

FIGURE 1

Use of optogenetics to engineer 2D ligand-receptor pairs. Cells displaying a phytochrome B fragment (PhyB_1-651) on their membrane resemble settings in which the ligand is cell bound. This ligand will bind to a receptor fused to PIF^S at red light but not at far-red light conditions. This will resemble the 2D interactions seen in many ligand-receptor pairs.

FIGURE 2

Engineering the PhyB-SpyTag and SpyCatcher-TMD-BFP system. (A) Scheme of the PhyB-SpyTag constructs without and with mCherry. (B) PhyB-SpyTag was expressed in bacteria and purified. The lysate (L), washing step (W), flow through (FT) and eluate (E) fractions of the Ni²⁺-affinity chromatography were separated by SDS-PAGE and proteins visualized by Coomassie (n = 2). (C) Purified PhyB-SpyTag and PhyB-mCherry-SpyTag were illuminated with 660 nm (red) or 740 nm (far-red) light and the absorbance spectra measured. (D) Scheme of the SpyCatcher-TMD-BFP construct. (E) SpyCatcher-TMD-BFP was lentivirally expressed in DAP-DR1-ICAM1 and HEK-293T cells, stained with an AlexaFluor647-coupled anti-HA tag antibody, and measured by flow cytometry. Dot plots of the anti-HA stain and BFP fluorescence intensity from untransduced and transduced cells are shown (n > 3). (F) Illustration of the covalent coupling of PhyB-mCherry-SpyTag to SpyCatcher-TMD-BFP expressed on the cell. (G) DAP-DR1-ICAM1 and HEK-293T cells expressing SpyCatcher-TMD-BFP or not (untransduced) were incubated with 10 μg/mL of the PhyB-mCherry-SpyTag protein or not (0 μg/mL) for 20 min at 37°C in RPMI 1640 medium. After washing, the cells were measured by flow cytometry and the fluorescence intensities of mCherry and BFP are shown (n > 3).

FIGURE 3

Binding of GFP-PIF to PhyB-mCherry-SpyTag-coupled cells. (A) Illustration of the binding of GFP-PIF to PhyB-mCherry-SpyTag, which was coupled to SpyCatcher-TMD-BFP-expressing cells. (B) SpyCatcher-TMD-BFP-expressing HEK-293T cells were loaded with 10 μg/mL PhyB-mCherry-SpyTag for 30 min at 37°C and 5% CO₂. After washing, the cells were incubated with 10 μg/mL GFP-PIF in PBS, 1% FBS, 0.1 mM EDTA and 0.1 mM TCEP. The samples were illuminated for 1 min with either 660 nm (red) or 740 nm (far-red) light and left in the dark for 30 min. The cells were then washed and analyzed by flow cytometry in the dark (n > 3). (C) Quantification of the GFP MFI from (B). (D) SpyCatcher-TMD-BFP-expressing HEK-293T cells were detached with EDTA in PBS and loaded or not with PhyB-mCherry-SpyTag as indicated. 100 nM GFP-PIF was added (right panels) or not (left panels) and cells were illuminated with red, 660 nm or far-red, 740 nm light and incubated for 20 min in the dark. Cells were imaged by confocal microscopy. Scale bar represents 20 µm (n > 3).

FIGURE 4

Improving the functionality of PhyB_1-651 coupled to cells in mammalian cell culture conditions. (A) HEK-293T SpyCatcher-TMD-BFP cells were coupled to 10 μg/mL PhyB-mCherry-SpyTag for 30 min and then incubated in RPMI medium with 1% FBS at 37°C for the time points indicated. Subsequently, cells were treated with 10 μg/mL GFP-PIF in PBS and illuminated for 1 min with 660 nm (red) or 740 nm (far-red) light, and left in the dark for 30 min. After washing, the cells were analyzed by flow cytometry and the MFI of GFP is displayed (n = 2). (B) In parallel, the cells from (A) without incubation with GFP-PIF were stained with an anti-His tag antibody as above and the MFI of this stain is shown (n = 2). (C) The experiment was performed as in (A), but adding the indicated amounts of TCEP during the incubation times. GFP-PIF binding was assessed with 660 nm light illumination (n = 2). (D) PhyB-mCherry-SpyTag was treated with IAA for 1 h, illuminated with 660 nm (red) or 740 nm (far-red) light and the absorbance spectra was measured. (E) The experiment was performed as in (C) at the indicated conditions (n = 1).

FIGURE 5

Stimulation of GFP-PIF^S-TCR T cells by the PhyB_1-651-coupled HEK-293T cells. (A) Scheme of the HEK-293T SpyCatcher-TMD-BFP cells coupled to PhyB-mCherry-SpyTag, called opto-APCs, stimulating GFP-PIF^S-TCR-expressing T cells at red light (630 nm), but not at far-red light (780 nm). (B) GFP-PIF^S-TCR-expressing Jurkat T cells were co-cultured at 37°C for 6 h with the following cells: no cells (−), un-coupled HEK-293T SpyCatcher-TMD-BFP cells (HEK Spy), opto-APCs with a 30 s 630 nm light illumination every 15 min (opto-APC, red light), opto-APCs with 1 min 780 nm light every 15 min (opto-APC, far-red light), or with the stimulating anti-TCR antibody Jovi3. Cells were stained with a PE-Cy7-conjugated anti-CD69 antibody and measured by flow cytometry. anti-CD69 fluorescence is shown on gated T cells. (C) The precent of CD69-positive T cells (left) and the MFI of the anti-CD69 stain (right) are shown for the experiment in (B); n > 3.

FIGURE 6

The timing of ligand binding determines the cellular response. (A) GFP-PIF^S-TCR-expressing Jurkat T cells were co-cultured at 37°C for 6 h with no additional cells (only T cells), or with the opto-APCs. At the start of the co-culture, cells were illuminated with 780 nm (far-red) light for 1 min and after 5 h and 55 min (5 min), 5 h and 30 min (30 min), 5h (1 h), 4h (2 h) or 2 h (4 h) cells were exposed to 30 s 630 nm (red) light allowing ligand binding (the ligand binding time is the value given in the brackets and in the right panel). For every light condition pulses were applied every 15 min. The time from the first 630 nm light illumination to the completion of the total 6 h is given. After 6 h, the cells were stained with a PE-Cy7-conjugated anti-CD69 antibody and measured by flow cytometry. After gating on the T cells, the anti-CD69 fluorescence is shown, n = 2. (B) The experiment was done as in (A) with the difference that the co-culture started under 630 nm red light. After 0 min, 5 min, 30 min, 1 h, 2 h or 4 h the cells were exposed to 1 min 780 nm far-red light disrupting ligand binding. After a total of 6 h the cell were measured and analysed.