Multiple TGF-β superfamily signals modulate the adult Drosophila immune response

Abstract

TGF-β superfamily signals play complex roles in regulation of tissue repair and inflammation in mammals [1]. Drosophila melanogaster is a well-established model for the study of innate immune function [2, 3] and wound healing [4-7]. Here, we explore the role and regulation of two TGF-β superfamily members, dawdle and decapentaplegic (dpp), in response to wounding and infection in adult Drosophila. We find that both TGF-β signals exhibit complex regulation in response to wounding and infection, each is expressed in a subset of phagocytes, and each inhibits a specific arm of the immune response. dpp is rapidly activated by wounds and represses the production of antimicrobial peptides; flies lacking dpp function display persistent, strong antimicrobial peptide expression after even a small wound. dawdle, in contrast, is activated by Gram-positive bacterial infection but repressed by Gram-negative infection or wounding; its role is to limit infection-induced melanization. Flies lacking dawdle function exhibit melanization even when uninfected. Together, these data imply a model in which the bone morphogenetic protein (BMP) dpp is an important inhibitor of inflammation following sterile injury whereas the activin-like dawdle determines the nature of the induced immune response.

Graphical abstract

Figure 1

dpp and daw Are Regulated by Immune Challenge (A and B) daw and dpp expression in wild-type flies following M. luteus infection or PBS injection, normalized to untreated controls. Times given are the interval between treatment and sample collection. (C and D) daw and dpp expression following M. luteus infection in Dif;Rel double mutants (Dif² cn bw;Rel^e20) and wild-type controls (repeated from A and B). Expression is normalized to untreated wild-type controls. (E) dpp expression following M. luteus infection or PBS injection in Rel mutants (Rel^e38/Rel^e20) and wild-type controls (repeated from B). Expression is normalized to untreated wild-type controls. (F) dpp and daw expression 3 hr after heat shock in flies carrying UAS-Tl^10b (w;UAS-Tl^10b.myc/tubulin-Gal80^ts;hs-Gal4/+) and driver-only controls (w;tubulin-Gal80^ts/+;hs-Gal4/+). Expression after heat shock is normalized to non-heat-shocked genotype controls. (G) daw expression following M. luteus infection in Tak1 mutants (Tak1¹) and wild-type controls (repeated from A). Means are shown ±SEM. Assays were performed by qRT-PCR, and expression was normalized to Rpl1. ^∗∗∗p < 0.001, ^∗∗p < 0.01, ^∗p < 0.05 by Mann-Whitney test. See also Figure S1.

Figure 2

dpp Suppresses AMP Expression (A) AMP expression following wounding in flies overexpressing Dpp and Gbb under the control of heat-shock Gal4 (tubulin-Gal80^ts/UAS-gbb;hs-Gal4/UAS-dpp) and in driver-only controls (tubulin-Gal80^ts/+;hs-Gal4/+). Flies were heat shocked for 30 min beginning 30 min after wounding, and RNA samples were collected 3 hr after wounding. (B) AMP expression in Mad knockdown flies (w¹¹¹⁸;UAS-Mad-IR/c564) and driver-only controls (w¹¹¹⁸;c564/+) following PBS injection. Means are shown ±SEM. Expression is normalized to untreated driver-only controls. Assays were performed by qRT-PCR, and expression was initially normalized to Rpl1. ^∗∗∗p < 0.001, ^∗∗p < 0.01, ^∗p < 0.05 by Mann-Whitney test. See also Figure S2.

Figure 3

Dawdle Signals via the Activin Pathway to Suppress Melanization (A) Melanotic tumors in daw knockdown flies (w;UAS-daw-IR/+;tubulin-Gal4/+). Two independent inverted repeat (IR) lines are shown (VDRC13420 and VDRC105309). Tumors are indicated by arrows. (B) Sp7 expression 5 days after eclosion in daw knockdowns and driver-only controls. (C) Sp7 expression in flies overexpressing daw or activated babo in adult fat body (w;UAS-daw/c564;tubulin-Gal80^ts/+ or c564/+;UAS-act.babo/tubulin-Gal80^ts) relative to controls (c564/+;tubulin-Gal80^ts/+). Animals were untreated or collected 6 hr postinjection with PBS or mixed E. coli and M. luteus. Expression is normalized to untreated driver-only controls. (D) daw expression following Listeria infection of wild-type flies. Expression is normalized to day 0 untreated levels. (E) Survival of daw- and activated babo-expressing flies following Listeria infection, relative to driver-only controls. (Lines labeled “OE” correspond to overexpressors.) Survival of both misexpression lines is different from controls (p < 0.001). For qRT-PCR assays in (B)–(D), expression was initially normalized to Rpl1, and means are shown ±SEM. ^∗∗∗p < 0.001, ^∗∗p < 0.01, ^∗p < 0.05 by Mann-Whitney test. See also Figure S3.

Figure 4

dpp and dawdle Are Expressed in Subsets of Phagocytes (A–T) Dorsal thoraxes of flies injected with Alexa Fluor 488-labeled S. aureus. (A–E) HmlΔ-dsRed flies (w;HmlΔdsRed.nuc). (F–J) crq>cherry flies (w¹¹¹⁸;;crq-Gal4,UAS-mCD8-cherry/+). (K–O) dpp>mRFP flies (w;UAS-myr.mRFP/+;dpp.blk1.40C.6-Gal4/+). (P–T) daw>mRFP flies (w;UAS-myr.mRFP/daw^NP4661). (A, F, K, and P) Bright field of the region imaged in each row. (B, G, L, and Q) Maximum projection, Alexa Fluor 488 S. aureus-labeled phagocytes (green). (C, H, M, and R) Maximum projection, RFP (magenta). (D, I, N, and S) Merge of red and green showing overlap (white) between Alexa Fluor 488 S. aureus and RFP. (E, J, O, and T) Close-up of the indicated region of the previous image (yellow square), taken from a single focal plane to clarify the overlap between S. aureus and marker expression. Example double-labeled cells are indicated with white arrowheads. (U) Proportions of cells showing green fluorescence (Alexa Fluor 488 S. aureus), red fluorescence (RFP), or both. Top: these subsets as proportions of the total number of labeled cells. Bottom: proportions of RFP⁺ cells that were also AF488⁺/AF488⁻. Counts were taken from single images covering the entire dorsal thorax and abdomen. Full counts are given in Table S1. (V) Top and bottom right: fluorescence-activated cell sorting strategy for cells marked by HmlΔ>eGFP (w¹¹¹⁸;HmlΔ-Gal4,UAS-2xeGFP/+). Gating (clockwise from top left): debris exclusion, doublet exclusion, eGFP expression. Bottom left: expression of Hml, crq, dpp, and daw in this sample (20,000 GFP⁺ cells) and purity analysis for this sample. See also Figure S4 and Table S1.