Signaling dynamics distinguish high- and low-priority neutrophil chemoattractant receptors

Abstract

Human neutrophils respond to multiple chemoattractants to guide their migration from the vasculature to sites of infection and injury, where they clear pathogens and amplify inflammation. To properly focus their responses during this complex navigation, neutrophils prioritize pathogen- and injury-derived signals over long-range inflammatory signals, such as the leukotriene LTB4, secreted by host cells. Different chemoattractants can also drive qualitatively different modes of migration even though their receptors couple to the same Gα_i family of G proteins. Here, we used live-cell imaging to demonstrate that the responses differed in their signaling dynamics. Low-priority chemoattractants caused transient responses, whereas responses to high-priority chemoattractants were sustained. We observed this difference in both primary neutrophils and differentiated HL-60 cells, for downstream signaling mediated by Ca²⁺, a major regulator of secretion, and Cdc42, a primary regulator of polarity and cell steering. The rapid attenuation of Cdc42 activation in response to LTB4 depended on the phosphorylation sites Thr³⁰⁸ and Ser³¹⁰ in the carboxyl-terminal tail of its receptor LTB4R in a manner independent of endocytosis. Mutation of these residues to alanine impaired chemoattractant prioritization, although it did not affect chemoattractant-dependent differences in migration persistence. Our results indicate that distinct temporal regulation of shared signaling pathways distinguishes between receptors and contributes to chemoattractant prioritization.

Fig. 1.. Low priority chemoattractant receptors have faster signal attenuation than high priority receptors.

(A) High and low priority receptor activation leads to a common downstream signaling cascade driven by effectors of Gα_i and Gβγ subunits. (B) Neutrophil-like differentiated HL-60 (dHL-60) cells expressing a Cdc42 FRET sensor were imaged at 5 second intervals before and after stimulation with 6 nM LTB4 or fMLF stimulus. Representative images show pseudo-colored relative FRET ratio above and grayscale of fluorescent intensity below. Scale bar, 100 μm. (C, D) Quantification of Cdc42 dynamics in response to stimulation with different concentrations of LTB4 or fMLF (n=3 biological replicates for each condition). (E) Boxplot showing the Cdc42 response durations calculated as the time from peak signal to half-maximal value. Dots indicate independent biological replicates. P-values for comparisons between fMLF and LTB4 samples at each concentration were calculated using a permutation approach and adjusted using the Bonferroni correction (7 comparisons). (F, G) Cytosolic Ca²⁺ dynamics in response to stimulation with the indicated chemoattractants were measured by imaging dHL-60 or primary human neutrophils stained with Fluo-3 dye at 5 second intervals. Cells were treated with a saturating dose of “end-target” chemoattractants fMLF (24 nM) or the C5aR agonist ChaCha peptide (1 μM), or “intermediary” chemoattractants LTB4 or IL-8 (24 nM). Data was normalized according to the pre-stimulus baseline and the mean maximum signal for all conditions to account for staining variability. (H) Boxplot of the times from peak signal to half-maximal value for primary neutrophil Ca²⁺ responses. P-values were calculated using Tukey’s range test. Dots indicate biological replicate measurements. Curves and shaded error regions represent the mean ± SEM over biological replicate measurements. Boxes in the boxplots indicate the 25^th and 75^th percentiles, with the center bar indicating the median, and whiskers indicating the range of the data aside from automatically determined outliers. No symbol for p-value > 0.05, * for p-value < 0.05, ** for p-value < 0.01, and *** for p-value < 0.005 for figure panels where statistical calculations are indicated.

Fig. 2.. Single-cell kinetics are consistent with bulk analysis and reveal a graded response.

Data presented in Fig. 2 are single cell information extracted from the experiments shown in Fig. 1C, D. (A) Heat maps displaying Cdc42 FRET ratios over time on a single cell basis. For each concentration, 500 cells were selected as an even distribution of all cells analyzed. Data shown is arranged in descending order from top to bottom by the mean of the FRET ratio after stimulus. (B, C) Histograms displaying the frequency distributions of the single cell fold-change in Cdc42 FRET ratio, comparing the maximum after stimulus to the baseline before stimulus. (D) Hill plots of Cdc42 response to chemoattractants. Dots indicate independent biological replicates and the lines indicate the Hill-fit. Hill coefficient (n_H) for fMLF = 1.80 ± 0.25 and for LTB4 = 1.28 ± 0.04. (E, F) Violin plots showing the distribution of Cdc42 response duration in response to fMLF (E) and LTB4 (F) among single cells.

Fig. 3.. Rapid signal attenuation in response to LTB4 depends on the phosphorylation sites Thr³⁰⁸ and Ser³¹⁰ in the LTB4R receptor.

