Figure 1.
Flies mutant for lnk are viable but small.
(A–D) lnk regulates organismal size throughout development. In comparison to ey-flp induced control clones (FRT82, A), FRT82 lnk4Q3 clones result in a small head phenotype (B). lnk4Q3/lnk4Q3 adult flies (C) and pupae (D) are smaller than the controls. (E) Flies lacking lnk function are strongly reduced in dry weight. Introduction of a genomic construct comprising the lnk locus rescues the lnk growth deficit. Significant changes relative to the control (p≤0.01, Student's t-test, n = 20) are marked by double asterisks; error bars represent the standard deviation. (F) Alignment of the Drosophila Lnk protein and its human homologs of the SH2B family of adaptor proteins. lnk codes for a 723 amino acid adaptor protein containing a PH domain and an SH2 domain. Black and grey boxes indicate amino acid identity and similarity, respectively. The SH2 domain is highlighted in red, the PH domain in blue, and the highly conserved Cbl binding motif in orange. Asterisks mark the mutations recovered in the screen, and arrowheads indicate the tyrosines of the two potential YXN Drk SH2 binding motifs and the conserved Cbl binding motif. (G) The mutations leading to an amino acid exchange in the PH domain (7K1) and to premature translational stops (4Q3, 6S2) or an amino acid exchange in the SH2 domain (4H2), respectively, genetically behave as null alleles, indicating that both the PH and the SH2 domain are essential for Lnk's function. Genotypes are: (A) y w ey-flp/y w; FRT82/FRT82 cl(3R3) w+, (B) y w ey-flp/y w; FRT82 lnk4Q3/FRT82 cl(3R3) w+.
Figure 2.
Drosophila lnk regulates cell number and cell size, reminiscent of low IIS activity.
SEM picture of a wild-type Drosophila head (A) compared to the head of a homozygous lnk mutant fly (B). The eye of the lnk mutant is smaller due to fewer and smaller ommatidia. (C) Whereas wild-type eyes consist of more than 700 ommatidia, the number of ommatidia is reduced to about 500 in lnk mutants. (D) Tangential sections through mosaic eyes consisting of homozygous mutant photoreceptors (marked by the absence of pigment) surrounded by heterozygous tissue. Ommatidia containing both wild-type and small homozygous lnk mutant photoreceptor cells (arrowheads) can be observed, pointing to a cell autonomous role of lnk in cell size regulation (wild-type cells are marked by circles and mutant cells by asterisks in a representative ommatidium). (E) The sizes of photoreceptor cells and of rhabdomeres are reduced in lnk mutant ommatidia compared to wild type. (F) lnk mutant females are sterile due to an arrest in oogenesis. Ovaries of homozygous lnk females are small and contain only few oocytes developed to previtellogenic stages (G). The ovarioles of lnk mutant females resemble those of chico mutants (G). (H) The lipid levels of lnk mutant males are strongly elevated compared to wild type, similar to the levels measured in chico mutant flies. Significant changes relative to the respective controls (p≤0.01, Student's t-test) are marked by double asterisks; error bars represent the standard deviation; n = 8 in (C), n = 9 in (E), n = 10 in (H).
Figure 3.
Clones of lnk mutant cells are smaller due to fewer and smaller cells.
(A) Twin-spot clone in the wing imaginal disc. A clone consisting of lnk mutant cells (marked by the absence of GFP) and its wild-type sister clone (marked by two copies of GFP) were induced by mitotic recombination using the hsFlp/FRT system. Nuclei are stained with DAPI (A′) and merged with the GFP signal (A″). Clones of lnk mutant cells consist of fewer cells (B) and cover smaller areas (C) than their corresponding wild-type sister clones, indicating that lnk is required for proper cellular growth during development.
Figure 4.
(A–D) The IIS activity is visualized by the localization of the tGPH reporter. Compared to the signal at the membrane of wild-type fat body cells (A), the signal is diffuse and mostly cytoplasmic in lnk mutant larvae (B), indicative of low PI3K activity. In clones of lnk4Q3 mutant cells (recognizable based on the size reduction and indicated by arrowheads), the signal is also almost absent from the plasma membrane (C). A similar effect is observed in chico mutant cells (D). (E–E″) The strong reduction of membrane-localized tGPH is not due to structural defects of the lnk mutant cells as shown by differential interference contrast (DIC) microscopy (E′–E″). (F) The phosphorylation of PKB is used to monitor IIS activity in larval extracts. Both lnk and chico mutants display a clear reduction of phosphorylated PKB. Note that the levels of PKB do not change. Actin is used as a loading control.
Figure 5.
lnk genetically interacts with components of the IIS pathway.
(A–F) Whereas the loss of lnk function suppresses the overgrowth phenotype caused by eye-specific expression of a constitutive active form of InR (compare E with B), it is not sufficient to suppress the overgrowth phenotype caused by an activated form of PI3K (compare F with C). (G,H) chico and lnk display synthetic lethality. Removing one copy of PTEN in a chico; lnk mutant background is sufficient to restore viability of chico; lnk double mutant flies. Re-introduction of a PTEN genomic rescue construct into this background results in lethality. Error bars represent the standard deviation; n = 20. Genotypes are: (A) GMR-Gal4/+, (B) GMR-Gal4/UAS-InRact, (C) GMR-Gal4/UAS-Dp110CAAX, (D) GMR-Gal4/+; lnk4Q3/lnk6S2, (E) GMR-Gal4/UAS-InRact; lnk4Q3/lnk6S2, (F) GMR-Gal4/UAS-Dp110CAAX; lnk4Q3/lnk6S2, (G) FRT82, lnk4Q3/6S2, chico1/2, Df(2L)Exel6026/+; lnk4Q3/lnk6S2 and chico1/Df(2L)Exel6026; lnk4Q3/lnk6S2.