Schnurri regulates hemocyte function to promote tissue recovery after DNA damage

Abstract

Tissue recovery after injury requires coordinated regulation of cell repair and apoptosis, removal of dead cells and regeneration. A critical step in this process is the recruitment of blood cells that mediate local inflammatory and immune responses, promoting tissue recovery. Here we identify a new role for the transcriptional regulator Schnurri (Shn) in the recovery of UV-damaged Drosophila retina. Using an experimental paradigm that allows precise quantification of tissue recovery after a defined dose of UV, we find that Shn activity in the retina is required to limit tissue damage. This function of Shn relies on its transcriptional induction of the PDGF-related growth factor Pvf1, which signals to tissue-associated hemocytes. We show that the Pvf1 receptor PVR acts in hemocytes to induce a macrophage-like morphology and that this is required to limit tissue loss after irradiation. Our results identify a new Shn-regulated paracrine signaling interaction between damaged retinal cells and hemocytes that ensures recovery and homeostasis of the challenged tissue.

Fig. 1.

shn gene dose influences JNK-induced tissue loss in the Drosophila retina. (A) Constitutive activation of the stress-responsive Jun N-terminal kinase (JNK) pathway in the photoreceptor and cone cells by Sep-Gal4 driving constitutively active Hep (abbreviated SeP>Hep^act) results in excessive apoptosis and disruption of the adult compound eye. This phenotype can be enhanced by loss of shn through mutant alleles or RNAi. Overexpression of endogenous shn using an EP line can rescue the eye to almost a wild-type phenotype. (A′) Tissue loss is represented as the fraction of the eye lacking ommatidia (‘cleared’ area, as indicated by the dotted line in A, divided by total eye area). shn mutants were compared with their wild-type siblings. Error bars indicate s.d.; P-values from Student's t-test. (B) In response to UV stress, the JNK cascade is activated through a series of phosphorylation reactions and ultimately activates transcription of target genes. This can lead to cell death when signaling through Hid. Activation of JNK is visualized by puc-lacZ staining at 27 hours after puparium formation (P27) pupa after UV treatment at P24. (C) Representative examples are shown of each genotype following UV irradiation of the left eye. Overexpression of Shn by the eye driver GMR-Gal4 reduces UV-triggered tissue loss in the retina, whereas knockdown of Shn by RNAi enhances the loss of tissue due to UV irradiation. Both RNAi and overexpression lines show a greater effect at 29°C. (C′) Tissue loss was quantified by calculating the size ratio of UV-treated eye to untreated eye. Error bars indicate s.e.m.; P-values from Student's t-test.

Fig. 2.

Activation of retina-associated hemocytes by Shn. (A) Real-time PCR of dissected larval eye discs showing induction of puc, but not hid, transcript levels (expression relative to Actin 5C transcript is shown). Error bars indicate s.d.; P-values from Student's t-test. (B) Accumulation of puc-expressing hemocytes in third instar eye discs overexpressing Shn in the retina, as visualized by the lacZ reporter puc^E69. Hemocytes are a cluster of round cells that are much larger than imaginal cells. (B′) Quantification of puc-expressing hemocytes in eye discs. Error bars indicate s.e.m.; P-values from Student's t-test. (C) Confocal images confirming the identity of puc-lacZ-positive cells as hemocytes. Eye disc-associated hemocytes were identified by the expression of RFP from an Hml promoter (red), while Shn was expressed using the ShnEP insertion in the retina using GMR-Gal4 in a puc^E69 background (right panels; eye discs without Shn expression are shown on the left). Scale bars shown in the first two panels are representative for these panels and for the two panels on the right. β-galactosidase expression was identified by immunostaining (green in top row and white in bottom row). Arrowheads point to clusters of hemocytes. (D) Retina-associated hemocytes change morphology in pupal stages. In the larva, hemocytes are compact and round, whereas by 24 hours after puparium formation they display a spread phenotype and many DNA punctae, indicating phagocytic activity. The left two panels are confocal z-stacks; the magnified images (right two panels) show only one focal plane. (E) In irradiated animals, puc-expressing hemocytes are seen at 27 hours after puparium formation, both attached to the retina and distal from the area of UV treatment (arrows point to several examples). By contrast, very few puc-expressing hemocytes are seen in mock-treated pupae. (E′) Quantification of puc-expressing hemocytes visible in P27 pupae, 3 hours after UV treatment. Error bars indicate s.e.m.; P-values from Student's t-test.

Fig. 3.

Phagocytosis in hemocytes is required to limit tissue loss after UV irradiation. (A) Hemocytes were ablated using the hemocyte driver Hml-Gal4 driving the apoptotic protein Hid. In these animals, tissue loss was severe following UV irradiation compared with wild-type siblings. (A′) Tissue loss was quantified by calculating the ratio of UV-treated eye size to untreated eye size of the same animal and then compared with siblings. Error bars indicate s.e.m.; P-values from Student's t-test. (B,B′) Expression of mbc RNAi under the control of Hml-Gal4 also increases sensitivity of the retina to UV irradiation. Error bars indicate s.e.m.; P-values from Student's t-test.

Fig. 4.

Pvf1/PVR signaling regulates hemocyte activity and tissue maintenance downstream of Shn. (A) Transcript levels of the VEGF ligand Pvf1 are elevated in larval eye discs overexpressing Shn by Sep-Gal4. Error bars indicate s.d.; P-values from Student's t-test. (B) Overexpressing PVR in hemocytes using the hemocyte driver Pxn-Gal4 is sufficient to promote early spreading of hemocytes in larval stages. (C) Knockdown of PVR by RNAi in the hemocytes does not diminish the hemocyte population. (D) Knockdown of PVR using RNAi prevents the spread phenotype typically seen at pupal stages. (B′,D′) Size of individual hemocytes. Error bars indicate s.e.m.; P-values from Student's t-test. (E) Tissue loss following UV irradiation is enhanced in animals with hemocyte-specific knockdown of PVR. (F) Overexpression of Pvf1 in the retina is sufficient to protect from UV-mediated tissue loss. (G) Overexpression of Pvf1 in the retina can compensate for Shn loss in modulating tissue recovery after UV irradiation. (E′, F′,G′) Tissue loss was quantified by calculating the size ratio of UV-treated eye to untreated eye, and then compared with siblings. Error bars indicate s.e.m.; P-values from Student's t-test. (H) Model for the role of Shn in tissue maintenance. Shn regulates the expression of pvf1 and egr to activate retina-associated hemocytes, promoting their transition into a macrophage-like morphology. This interaction promotes tissue recovery.