Monomeric Yan has reduced activity in repressing reporter gene expression in S2 cells. Transcription assays were performed with wild-type Yan, ML mutant (A86D) Yan, EH mutant (V105R or V105E) Yan, or the combination of both (A86D+V105R; A86D+V105E) (V105R+V105E served as a control). Error bars indicate standard deviations between the results of two transfections. Results are shown for mae-luciferase (A), eve-luciferase (B), and argos-luciferase (C).

Monomeric Yan fails to repress the expression of target genes argos and even skipped in embryos. (A to D) In situ hybridization of argos in wild-type embryos (A) and embryos expressing Yan^Act (B), two copies of Yan^{Act, V105R} (C), or Yan^{Act, A86D} + Yan^{Act, V105R} (D) under the control of the rhomboid-Gal4 driver. (E to H) Eve staining in wild-type embryos (E) and embryos expressing Yan^Act (F), two copies of Yan^{Act, V105R} (G), or Yan^{Act, A86D} + Yan^{Act, V105R} (H) under the control of the twist-Gal4 driver. Yan^Act and Yan^{Act, A86D} + Yan^{Act, V105R} strongly repress the expression of argos in the ventral epidermis (arrows) and Eve in the mesoderm (dashed circles), whereas Yan^{Act, V105R} has minimal effects.

Yan forms a large complex in living cells. Recovery time measurements by FRAP of Yan-mEGFP in Drosophila S2 cells. Data points in gray are for individual cells. Black points are averages of between 49 and 101 cells per condition. The mean ± standard error of the mean are reported above each black point. (A) Recovery time of wild-type, A86D, V105R, and coexpressed A86D and V105R Yan-mEGFP and the effect of Mae on Yan's diffusion rate. (B) The effect of Ras^V12 on the diffusion rates of Yan and Yan^Act.

Rescue of yan null mutants. (A) Wild-type Yan, ML mutant (A86D) Yan, EH mutant (V105R or V105E) Yan, or the combination of both (A86D+V105R; A86D+V105E) was expressed under the control of the armadillo-Gal4 driver in yan^ER443/yan^E833 null mutant embryos, and the percentages of animals that developed to the larval and pupal stages were calculated (see Materials and Methods). Pupal survival for A86D was not quantified because of the low rate of larval hatching. N, number of animals scored in survival rate calculation. (B) Quantification of Yan protein levels by Western blotting. Numbers below the blots are relative Yan/α-tubulin ratios. The ratio for WT is set to 1.

Monomeric Yan fails to inhibit eye development and repress target genes prospero and D-Pax2 in third instar eye imaginal discs. (A to D) Pictures of eyes of adult wild-type flies (A) and flies expressing Yan^Act (B), two copies of Yan^{Act, V105R} (C), or Yan^{Act, A86D} + Yan^{Act, V105R} (D) under the lozenge-Gal4 driver. (E to L) Third instar wild-type eye discs (E and I) and discs expressing Yan^Act (F and J), two copies of Yan^{Act, V105R} (G and K), or Yan^{Act, A86D} + Yan^{Act, V105R} (H and L) under the GMR-Gal4 driver. (E to H) Prospero expression. (I to L) β-Galactosidase staining revealing the expression of the D-Pax2 gene. (M) Quantification of Yan protein levels by immunostaining (see Materials and Methods). Error bars represent standard deviations. N, number of discs measured for each genotype. (N) Quantification of Yan protein levels by Western blotting. Numbers below the blots are relative Yan/α-tubulin ratios. The ratio for Yan^Act is set to 1.

ML and EH surfaces mediate self-association of Yan. (A) Regulation of Yan and Pnt by RTK signaling. In the absence of RTK signaling, Yan represses target gene expression. Activation of RTK signaling leads to Yan's phosphorylation by MAPK and nuclear export facilitated by Mae. In parallel, MAPK phosphorylates and activates Pnt, which then induces the expression of target genes previously repressed by Yan. (B) Model of self-association between ML and EH surfaces of the Yan-SAM domain. ML (A86D) and EH (V105R) mutants exist as monomers, while they interact with each other to form dimers. Mae-SAM interacts with the EH but not the ML surface of Yan-SAM. (C and D) Different combinations of HA- and Flag-tagged wild-type (WT), ML mutant (A86D), EH mutant (V105R; V105E), and ML, EH double mutant (A86D, V105R; A86D, V105E) Yan were cotransfected into Drosophila S2 cells. Immunoprecipitation (IP) was performed using anti-Flag agarose, and the precipitated proteins were detected by immunoblotting with anti-Flag or anti-HA antibodies. (C) Co-IP between wild-type HA-Yan and different mutant forms of Flag-Yan. (D) Co-IP between mutant forms of Yan. IB, Western blotting; input, 2% of the lysate. Numbers below the blots are IP/input ratios.

Yan self-association functions downstream of nuclear localization and phosphorylation. (A) Localization of the Yan^WT, Yan^A86D, Yan^V105R, and Yan “dimer” made by cotransfecting Yan^A86D and Yan^V105R. (B) Localization of Yan^Act, Yan^{Act, A86D}, Yan^{Act, V105R}, and Yan^{Act, A86D} + Yan^{Act, V105R}. Subcellular localization of Yan was scored by eye with reference to nuclear staining by DAPI. Localization was judged nuclear if anti-Yan staining was clearly brighter in the region of the nucleus than in the surrounding cytoplasm. Conversely, localization was judged cytoplasmic if the staining intensity in the nuclear region appeared to be less than that in the surrounding cytoplasm. Cells were classified as having both nuclear and cytoplasmic staining if the relative intensities of the signals in the two regions appeared comparable. Scoring was performed without knowledge of the identity of the samples. Yan is expressed endogenously in S2 cells but at low levels that are essentially undetectable relative to the proteins expressed from our transfected constructs. N, number of cells scored. (C to H) Immunostaining of Yan (red) and dRAQ5 (blue), which stains DNA, in third instar wing discs expressing Yan^WT (C), Yan^V105R (D), Yan^A86D + Yan^V105R (E), Yan^Act (F), Yan^{Act, V105R} (G), and Yan^{Act, A86D} + Yan^{Act, V105R} (H) under the dpp-Gal4 driver. (I) Repression of the PntP1-activated eve-luciferase reporter in Drosophila S2 cells by various Yan constructs. The data have been normalized such that the fully activated reporter (with PntP1) is set to 1. All Yan constructs with a wild-type SAM domain (white bars) repress the reporter approximately fivefold, while all monomeric constructs (gray bars) repress the reporter approximately twofold. Even when the protein is rendered nonphosphorylatable (Act), restricted to the nucleus (IntNLS), or both (Act, IntNLS), the monomeric mutants are still less active than wild-type Yan.