The DPP and EGFR pathways are jointly required for vein differentiation and reciprocally induce each other's ligands. (A–C) Pupal wings at 22 h APF stained for pMAD (green) and dpERK (red), with anterior (up) and distal (to the right), showing that both pMAD and dpERK are detected in the provein cells. 1–5, longitudinal veins L1–L5, and m, the wing margin. (D) Flip-out clones expressing EGFR^DN-interrupted veins L2, L3, and L4. (E) Expressing Put^ΔI under control of the engrailed–Gal4 (en–Gal4) driver, which is expressed only in the posterior compartment of the wing, caused loss of vein tissues. (F) An adult wing with truncated veins resulting from co-expressing TKV^QD and EGFR^DN. In contrast to the wing with clones expressing TKV^QD alone (Supplementary Figure S1A–C), few, if any, ectopic vein cells are observed (data not shown). (G–I) Ectopic EGFR signaling induces dpp expression. dpp–lacZ expression (magenta in H and I) is detected in the EGFR^λ-expressing clones. (J–L) TKV^QD activates EGFR by inducing rho expression.

Transient EGFR or DPP pathway activation leads to vein differentiation. (A) Model of vein differentiation, which depends on the joint activation of the EGFR and DPP systems and is mediated by positive feedback. (B) Phase plot showing three steady states marked by white circles (stable) or a black circle (unstable). The red and blue lines are the nullclines. (C) Depending on network parameters, transient perturbations (left panel) can lead to irreversible switching from an ‘off' to an ‘on' state (right panel). Parameters for (C): u_x=1, Δt=2, α=1, ɛ=1, and σ₁=σ₂=0.5. (D) Pupae were heat-shocked for 20 min at 37°C at 18 h APF. Adult wings (left) of indicated genotypes and a higher magnification of the boxed region (right) are shown. Transient expression of activated DPP signaling (TKV^QD) or activated EGFR signaling (DRAF^ΔN) by heat shock-induced GAL4, resulted in the formation of ectopic vein cells (discrete darker spots). Transient co-expression of a dominant-negative receptor, EGFR^DN or PUT^ΔI, suppressed the ectopic vein formation and disrupted the formation of endogenous veins (gap in L4). (E, F) Quantification of ectopic vein induction as a function of heat-shock duration of 18-h-APF pupae bearing hsp70–GAL4 and the indicated UAS transgenes.

Symmetry in EGFR and DPP signaling dynamics. (A) Plot of parameter space shows that bounded values for ɛ exist for a measured stimulus of duration, Δt, required for ensuring vein fate induction, given that ectopic activation of either pathway is sufficient. The shaded region includes values of Δt and the ratio of time scales, ɛ, that are consistent with symmetric bistable switching from transient activation of either of the two signaling pathways, X or Y. The cross-activation thresholds, σ₁=σ₂=0.5, the exogenous signal strength (u_x or u_y), and α are set to 1. (B) Plot of parameter space showing that for a given transient activation of time, ɛ, is bound and centered on 1, even for a large variation in the sensitivity ratio, α. (C) Pupae with transient expression of activated DPP signaling (TKV^QD) by hs–GAL4 were heat-shocked for 30 min at 37°C at 18 h APF. Representative pupal wings (left) are shown with pMAD staining of indicated pupal stages and with a higher magnification of the boxed region (right) between L4 and L5. (D) Representative 22.5-h APF pupal wing between L4 and L5 with pMAD and dpERK staining of pupae bearing hsp70–Gal4 and UAS–tkv^QD after 30-min heat-shock treatment at 18 h APF. (E, F) Quantification of ectopic pMAD-positive cell number in selected boxed areas at different pupal stages for (E) hsp70>tkv^QD (n=3 pupal wings) and (F) hsp70>dRaf^ΔN (n=3), which were heat-shocked at 37°C for 30 min at 18 h APF. (G, H) Quantification of ectopic dpERK and pMAD intensities at different pupal stages of a selected boxed area between L4 and L5 of 18 h APF pupae for (G) hsp>tkv^QD and (H) hsp70>dRaf^ΔN that were heat-shocked at 37°C for 30 min at 18 h APF (n=4).