Protein phosphatase 4 coordinates glial membrane recruitment and phagocytic clearance of degenerating axons in Drosophila

Abstract

Neuronal damage induced by injury, stroke, or neurodegenerative disease elicits swift immune responses from glial cells, including altered gene expression, directed migration to injury sites, and glial clearance of damaged neurons through phagocytic engulfment. Collectively, these responses hinder further cellular damage, but the mechanisms that underlie these important protective glial reactions are still unclear. Here, we show that the evolutionarily conserved trimeric protein phosphatase 4 (PP4) serine/threonine phosphatase complex is a novel set of factors required for proper glial responses to nerve injury in the adult Drosophila brain. Glial-specific knockdown of PP4 results in reduced recruitment of glia to severed axons and delayed glial clearance of degenerating axonal debris. We show that PP4 functions downstream of the the glial engulfment receptor Draper to drive glial morphogenesis through the guanine nucleotide exchange factor SOS and the Rho GTPase Rac1, revealing that PP4 molecularly couples Draper to Rac1-mediated cytoskeletal remodeling to ensure glial infiltration of injury sites and timely removal of damaged neurons from the CNS.

Figure 1

The PP4 complex subunits Falafel, PP4c, and PP4r2 are required for glial clearance of degenerating axons. (a–d) Representative maximum intensity projection confocal images (z-stack, 15 μm) show green fluorescent protein (GFP)-labeled OR85e axonal projections (green) in antennal lobes of uninjured (a and c) and injured (b and d) adult flies. (e) Quantification of OR85e GFP fluorescence from experiment in panels (a–d) normalized to uninjured; mean±S.E.M. plotted; one-way analysis of variance (ANOVA). ***P<0.001. (f–k) Representative maximum intensity projection images (z-stack, 15 μm) shown in antennal lobes of uninjured (f, h, and j) and injured (g, i and k) adult flies. (l) Quantification of OR85e axonal debris (GFP fluorescence) from experiment in panels (f–k) normalized to uninjured conditions; mean±S.E.M. plotted; one-way ANOVA. **P<0.01. (m–n) Representative antennal lobe regions (z=3 μm) show cells stained with anti-Falafel (magenta) and anti-Repo (glia nuclear marker, green) in control (m) and Falafel RNAi (n). Falafel fluorescence intensity in Repo+ glia nuclei was quantified from the entire central brain. See values (white font) in panels (m″ and n″). Unpaired t-test. P<0.0001. Scale bars=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. Falafel RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/UAS-falafel^RNAi. RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/+. PP4r2 RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4r2^RNAi; repo-Gal4/+

Figure 2

PP4 is required for proper recruitment of Draper to degenerating axons. (a–h) Representative single z-slice (1 μm) show anti-Draper fluorescence in one antennal lobe of uninjured (a, c, e, and g) and injured (b, d, f and h) adult flies. White dotted outlines show OR85e glomeruli (visualized with OR85e-mCD::GFP, not shown) and representative areas of Draper fluorescence quantification. (i) Draper fluorescence quantified in z-stack of 15 μm, normalized to uninjured conditions. Uninjured Draper set at a value of 1; mean±S.E.M. plotted; one-way analysis of variance (ANOVA). ****P<0.0001. Scale bars=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. Falafel RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/UAS-falafel^RNAi. PP4c RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/+. PP4r2 RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4r2^RNAi; repo-Gal4/+. GFP, green fluorescent protein

