A genetically encoded sensor for visualizing leukotriene B4 gradients in vivo

Abstract

Leukotriene B₄ (LTB₄) is a potent lipid chemoattractant driving inflammatory responses during host defense, allergy, autoimmune and metabolic diseases. Gradients of LTB₄ orchestrate leukocyte recruitment and swarming to sites of tissue damage and infection. How LTB₄ gradients form and spread in live tissues to regulate these processes remains largely elusive due to the lack of suitable tools for monitoring LTB₄ levels in vivo. Here, we develop GEM-LTB₄, a genetically encoded green fluorescent LTB₄ biosensor based on the human G-protein-coupled receptor BLT1. GEM-LTB₄ shows high sensitivity, specificity and a robust fluorescence increase in response to LTB₄ without affecting downstream signaling pathways. We use GEM-LTB₄ to measure ex vivo LTB₄ production of murine neutrophils. Transgenic expression of GEM-LTB₄ in zebrafish allows the real-time visualization of both exogenously applied and endogenously produced LTB₄ gradients. GEM-LTB₄ thus serves as a broadly applicable tool for analyzing LTB₄ dynamics in various experimental systems and model organisms.

Fig. 1. Development and characterization of the GEM-LTB₄ sensor in HEK293A cells.

a Schematic diagram of LTB₄-sensor design showing fluorescence increase upon ligand binding. The sensor consists of cpEGFP inserted with linkers into the 3rd intracellular loop (ICL3) of BLT1. b Summary of ΔF/F₀ fluorescence responses in all LTB₄-sensor variants tested in this study. Data shown as mean ± SEM for n = 106, 134, 61, 90, 17, 105, 30, 62, 135, 98, 139, 62, 47, 26, 97 and 86 cells from 3 independent experiments, respectively. GEM-LTB₄ is shown in blue. c Representative GEM-LTB₄ and GEM-LTB₄mut confocal fluorescence and corresponding ΔF/F₀ ratio images in HEK293A cells before and after 100 nM LTB₄ stimulation. Scale bars, 25 µm. d Excitation and emission spectra of GEM-LTB₄ in the absence (dotted lines) and presence (continuous lines) of 100 nM LTB₄. Insert shows ratio of excitation spectra. Each trace is the average of n = 3 independent experiments (arb. units=arbitrary units). e Dose-response measurements of GEM-LTB₄ and GEM-LTB₄mut, with the corresponding EC₅₀ value. Data shown as mean ± SEM for n = 66 and 218 cells per condition, respectively, from 3 independent experiments. EC₅₀ value was obtained by fitting the data to a four-parameter log-logistic function. f Average ΔF/F₀ responses of GEM-LTB₄ and GEM-LTB₄mut to sequentially added increasing doses of LTB₄. Data shown as mean ± SEM for n = 113 and 181 cells, respectively, from 3 independent experiments. g GEM-LTB₄ response to 100 nM LTB₄ stimulation followed by treatment with 1 µM of the BLT1 inhibitor CP-105,696. Data shown as mean ± SEM for n = 85 cells from 3 independent experiments. h Kinetic analysis from high-speed acquisition of GEM-LTB₄ fluorescence in HEK293A cells during 100 nM LTB₄ stimulation. All measured normalized data points and the average fitted curve are shown from n = 13 cells from 7 independent experiments. i Maximal ΔF/F₀ responses of GEM-LTB₄ and GEM-LTB₄mut to 100 nM of the indicated eicosanoid compounds. Data shown as mean ± SEM for n = 45 and 88 cells, respectively, from 3 independent experiments. Statistical analysis were performed with One-way ANOVA (F = 761.3, p = 2.0 × 10⁻¹⁸¹) with Dunnett’s correction (20-OH-LTB₄ and LTB₄ are different from control for GEM-LTB₄). Source data are provided as a Source Data file.

Fig. 2. Real-time measurements of LTB₄ release from neutrophils with GEM-LTB₄.

