A spared synaptic branch restrains Wnd-dependent injury signaling in SNc motoneurons.

A) Schematic representation of the two SNc motoneurons innervating muscles 26 and 29 (MNSNc-26/29, red) and muscle 27 (MNSNc-27, blue), which are labeled by expression of the m12-Gal4 driver.

B) Example images of NMJ terminals from m12-Gal4/+; UAS-BitBow2(Li et al., 2021) /+ third instar larvae, used to define the connectivity shown in A. The neuron that innervates muscle 27 (MNSNc-27) expresses a distinct set of colors from the Bitbow2(Li et al., 2021) reporter than the neuron that innervates muscles 26 and 29 (MNSNc-26/29).

C) Example NMJs (top row) and cell bodies (bottom row) of MNSNc neurons (from m12-Gal4, UAS-mCD8GFP/puc-lacZ larvae) following injury at a location upstream of all of the synaptic branches, indicated by the arrow. The left column shows uninjured wild type animals, the middle column shows wild type animals 24h post injury. The far right column is from an animal that co- expresses UAS-wnd-RNAi. LacZ expression (red channel) in MNSNc nuclei indicates the induction of Wnd-dependent signaling in the injured neurons.

D) Example MNSNc neurons injured at different locations. The top row shows lucida drawings of the NMJ terminals in the middle row, while the bottom row shows puc-lacZ expression (red), in GFP(gray)-expressing cell bodies (from m12-Gal4, UAS-mCD8GFP/puc-lacZ larvae).

E) Quantification of puc-lacZ intensity measurements in MNSNc-26/29 (blue) and MNSNc- 27(red) following injuries that remove all synaptic branches versus injuries that leave a spared synaptic branch. A one-way ANOVA with Tukey test for multiple comparisons was performed for each neuron. **** p < 0.0001; ** p = 0.0011; comparisons not denoted were not significant.

Restraint of Wnd-mediated injury signaling by spared branch in bifurcated neurons.

A) Cartoon of Ventral Unpaired Motoneurons (VUM), which have bifurcated axons that symmetrically innervate body wall muscles on both the left and right sides of the animal. Nerve crush to either left or right side of the animal can axotomize a single bifurcation while leaving the other bifurcated axon intact.

B) Example images of VUM axons (visualized in Tdc2-Gal4, UAS-mCD8-GFP larvae) in segmental nerves on the uninjured and injured sides following nerve crush to a single side.

C) Example images of puc-lacZ expression in the VNC (ventral nerve cord) of larvae following nerve crush to a single side (half crush) versus crush to all the segmental nerves (full crush). puc-lacZ expression (red) is induced in VUM neurons (white) only after full crush. In contrast, other motoneurons, which innervate a single side, are induced by both half and full crush injuries. Co-expression of UAS-wnd-RNAi in VUM neurons cell autonomously inhibits puc-lacZ induction.

D) Quantification of puc-lacZ intensity measurements in VUM neurons. A one-way ANOVA with Tukey test for multiple comparisons was performed. **** p < 0.0001. Scale bars = 20 μm.

Presence of spared synaptic branch restrains Wnd signaling independently of Hiw.

A) Laser axotomy is carried out to MNSNc neurons at a location that completely removes the synaptic terminal of MNSNc-27 (red neuron). The injury also leads to loss of the MNSNc-26/29 (blue) terminal on muscle 29 but not 26 (hence leaves a spared synaptic branch. The final column shows an axotomy that fully removes the terminals for both MNSNc neurons. These injuries were repeated in control animals versus the background of a hiw null mutant, hiwΔN.

B) Quantification of puc-lacZ expression for individual MNSNc neurons after full versus spared axotomies, compared to uninjured neurons. Basal puc-lacZ expression is already elevated in uninjured hiwΔN neurons compared to control. This can be further elevated in axotomies that remove all synapses, but not in axotomies that leave spared branches.

C) Quantification of puc-lacZ in VUM neurons (labeled by Tdc-2-Gal4; UAS-mCD8-GFP) 24h and 48h following full nerve crush in control versus hiwΔN mutants. A two-way ANOVA with Tukey test for multiple comparisons was performed. **** p < 0.0001; *** p < 0.001; **p<0.01.

Potential mechanisms for regulation of Wnd signaling from synaptic terminals.