Figure 3

PP4 is necessary for glial membrane expansion and recruitment to severed axons after nerve injury. (a–d) Representative antennal lobe z-stacks (3 μm) show glial membrane RFP in gray scale in uninjured (a and c) and injured (b and d) adult flies. (b′ and d′) Zoomed in view of white boxed regions in (b) and (d). (e) Quantification of RFP+ glial membrane fluorescence, normalized to uninjured condition. Uninjured RFP fluorescence set to 1; N: control: 4, control injured: 6, PP4c: 18, PP4c injured: 24; individual data points with mean±S.E.M. plotted; one-way analysis of variance (ANOVA). **P<0.01. (f–i) Representative antennal lobe z-stacks (15 μm) show RFP-labeled glial membranes (gray scale) and OR85e (green) in uninjured (f and h) and injured (g and i) adult flies. (f′, g′, h′, and i′) Zoomed in view of blue boxed regions in (f–i) show glial membrane accumulation on severed maxillary nerves in control animals (g′). (j) RFP membrane fluorescence quantified, normalized to uninjured condition. Uninjured RFP fluorescence set to 1; individual data points with mean±S.E.M. plotted; one-way ANOVA. ****P<0.001. Scale bars=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+, repo-Gal4/repo-LexA, LexAop-mCD2::RFP. PP4c RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/PP4c^RNAi; repo-Gal4/repo-LexA, LexAop-mCD2::RFP. GFP, green fluorescent protein; NS, nonsignificant

Figure 4

PP4c overexpression partially reverses axon clearance defects in draper^RNAi animals. (a–h) Representative maximum intensity projection confocal images (z-stack, 15 μm) show green fluorescent protein (GFP)-labeled OR85e axonal projections (green) in antennal lobes of uninjured (a–d) and injured (e–h) control (a and e), Draper RNAi alone (b and f), PP4cHA alone (cand g), Draper RNAi+PP4cHA (d and h). (i) Quantification of OR85e axonal debris (GFP) normalized to uninjured conditions. Uninjured GFP fluorescence values set at 1. Individual data points with mean±S.E.M. plotted; one-way analysis of variance (ANOVA). *P<0.05, ***P<0.001, and ****P<0.0001. Scale bar=20 μm. Genotypes: control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. Drpr RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-draper^RNAi; repo-Gal4/UAS-LacZ. PP4cHA=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-LacZ; repo-Gal4/UAS-PP4cHA. Drpr RNAi+PP4cHA=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-draper^RNAi; repo-Gal4/UAS-PP4cHA. NS, nonsignificant

Figure 5

Injury-induced activation of STAT92E in ensheathing glia does not require PP4c. (a–d) Expression of the STAT92E transcriptional reporter 10XSTAT92E-dGFP. Representative z-stacks (10 μm) show anti-green fluorescent protein (GFP) (green) and anti-Draper fluorescence (magenta) in one antennal lobe of uninjured (a and c) and injured (b and d) adult flies. (e) Quantification of Draper, normalized to uninjured conditions. Uninjured Draper set at a value of 1. (f) Quantification of dGFP levels, normalized to uninjured conditions. Uninjured dGFP set at a value of 1. Individual data points with mean±S.E.M. plotted; one-way analysis of variance (ANOVA), ****P<0.0001. Scale bar=20 μm. Genotypes: Control=10XSTAT92E-dGFP,tub-Gal80^ts/+ repo-Gal4/+. PP4c RNAi= 10XSTAT92E-dGFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/+. NS, nonsignificant

Figure 6

Forced glial expression of Rac1 rescues axonal clearance and Draper recruitment defects in PP4 knockdown flies. (a–f) Representative single z-slice (1 μm) show anti-Draper fluorescence in one antennal lobe of uninjured (a, c, and e) and injured (b, d, and f) adult flies. Yellow dotted outlines show representative areas of Draper fluorescence quantified in OR85e glomeruli. (g) Draper fluorescence quantification in z-stack of 15 μm, normalized to uninjured conditions. Uninjured Draper set at a value of 1; mean±S.E.M. plotted; one-way analysis of variance (ANOVA), ***P<0.001. (h–m) Representative maximum intensity projection confocal images (z-stack, 15 μm) show green fluorescent protein (GFP)-labeled OR85e axonal projections (green) in antennal lobes of uninjured (h, j, and l) and injured (I, k, and m) adult flies. (n) GFP fluorescence quantification, normalized to uninjured condition. Uninjured GFP fluorescence set to 1; mean±S.E.M. plotted; *P<0.05 and ****P<0.0001. Scale bars=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. PP4c RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/UAS-LacZ. PP4c RNAi+UAS-Rac1=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/UAS-Rac1. NS, nonsignificant

