Figures and data
A mosaic genetic screen identifies GlcT as a tumor suppressor in adult Drosophila midgut.
(A) Diagram showing the lineage hierarchy of intestinal stem cells (ISCs) in the Drosophila intestine. Abbreviations: EB, enteroblast; EEP, enteroendocrine progenitor; EC, enterocyte; EE, enteroendocrine cell. (B to D) MARCM clones (green) induced for 5 days with Pros (red) staining (B), together with quantification of cell number (C) and the percentage of Pros⁺ EE cells per clone (D) in EA30, E230, and GlcT null allele (GlcTΔ8) mutant. (E to F) Overexpression of GlcT in EA30 or E230 mutant clones with Pros staining, along with quantification of the percentage of Pros⁺ EE cells per clone (F). (G to J) esg>GlcT-IR in adult fly gut for 14 days. PH3 staining (G) and Pros staining (I) with quantification of positively stained cells. Error bars represent mean ± SEM, with p-values indicated in the figure (two-tailed Student’s t-test). Scale bars: 25 μm.
Cross scheme for the mosaic genetic screen on the 2R chromosome.
(A) A schematic illustrating the principles of FLPase-induced mitotic recombination for generating homozygous mutant cells in heterozygous animals. (B) The cross scheme and workflow for the forward genetic mosaic screen conducted in this study. Abbreviation: EMS, ethyl methanesulfonate.
Summary of tumor suppression genes identified in the screen.
Genetic analysis of the components of the GSL synthesis pathway reveals specific tumor-suppressive activity for egh, in addition to GlcT.
(A) Diagram illustrating a portion of the glycosphingolipid (GSL) metabolic pathway synthesized from ceramide, highlighting the key enzymes involved at each step. (B to D) GlcTΔ8clones induced for 7 days without or with co-expression of p35 or CDase, stained with anti-DCP-1 or anti-Pros (B). Quantification of the percentage of Pros+ cells in clones of the indicated genotypes (C, D). (E to G) MARCM clones induced in several mutants of key enzymes in the GSL synthesis pathway. Pros staining in egh mutant clones (eghA, eghB, egh7), brn228, β4GalNAcTA-IR, and α4GT1-IR clones (E). Quantification of cell number (F) and the percentage of Pros+ cells (G) in clones of the indicated genotypes. Error bars represent mean ± SEM, with p-values indicated (two-tailed Student’s t-test). Scale bars: 25 μm.
The EE tumor phenotype is a result of MacCer deficiency.
(A to C) LacCer feeding after MARCM clone induction. Anti-Pros staining of anterior midgut (AM) and posterior midgut (PM) in GlcTΔ8 clones (A). Quantification of cell number (B) and the percentage of Pros+ cells (C) in clones of the indicated genotypes. Error bars represent mean ± SEM, with p-values indicated (two-tailed Student’s t-test). Scale bars: 25 μm.
Loss of GlcT leads to reduced activation of Notch signaling in ISC progenies.
(A and B) Expression of NRE-lacZ, the Notch activity reporter, in GlcTΔ8clones (A) and quantification of LacZ fluorescence intensity for signal positive cell (B). (C) Tk and AstC staining in GlcTΔ8 clones in the posterior (R5) region of the Drosophila midgut. (D) Pros staining in Nintra-overexpressing GlcTΔ8 clones. Error bars represent mean ± SEM, with p-values indicated (two-tailed Student’s t-test). Scale bars: 25 μm.
The EGFR/Ras/MAPK and JAK/STAT signaling activities in GlcT mutant clones.
GlcT mutant clones stained with pERK (red, upper panels), a direct indicator of EGFR/Ras/MAPK signaling activity, and STAT92E (red, lower panels), whose nuclear localization is indicative of JAK/STAT signaling activity. Note that pERK was not significantly increased in GlcT mutant cells compared to normal cells outside of the clones, and STAT92E activity was also not obviously altered in GlcT mutant cells compared to normal cells outside the clones. Scale bar: 25 μm.
GlcT regulates the endocytic trafficking of Delta.
(A and B) Pros staining in flies carrying GlcT-IR driven by Dltsand NREts respectively (A), and quantification of the percentage of Pros+ cells (B). (C and D) Dl staining in an antibody uptake assay performed in GlcTΔ8 flies after 3-hour clone induction (C) and quantification of Dl fluorescence intensity inside and outside of the clones (D). (E and F) Dl staining in esg>GlcT-IR flies co-stained with Rab5-GFP, Rab7-GFP or Rab11-GFP (E); yellow arrows indicate co-localization of Dl and GFP. Quantification of Dl and GFP co-localization (F). Error bars represent mean ± SEM, with p-values indicated (two-tailed Student’s t-test). Scale bars: 25 μm (10 μm in E).
GlcT shows tissue specificity in regulating Notch signaling activity.
(A) Cut staining in wing disc of GlcTΔ8 clones. (B) EYA staining in follicle cell clones (non-green areas) of GlcTΔ8 flies. (C) Dl staining in germ cell clones (non-green area) of GlcTΔ8 mutant clone, yellow arrows indicate punctate Dl accumulation. Scale bars: 25 μm.
Transient loss of UGCG in mouse small intestine causes reduced number of ISCs and increased number of goblet cells.
(A and B) Olfm4 staining in the duodenum and ileum of VillinCreERT2; UGCGflox/flox mice 48 hours after tamoxifen injection (n = 3, A) and quantification of Olfm4 fluorescence intensity (B). (C and D) Muc2 staining of goblet cells in the duodenum and ileum of UGCG-CKO mice (C) and quantification of Muc2+ goblet cells per villus (D). (E) Schematic model illustrating the role of MacCer in regulating the Notch signaling pathway. Error bars represent mean ± SEM, with p-values indicated (two-tailed Student’s t-test). Scale bars: 100 μm.
The Crypt-Villus morphology in UGCG CKO mice. The duodenal structure exhibits significant abnormalities in CKO mice.
Small intestine (duodenum) from control and villin-creERT2, UGCGflox/flox mice stained with K20, which marks enterocytes and goblet cells, and Ki67, which marks proliferative cells. Note that the mutant intestine has shortened villi and elongated crypts. Dotted lines indicate the boundary between the crypt and the villus. Samples were collected at 48h after tamoxifen injection. Scale bar: 100 μm.