In green, Wnd signaling may be regulated in the cell body downstream of a retrogradely transported signal. (For example, neurotrophin signaling). In blue, Wnd signaling activation is restrained locally at synaptic terminals, perhaps by regulating the levels or activation of Wnd itself. Activated Wnd or a downstream signaling factor is then retrogradely transported to the cell body. Previous observations that inhibition of retrograde transport blocks the induction of Wnd signaling following axonal injury favors the latter (blue) possibility. However the restraint conferred by a physically separate bifurcation suggests that an inhibitor of Wnd signaling activation can be retrogradely transported (green). We speculate that both mechanisms act as dual checkpoints to restrain Wnd signaling activation in the context of healthy circuits.

Further characterization of branched injury assays to SNc and aCC motoneurons.

A) cartoon denoting nomenclature of MNSNc-26/29 (red) and MNSNc-27 (blue) branches.

B) Example images of nerve terminals and cell bodies of m12-Gal4; UAS-Bitbow2 labeled MNSNc neurons, used to confirm the anatomy.

C) Similarly to laser axotomy in Figure 1C, nerve crush injury (24h) induces puc-lacZ expression in both MNSNc neurons, but not in MNSNc neurons that co-express wnd-RNAi.

D) Synaptic terminals (top row) and cell bodies (bottom row) of aCC motoneurons on muscle 1, labeled in Dpr4-Gal4, UAS-mCD8-GFP larvae. puc-lacZ expression (red) is induced following injuries that result in loss of all synaptic boutons but not following injuries to one branch that leave the other branch intact.

E) Quantification of puc-lacZ intensities in aCC neurons. A one-way ANOVA with Tukey test for multiple comparisons was performed. **** p < 0.0001.

F-G) Full (F) but not partial (G) removal of synaptic branches induces stability and trafficking ectopically expressed kinase-dead GFP-Wnd-KD (in UAS-GFP-Wnd-KD; m12-Gal4, UAS- mCD8-RFP animals). Synaptic branches from muscle 27 (F), or muscle 29 (G) were axotomized by laser surgery and imaged following 24h.

F) GFP-Wnd-KD protein accumulates at the proximal tip of axons that have lost all synaptic boutons.

G) GFP-Wnd-KD is barely detectable in axons following injuries that leave spared synaptic branches. Not shown, GFP-Wnd-KD levels in G are similar in uninjured MNSNc axons.

Anatomy and laser surgery of Tdc2 bifurcated neurons.

A-E) Views of Tdc2-Gal4, UAS-mCD8-GFP expressing VUM neurons to illustrate their anatomy.

A) Individual confocal plane that shows 3 cell bodies in individual segments, which lie in the middle of the nerve cord. B) Side view and C) top view that show the locations of the bifurcations.

D) Cartoon of the 3 neurons from one segment, showing their bifurcations to symmetrical sides of the animal.

E) composite and camera lucida views of the NMJ terminals for the 3 neurons on one abdominal hemisegment. The 3 Tdc2 neurons each form stereotyped branches to innervate a unique group of muscles.

F) Example results from laser axotomies to individual Tdc2/VUM neurons (24h after injury) on either one side or both sides of the animal. Surgeries were carried out at the indicated locations which lie upstream of the final synaptic branches (at the transition zone between the segmental nerve and abdominal muscles). The stereotyped anatomy allows for identification of each VUM neuron (labeled 1,2 and 3). Injured branches are marked with an asterisk while spared are marked with squares. For each neuron, only injuries to both bifurcations allowed for induction of puc-lacZ. Scale bars = 20 μm.

Confirmation that elevated puc-lacZ expression in hiw mutants requires Wnd.

A) Top row shows puc-lacZ expression (red) together with a marker for nuclei (Draq5, gray). Control (lexA)-RNAi or wnd-RNAi UAS lines are driven by the BG380Gal4 driver in the background of control versus hiwΔN. Bottom row shows example NMJs in these genotypes. Scale bars = 20 μm.

B) Quantification of puc-lacZ expression in A. A one-way ANOVA with Tukey test for multiple comparisons was performed. **** p < 0.0001.

Summary of negative results from genetic manipulations that impair synaptic transmission and/or signaling at synapses.

The listed UAS lines, UAS-RNAi lines and genetic mutations were tested for their ability to alter the expression of puc-lacZ in motoneurons. Outcomes and experimental details are noted.