FRAT2

FRAT2 is a positive regulator of the canonical Wnt/β-catenin–TCF signaling pathway that functions by binding GSK-3β and Dishevelled to modulate GSK-3β substrate specificity and downstream signal transduction. FRAT2 contains a GSK-3β binding domain identical to that of FRAT1, and its acidic and proline-rich domains are required for axis-inducing activity; it selectively enhances GSK-3β-mediated phosphorylation of primed substrates such as tau while being itself phosphorylated by GSK-3β (PMID:11237732, PMID:15522877). Beyond canonical Wnt signaling, FRAT2 activates Rac GTPase through a GSK-3/Dishevelled-dependent mechanism downstream of MLL fusion oncogenes in hematopoietic progenitors, linking it to leukemogenesis and chemosensitivity (PMID:23074275). FRAT2 physically interacts with FRAT1 to cooperatively drive invasion and is negatively regulated at the protein level by TMEM98 and at the mRNA level by miR-29 family members and miR-3648, establishing multiple feedback controls on Wnt pathway output (PMID:36153370, PMID:31961879, PMID:25605017).

1. 2001 High

   Establishing that FRAT2 is a functional Wnt pathway activator whose activity depends on its acidic and proline-rich domains resolved the question of whether FRAT2 — newly cloned alongside FRAT1 — shared its paralog's axis-inducing capacity.

   Evidence Xenopus axis duplication assay comparing wild-type and domain-deletion mutant FRAT2 mRNA injections

   PMID:11237732

   Open questions at the time

   • Endogenous expression pattern and physiological requirement for FRAT2 in vertebrate Wnt signaling not addressed
   • Mechanism by which the acidic/proline-rich domains contribute to signaling undefined
2. 2002 Medium

   Demonstrating that FRAT2 directly binds both GSK-3 and Dishevelled placed it at the core signal transduction node where Wnt signals converge, answering whether FRAT2 engages the same molecular partners as FRAT1.

   Evidence Protein interaction studies and subcellular fractionation in transiently transfected COS-1 cells

   PMID:12095675

   Open questions at the time

   • Binding shown only with overexpressed proteins; endogenous interaction not confirmed
   • Relative affinities for GSK-3 versus Dishevelled not determined
3. 2004 High

   Two independent studies resolved FRAT2's enzymatic relationship with GSK-3β: FRAT2 is a weaker Wnt–TCF activator than FRAT1 yet selectively enhances GSK-3β phosphorylation of primed substrates (e.g., tau) and is itself a GSK-3β substrate, suggesting a non-redundant modulatory role.

   Evidence Co-immunoprecipitation, β-catenin/TCF reporter assays, in vitro kinase assays with recombinant proteins, and phosphorylation detection

   PMID:15073180 PMID:15522877

   Open questions at the time

   • The basis for differential potency between FRAT1 and FRAT2 in Wnt activation is unknown
   • Functional consequence of FRAT2 phosphorylation by GSK-3β on its activity or stability not determined
   • Whether FRAT2 enhances primed phosphorylation of endogenous substrates in vivo is untested
4. 2012 High

   Genetic loss-of-function experiments revealed a Wnt-independent output of FRAT2: activation of Rac GTPase downstream of MLL fusion oncogenes via GSK-3 and Dishevelled, explaining how FRAT2 contributes to leukemic transformation and chemoresistance.

   Evidence Frat-knockout hematopoietic progenitors transformed by MLL fusions, Rac activity assays, epistasis analysis with GSK3/DVL perturbation

   PMID:23074275

   Open questions at the time

   • Direct mechanism linking FRAT2–GSK3–DVL to Rac GEF activation unresolved
   • Relative contributions of FRAT1 versus FRAT2 in MLL-driven Rac activation not separated
5. 2015 Medium

   Identification of miR-29 family members as direct negative regulators of FRAT2 in chondrocytes and pancreatic cancer cells established a post-transcriptional control layer that tunes Wnt pathway output, with TGF-β acting upstream to relieve miR-29-mediated FRAT2 repression.

   Evidence Luciferase reporter assays validating direct miR-29 targeting of FRAT2 3′-UTR; miR-29c overexpression/knockdown in pancreatic cancer cells

   PMID:25605017 PMID:26687115

   Open questions at the time

   • Quantitative contribution of FRAT2 repression versus co-targeted Wnt components (LRP6, FZD4/5) not dissected
   • In vivo relevance of miR-29–FRAT2 axis in cartilage or pancreatic homeostasis not tested
6. 2020 Medium

   Discovery that TMEM98 physically interacts with FRAT2 and reduces its protein levels provided the first evidence of a protein-level negative regulator that constrains FRAT2-mediated β-catenin/TCF signaling.

   Evidence Co-immunoprecipitation, TMEM98 overexpression with FRAT2 protein quantification, β-catenin/TCF reporter assays

   PMID:31961879

   Open questions at the time

   • Mechanism of FRAT2 protein reduction (degradation versus translational inhibition) by TMEM98 unknown
   • Endogenous co-expression and interaction of TMEM98 and FRAT2 not demonstrated
   • Single-lab finding without independent replication
7. 2022 High

   Demonstration that FRAT1 and FRAT2 physically interact and cooperatively promote invasion — with FRAT2 knockdown reversing FRAT1-driven invasiveness — established a functional heteromeric relationship and identified a miR-3648/FRAT1-FRAT2/c-Myc negative feedback loop in gastric cancer.

   Evidence Reciprocal co-immunoprecipitation, siRNA knockdown of FRAT2 in FRAT1-overexpressing gastric cancer cells, in vitro invasion and in vivo metastasis assays, miR-3648 direct target validation

   PMID:36153370

   Open questions at the time

   • Stoichiometry and structural basis of the FRAT1–FRAT2 complex unknown
   • Whether FRAT1–FRAT2 interaction modulates GSK-3β substrate selection or Rac activation not tested

• The structural basis of the FRAT1–FRAT2 heteromeric complex, the mechanism by which FRAT2 differentially channels GSK-3β activity toward primed substrates versus β-catenin, and the physiological requirement for FRAT2 in normal development remain unresolved.
• No structural data for FRAT2 or FRAT1–FRAT2 complex
• No FRAT2-specific knockout phenotype in whole-organism models
• Direct GEF or effector linking FRAT2–GSK3–DVL axis to Rac activation not identified