a ELISA measurement of LTB₄ secretion by murine neutrophils stimulated with 2 µM fMLP for 28 min. Data shown as mean ± SEM for 3 independent experiments, two-tailed unpaired t-test, ^**P = 0.0012. b Brightfield microscopy and corresponding ΔF/F₀ ratio images of murine neutrophils seeded over HEK293A cells expressing GEM-LTB₄ (left) and GEM-LTB₄mut (right). Scale bars, 100 µm. Representative images were taken before neutrophil addition (basal), after adding neutrophils and stimulating with 2 µM fMLP (20 min) and followed by 100 nM LTB₄ stimulation (35 min). c Average traces of ΔF/F₀ responses in GEM-LTB₄ and GEM-LTB₄mut expressing cells shown in b. Data are presented as mean ± SEM for n = 600 and 465 cells respectively from 3 independent experiments. d Relative surface area of GEM-LTB₄ expressing cells shown in b, reacting with over 50% increase in normalized ΔF/F₀ as a response to 2 × 10⁶ neutrophils stimulated with 2 µM fMLP in a 1 cm² chamber. Data shown as mean ± SEM for n = 15-15 fields of view from 3-3 independent experiments with two-tailed unpaired t-test, ^***P = 0.001159. e Brightfield microscopy and corresponding ΔF/F₀ ratio images of GEM-LTB₄ expressing HEK293A with neutrophils seeded over them at a ~ 1/10 density compared to b. Representative ΔF/F₀ images taken before neutrophil addition (basal) and at two time points after stimulation with 2 µM fMLP followed by 100 nM LTB₄. White asterisk (*) refers to the center of the analysis shown in f. Scale bars, 20 µm. f Representative spatiotemporal traces of pixelwise ΔF/F₀ GEM-LTB₄ values from e. The spatial origo is the center of the marked neutrophil (*) also shown in the inlay image. Similar results were obtained in n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 3. Expression of GEM-LTB₄ in zebrafish and exogenous LTB₄ penetration detection.

a Representative confocal fluorescence imaging and quantification of ΔF/F₀ responses of GEM-LTB₄ in Tg(krt4:QF2 x QUAS:GEM-LTB₄) (top) and Tg(krt19:QF2 x QUAS:GEM-LTB₄) (bottom) zebrafish larvae before and after 1 µM LTB₄ stimulation. Enlarged images show the cellular distribution of GEM-LTB₄ expression in the respective epithelial layers. Scale bars, 100 µm and data are presented as mean ± SEM for n = 12 and 16 cells from 3 and 4 independent fish, respectively. b Representative ΔF/F₀ of time-lapse images of amputated zebrafish larvae Tg(krt4:QF2 x QUAS:GEM-LTB₄) after stimulation with 1 µM LTB₄. Scale bar, 100 µm. c Averaged spatiotemporal profile plot of GEM-LTB₄ ΔF/F₀ responses after stimulation of intact tail fins with 1 µM LTB₄ in Tg(krt4:QF2 x QUAS:GEM-LTB₄) larvae. n = 3 larvae. d Averaged spatiotemporal profile plot of GEM-LTB₄ (left) and GEM-LTB₄mut (right) ΔF/F₀ responses after stimulation of amputated tail fins with 1 µM LTB₄ in Tg(krt4:QF2 x QUAS:GEM-LTB₄) and Tg(krt4:QF2 x QUAS:GEM-LTB₄mut) larvae. The amputation and stimulation were performed under isotonic conditions (see Methods for details). n = 3 larvae. e Measurement of neutrophil movement triggered by control or LTB₄ towards amputational tail fin wounds imaged in Tg(mpx:GFP)i114 zebrafish larvae by light transmission and fluorescence microscopy. Top left: scheme of neutrophil movement quantification towards the wound. Left: representative leukocyte tracks capturing all visible cell movements during imaging in control and 1 µM LTB₄ treated samples. Right: Time course of average neutrophil movement towards the wound shown in b, in control and in 1 µM LTB₄ stimulated larvae. Data are shown as mean ± SEM for n = 25 and 38 cells from 5 and 4 independent experiments, respectively, with a two-tailed unpaired t-test, ^****P = 1.6 × 10⁻⁹ performed at the endpoint of the measurement. Source data are provided as a Source Data file.

Fig. 4. Endogenous LTB₄ release measured by GEM-LTB₄ in zebrafish.

a Schematics of experimental design for measuring endogenous LTB₄ release. Larvae were wounded and pre-incubated in isotonic E3 with 20 µM arachidonic acid for 90 min alone or in combination with 20 µM of the 5-lipoxygenase inhibitor zileuton, and then stimulated with 100 µM A23187 after mounting in hypotonic agarose. b Representative pseudo-color of green/red ratio of time-lapse images of zebrafish larvae Tg(krt19:QF2 x QUAS:GEM-LTB₄) (top and bottom) and Tg(krt19:QF2 x QUAS:GEM-LTB₄mut) (middle) in the basal epithelial cells. Scale bars, 50 µm. c Brightfield images of the corresponding wounded tail fins in b, at 90 min with manually traced leukocyte time lapse tracks shown in color. Scale bars, 50 µm. d Averaged spatiotemporal profile plots of corresponding experiments shown in b. n = 3 larvae each.