Frat is dispensable for canonical Wnt signaling in mammals

Abstract

Wnt-signal transduction through beta-catenin is thought to require the inhibition of GSK3 by Frat/GBP. To investigate the role of Frat in mammalian development, we have generated mice with targeted mutations in all three murine Frat homologs. We show that Frat is normally expressed at sites of active Wnt signaling. Surprisingly, Frat-deficient mice do not display gross abnormalities. Moreover, canonical Wnt signaling in primary cells is unaffected by the loss of Frat. These studies show that Frat is not an essential component of the canonical Wnt pathway in higher organisms, despite the strict requirement of Frat/GBP for maternal Wnt signaling in Xenopus.

Figure 4.

Model for the physiological role of Frat. (A) Current model of canonical Wnt-signal transduction. See text for details. (B) Several models could explain the observation that Frat is not critically required for canonical Wnt signaling in mammals. (Left) Frat might be induced under specific conditions or it might feed into the canonical Wnt pathway from a parallel route independent from upstream Wnt signals. (Right) Alternatively, Frat might be involved in any of the numerous other GSK3-dependent cellular activities, such as the NFκB and NFAT signaling pathways.

Figure 1.

Frat1 and Frat2 show an identical expression pattern in neural and epithelial tissues of developing and adult mice. Comparison of lacZ activity in Frat1^+/lacZ (A–D) and Frat2^+/lacZ (E–H) mice reveals overlapping expression between the two homologs. (A,E) Embryonic day 8.5 (E8.5) embryo. (B,F) E10.5 embryo. (C,G) Adult hippocampus. (D,H) Adult skin.

Figure 2.

Loss of Frat does not affect the intestinal Wnt gradient. (A–D) LacZ staining on duodenum (A,B) and colon (C,D) shows that Frat2 expression is restricted to intestinal crypt cells. (A,C) Frat2^+/lacZ. (B,D) Littermate control. (E–P) The expression of different Wnt-related markers is unchanged in Frat-TKO intestine. (E–J) Frat-TKO. (K–P) Littermate control. (E,K) β-Catenin, 40×. F and L are close-ups of crypts depicted in E and K, respectively. (G,M) Ki67, 40×.(H,N) EphrinB1, 40×.(I,O) EphB2, 40×. J and P are close-ups of crypts depicted in I and O, respectively.

Figure 3.

Sensitivity to Wnt ligand and the β-catenin/TCF response are unimpaired in Frat-TKO cells. TKO and littermate control MEFs were stimulated with control L-cell (lanes labeled C) or Wnt3A-CM (Wnt3A). Activation of canonical Wnt signaling was determined by Western blot analysis of unphosphorylated β-catenin (A), TOPFLASH luciferase reporter activity (shown are the averages of two experiments on independent MEF isolates, each performed in triplicate) (B), and immunofluorescence to visualize the nuclear accumulation of β-catenin following titration of different dilutions of Wnt3A-CM (ranging from 1:100 to 1:2) (C).