Lyn is a redox sensor that mediates leukocyte wound attraction in vivo

Abstract

Tissue wounding induces the rapid recruitment of leukocytes. Wounds and tumours--a type of 'unhealed wound'--generate hydrogen peroxide (H(2)O(2)) through an NADPH oxidase (NOX). This extracellular H(2)O(2) mediates recruitment of leukocytes, particularly the first responders of innate immunity, neutrophils, to injured tissue. However, the sensor that neutrophils use to detect the redox state at wounds is unknown. Here we identify the Src family kinase (SFK) Lyn as a redox sensor that mediates initial neutrophil recruitment to wounds in zebrafish larvae. Lyn activation in neutrophils is dependent on wound-derived H(2)O(2) after tissue injury, and inhibition of Lyn attenuates neutrophil wound recruitment. Inhibition of SFKs also disrupted H(2)O(2)-mediated chemotaxis of primary human neutrophils. In vitro analysis identified a single cysteine residue, C466, as being responsible for direct oxidation-mediated activation of Lyn. Furthermore, transgenic-tissue-specific reconstitution with wild-type Lyn and a cysteine mutant revealed that Lyn C466 is important for the neutrophil wound response and downstream signalling in vivo. This is the first identification, to our knowledge, of a physiological redox sensor that mediates leukocyte wound attraction in multicellular organisms.

Figure 1. SFKs mediate neutrophil wound response

a, Diagram of tail-transection in 3 dpf zebrafish larvae. b, Immunofluorescence of pSFK (phosphorylation of SFK activation loop tyrosine) in Tg(mpx:mCherry). mCherry labels neutrophils. c, Quantification of fluorescence intensity of pSFK in neutrophils (wound: 40 cells (5 larvae), no wound: 38 cells (7 larvae)). d, Immunofluorescence of pSFK in Tg(mpx:mCherry) with/without duox morpholino. e, RT-PCR of duox mRNA with/without duox morpholino (ef1α is a loading control) and quantification of fluorescence intensity of pSFK in neutrophils (ctrl: 110 cells (13 larvae), duox MO: 98 cells (19 larvae). f, Neutrophil recruitment to wounds and fins at 1h post wounding with/without PP2 (DMSO: 25 larvae, PP2: 17 larvae). g, Representative pictures of Sudan Black staining. h, H₂O₂ imaging with HyPer probe with/without PP2. i, Under-agarose assay using human neutrophils (n=3). Error bars indicate SEM. Asterisk, P<0.05, two-tailed unpaired t-test (c,e,f,i). Scale bars: 10 μm (b), 50 μm (d,h)

Figure 2. Lyn mediates neutrophil wound responses

a, RT-PCR of SFKs. c-fms and mpx are markers of macrophages and neutrophils respectively and ef1α is a loading control. b, In situ hybridization of mpx and lyn mRNA. Green arrowheads: pronephric ducts, purple arrows: neuromasts. c, RT-PCR of lyn with/without lyn morpholinos. d, Neutrophil wound recruitment with/without lyn morpholinos at 30 minutes post wounding (ctrl: 49 larvae, lyn MO1: 38 larvae, lyn MO2: 50 larvae). Pictures display Sudan Black staining. e, The percentage of neutrophils that accumulate at wounds during 30 minutes post wounding (14 larvae each). f, Neutrophil tracking for 25 minutes post wounding (29 cells, 6 larvae each). g, Neutrophil velocity during wound responses (29 cells, 6 larvae each). h, Immunofluorescence of pSFK in Tg(mpx:mCherry) with/without lyn morpholino. i, Quantification of fluorescence intensity of pSFK in neutrophils (ctrl: 39 cells (4 larvae), lyn MO: 36 cells (5 larvae)). Error bars indicate SEM. Asterisk, P<0.05, one-way ANOVA with Dunnett post-test (d) and two-tailed unpaired t-test (e,g,i). Scale bar: 50 μm.

Figure 3. H₂O₂ activates Lyn in a C466-dependent manner

a, Autophosphorylation of Lyn activation loop tyrosine with/without DPI in HEK293 cells. b, Quantification of a (n=4). c, Autophosphorylation of Lyn mutants in HEK 293 cells. Y506F is a constitutively active mutant as a positive control of pLyn. d, Quantification of c (n=3). e, In vitro kinase assay of Lyn WT and C466A with/without H₂O₂. f, Quantification of e (n=6). g, In vitro kinase assay of Lyn WT and C466A with/without magnesium. h, Quantification of g (n=5). Error bars indicate SEM. Asterisk, P<0.05, one-way ANOVA with Dunnett post test (b), Bonferroni post-test (d) and two-tailed unpaired t-test (f,h).

Figure 4. Lyn regulates neutrophil wound responses in a C466-dependent manner

a, Neutrophil wound recruitment with/without lyn morpholino in Tg(mpx:lyn-GFP) and wild-type clutchmates (Lyn-GFP(−)/ctrl: 38 larvae, Lyn-GFP(−)/lyn MO: 29 larvae, Lyn-GFP(+)/ctrl: 26 larvae, Lyn-GFP(+)/lyn MO: 30 larvae). b, Representative pictures of Sudan Black staining in a. c, Neutrophil wound recruitment with/without lyn morpholino in Tg(mpx:lyn C466A-GFP) and wild-type clutchmates (C466A-GFP(−)/ctrl: 41 larvae, C466A-GFP(−)/lyn MO: 41larvae, C466A-GFP(+)/ctrl: 34 larvae, C466A-GFP(+)/lyn MO: 32 larvae). d, Representative pictures of Sudan Black staining in c. Asterisk, P<0.05, one-way ANOVA with Bonferroni post-test (a,c).