(A) Schematic showing the LTB4R GPCR and phosphorylation sites in the C-terminal tail of the receptor. Phosphorylation sites highlighted in purple are ligand-dependent, and those in grey are basally phosphorylated. (B) Amino acid sequence alignment of a section of LTB4R showing conservation across multiple metazoan species. Phosphorylation sites are highlighted using the color scheme in (A). Phosphorylation sites substituted with alanine residues by site directed mutagenesis are shown in red. TM7= trans-membrane helix 7, H8 = helix 8. (C) Plots showing the relative Cdc42 FRET ratio of dHL-60 cells expressing versions of the LTB4 receptor under the same promoter. Images were acquired at 5 second intervals before and after 6 nM LTB4 stimulus. Curves and shaded error regions represent the mean ± SEM over at least five biological replicate measurements per group. (D) Signal duration of the data in (C) was measured as time to half maximum as a dose response curve comparing mutant versions of the LTB4 receptor. P-values were calculated for comparisons between each pair of receptors within each concentration using a permutation approach with the Bonferroni correction (12 comparisons). (E) Single cell data was extracted from the experiment shown in (C) and (D). Single cell changes in Cdc42 activity are shown on heat maps, which compare different versions of the LTB4 receptor stimulated with 6 nM LTB4. 500 cells shown were selected as an even distribution from all cells.

Fig. 4.. The phosphorylation sites Thr³⁰⁸ and Ser³¹⁰ limit LTB4 receptor endocytosis.

(A) A schematic depicting detection of LTB4R endocytosis measured by loss of fluorescent signal. Versions of LTB4R tagged with 3xHA (hemagglutinin-tag) were expressed in dHL-60 cells for quantification by immunocytometry for HA. Comparison of cells stained without compared to with LTB4 stimulation was used to determine the percentage of receptor remaining on the surface. Allophycocyanin (APC) signal (LTB4R on the cell surface) was measured by flow cytometry. (B) Representative histograms showing LTB4R cell surface levels. Cells were either unstimulated or stimulated with 12 nM LTB4 for 10 minutes prior to immunostaining for HA (left). Boxplot (right) quantifying receptor internalization of LTB4R. 100% is defined as the surface LTB4R level in unstimulated cells and indicates no internalization of the receptor. Data represent 5 independent biological replicates per group. (C) Data showing experiments as performed in (B) but with cells stimulated with 250 nM LTB4 for 30 minutes to obtain maximal internalization of the receptors. Data represent 5 independent biological replicates per group. P-values were calculated using Tukey’s range test.

Fig. 5.. The phosphorylation sites Thr³⁰⁸ and Ser³¹⁰ in LTB4R are necessary for chemoattractant prioritization.

(A) A diagram displaying the layout of a microscopy-based competing chemoattractant chemotaxis assay performed in a 24-well plate format. Cells move across the imaging plate surface under 1.5% agarose. Diffusion-based gradients were created by adding chemoattractant to “reservoirs” on one or both sides of the well and cell movement can be tracked over time by imaging (B) Chemotaxis of dHL-60 cells expressing LTB4R wild-type (WT) or LTB4R (T308A/S310A) was compared using the assay described in (A). Cells were stained with Hoechst and images were acquired for 60 minutes at a rate of 1 frame/minute. Net chemotaxis was defined as directed speed of cells toward fMLF (positive) or LTB4 (negative) and was measured as the rate of movement of the cell in the direction of the gradient source. Dots indicate independent biological replicates (n = 6 per group). P-values for comparisons between WT and T308A/S310A were calculated using Welch’s t-test and corrected with the Bonferroni method (4 comparisons). (C) Bar graph showing the average speed of cells in (B). (D) Images showing representative traces (green) of cell movement over 60 minutes. The initial image (blue) was merged with the final image (yellow) to show the starting and ending locations of the cells. (E) A plot showing the directional persistence of cell movement, defined as the mean cosine of the angle between a migrating cell’s movement direction at two different time points was measured as a function of the difference in time between the two measurements. Cells migrating in a straight line would have a mean cosine angle of 1. Curves and shaded error regions represent the means ± SEM for 6 biological replicates per group.

Fig. 6.. Shared signaling networks between chemoattractant receptors allow for prioritization of fMLF over LTB4 signals.

A diagram displaying our model of signal transduction in response to multiple chemoattractant stimuli. In a resting state (top), receptors are inactive with guanosine diphosphate (GDP) bound G-proteins. Exposure to chemoattractant gradients (middle) activates the receptors, leading to dissociation of G-proteins in their guanosine triphosphate (GTP)-bound state and activation of downstream signaling pathways. Following stimulus, GRKs rapidly phosphorylate LTB4R (bottom), thereby switching the receptor off. The rapid desensitization of LTB4R allows the neutrophil to prioritize fMLF signals, and the cell will move in an fMLF-directed manner.