Interactions between the myosin Dachs, the adaptor Dlish, and the palmitoyltransferase Approximated mediate Fat-Dachsous signaling

1. Problem Statement

The Drosophila protocadherins Fat (Ft) and Dachsous (Ds) regulate tissue growth via the Hippo pathway and Planar Cell Polarity (PCP) through heterophilic binding. Their intracellular domains (ICDs) signal by modulating the levels and localization of three key downstream effectors: the atypical myosin Dachs, the SH3-domain adaptor Dlish (Vamana), and the palmitoyltransferase Approximated (App).

While previous studies established that Ft and Ds ICDs alter the subapical cortical levels of these proteins, the precise mechanistic interactions between Dachs, Dlish, and App remained unclear. Key unresolved questions included:

• How are Dachs and Dlish tethered to the cell cortex in the absence of Ft/Ds signaling?
• Does App palmitoylate Dlish to anchor it to the membrane, or does App regulate Dachs directly?
• Does the Ft ICD regulate the complex by altering App levels/activity, or through a different mechanism?
• What is the reciprocal relationship between Dachs and Dlish regarding protein stability and localization?

2. Methodology

The authors employed a combination of in vivo genetics in Drosophila wing imaginal discs and in vitro biochemical assays:

• Genetic Manipulation: Generation of various mutant clones (e.g., fat, ds, dlish, dachs, app) and RNAi knockdowns. They utilized specific Gal4 drivers (hh-gal4, ap-gal4, en-gal4) to overexpress wild-type and mutant constructs in specific regions of the wing disc.
• Construct Engineering: Creation of novel transgenic constructs including:
  • Dachs lacking its C-terminal SH3-binding motif (DachsΔSH3bd).
  • Dlish lacking specific SH3 domains (DlishΔSH3-2).
  • Dlish with all cysteines mutated to serine (Dlish-all C to S) to abolish palmitoylation.
  • Dlish fused with a CAAX prenylation motif to force membrane localization independent of palmitoylation.
  • Truncated Dlish constructs to map binding domains.
• Localization & Phenotypic Analysis: Immunostaining of wing discs to visualize protein localization (subapical cortex vs. cytoplasm) and quantification of growth phenotypes (wing overgrowth/undergrowth) and PCP defects (crossvein spacing).
• Biochemical Assays:
  • Co-Immunoprecipitation (Co-IP): To test physical interactions between Dlish, Dachs, App, and Fat/Ds ICDs in S2 cells.
  • Acyl-Biotin Exchange (ABE) & Click Chemistry: To confirm and map the palmitoylation sites on Dlish.
  • In vitro Pull-downs: Using GST/MBP-tagged proteins to map binding domains between Dlish and App.

3. Key Contributions & Results

A. Dlish and Dachs Interaction is Essential for Cortical Localization

• Direct Binding: The study confirms that Dlish binds Dachs directly via Dlish's second SH3 domain and Dachs' C-terminal motif.
• Localization Requirement: Overexpression of DachsΔSH3bd (unable to bind Dlish) resulted in diffuse cytoplasmic localization and failed to induce overgrowth, whereas wild-type Dachs concentrated at the subapical cortex and induced overgrowth.
• Reciprocal Stability: While Dlish is required to concentrate Dachs at the cortex, Dachs is required to stabilize total Dlish protein levels. In dachs mutants, Dlish levels drop drastically in both the cytoplasm and the cortex, suggesting Dachs protects Dlish from degradation. Conversely, Dlish is not required for Dlish's own localization; overexpressed Dlish localizes to the cortex even in dachs mutants.

B. Dlish is Palmitoylated by App, and Lipidation is Critical for Function

• Palmitoylation Confirmation: Using ABE and click chemistry assays, the authors confirmed Dlish is palmitoylated at multiple cysteine residues.
• Functional Necessity: A Dlish-all C to S mutant (unable to be palmitoylated) failed to localize to the cortex, remained cytoplasmic, and could not rescue dlish mutant phenotypes (no overgrowth, no Dachs cortical concentration).
• Rescue by CAAX: Adding a CAAX prenylation motif to the Dlish-all C to S mutant restored membrane localization and overgrowth activity, proving that the loss of function in the cysteine mutant was due to the loss of lipidation/membrane anchoring, not structural disruption.
• App is Not Limiting: While App is required for Dlish palmitoylation, App levels are not the limiting factor in the pathway. Strong RNAi knockdown of App did not phenocopy dlish or dachs mutants, and small app null clones retained sufficient activity. This suggests App activity is permissive rather than instructive in this context.

C. Mechanism of Fat and Ds Signaling

• Fat's Role: The Ft ICD does not regulate the complex by changing App levels. Instead, Ft likely acts by destabilizing the Dlish-Dachs complex.
• Dlish as a Mediator: In dlish mutants, residual cortical Dachs exists but is insensitive to the loss of Fat or the presence of Ds. This indicates that Dlish is not just a scaffold for Dachs but is the essential mediator coupling Dachs to the regulatory effects of the Ft and Ds ICDs.
• Ds's Role: The stabilization of Dachs by the Ds ICD also requires Dlish. Without Dlish, Ds overexpression fails to accumulate or polarize Dachs.

D. Novel Models for Regulation

The authors propose a "Localization and Destabilization" model:

1. Palmitoylated Dlish anchors the Dachs-Dlish complex to the subapical cortex.
2. Fat ICD promotes the destabilization (likely via ubiquitination) of this cortical complex.
3. Loss of Fat stabilizes the complex, leading to high cortical Dachs/Dlish and overgrowth.
4. Loss of Dlish prevents cortical anchoring (Dachs becomes cytoplasmic) but also stabilizes total Dachs protein levels (by removing the target for degradation), though the cytoplasmic Dachs is less effective at driving growth.

4. Significance

• Mechanistic Clarity: This study resolves the hierarchy of the Fat/Ds signaling pathway, establishing that Dlish is the central adaptor that couples Dachs to the membrane and to the regulatory inputs of Ft and Ds.
• Post-Translational Regulation: It definitively proves that Dlish is a palmitoylated protein and that this lipidation is the primary mechanism for its cortical tethering, a mechanism that can be bypassed by artificial prenylation.
• Protein Stability vs. Localization: The work highlights a distinct regulatory mechanism where Dachs acts as a stabilizer for Dlish, contrasting with Dlish's role as a localizer for Dachs. This reciprocal regulation ensures the robustness of the signaling complex.
• Hippo Pathway Insights: By clarifying how Dlish couples Dachs to the Hippo pathway (via Ex and E3 ligases like Fbxl7/Slimb), the paper provides a more complete picture of how cell-cell adhesion signals are transduced into growth control, with implications for understanding human homologs (FAT4, DCHS1) involved in syndromes and cancer.

In summary, the paper demonstrates that App-mediated palmitoylation of Dlish is the critical switch for membrane localization, while Dlish serves as the essential adaptor that links Dachs to the regulatory effects of Fat and Ds, controlling both the localization and the stability of the signaling complex.