Figure 7

PP4 is an upstream effector of SOS and Rac1-mediated cytoskeletal remodeling. (a–h) Representative confocal images of Draper staining in single antennal lobes of control (a and b), PP4c RNAi (c and d), UAS-SOS (e and f), and PP4c RNAi+UAS-SOS (g and h) uninjured and injured brains. Single 1 μm slice shown. Dotted lines show region of interest (OR85e glomerulus) for Draper quantification. (i) Quantification of Draper fluorescence from experiment shown in (a–h), on 15 μm Z-stack, normalized to uninjured conditions. Mean±S.E.M. plotted; one-way analysis of variance (ANOVA), Sidak's comparison test. **P<0.01, ****P<0.0001. Scale bar=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80ts/+ repo-Gal4/+. PP4c RNAi=OR85e-mCD8::GFP,tub-Gal80ts/UAS-PP4c^RNAi; repo-Gal4/UAS-LacZ. SOS=OR85e-mCD8::GFP,tub-Gal80ts/UAS-SOS-Myc; repo-Gal4/+. PP4c RNAi+SOS=UAS-SOS-Myc/UAS-PP4c^RNAi; repo-Gal4/tub-Gal80ts. (j–q) Representative confocal images of anti-Rac1 in single antennal lobes of control (j, l, n, and p) and PP4c RNAi (k, m, o, and q) uninjured (j–m) and injured (n–q) adult flies. Merge panels (j, k, n, and o) show anti-Rac1 (gray scale) and OR85e glomeruli (green). (l, m, p, and q) show the Rac1 channel alone. (r) Quantification of basal anti-Rac1 levels from cortical areas. (s) Quantification of anti-Rac1 fluorescence in region of interest (dotted line around 85e glomeruli) Z-stack of 15 μm. Mean±S.E.M. plotted; one-way ANOVA. *P<0.05, Scale bars=20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. PP4c RNAi=OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/+. (t–w) Phalloidin F-actin (gray scale) staining on control and PP4c RNAi brains, uninjured (t and v) and 1 day after antennal nerve injury (u and w). Representative antennal lobes shown; 20 μm z-stacks. (x) Phalloidin fluorescent intensity quantification. Dotted outline in (t) shows representative area of quantification (single antennal lobe). Mean±S.E.M. plotted; one-way ANOVA. *P<0.05 and ****P<0.0001. Scale bar: 20 μm. Genotypes: Control=OR85e-mCD8::GFP,tub-Gal80^ts/+ repo-Gal4/+. OR85e-mCD8::GFP,tub-Gal80^ts/UAS-PP4c^RNAi; repo-Gal4/+. GFP, green fluorescent protein; NS, nonsignificant

Figure 8

Nuclear levels of Falafel decrease in ensheathing glia after axon injury. (a–c) Representative confocal images of antennal lobe regions. Brains were stained with anti-Falafel and with anti-β-gal to visualize ensheathing glial nuclei. Dotted squares in (a–c) outline higher magnification images in (a′, a″, b′, b″, c′, and c″). White boxed regions in (a″–c″) show Falafel fluorescence in isolated glial cells. Arrows identify representative glial cells that were quantified. (d) Quantification of Falafel fluorescence in ensheathing glial nuclei. (e–h) Representative images of Falafel fluorescence in cortex glia, identified by anti-β-gal expression. (i) Quantification of Falafel fluorescence in cortex glial nuclei. (j–k) Representative images of Falafel fluorescence in astrocyte nuclei. Eleven micrometer of z-stacks. Mean±S.E.M. plotted; one-way analysis of variance (ANOVA). *P<0.05. Scale bar: 20 μm. Genotypes: (a–d) UAS-LacZ::NLS; TIFR-Gal4. (e–i) UAS-LacZ::NLS; NP2222-Gal4. (j and k) UAS-LacZ::NLS; alrm-Gal4. NS, nonsignificant