Psi promotes Drosophila wing growth via direct transcriptional activation of cell cycle targets and repression of growth inhibitors

Abstract

The first characterised FUSE Binding Protein family member, FUBP1, binds single-stranded DNA to activate MYC transcription. Psi, the sole FUBP protein in Drosophila, binds RNA to regulate P-element and mRNA splicing. Our previous work revealed pro-growth functions for Psi, which depend, in part, on transcriptional activation of Myc. Genome-wide functions for FUBP family proteins in transcriptional control remain obscure. Here, through the first genome-wide binding and expression profiles obtained for a FUBP family protein, we demonstrate that, in addition to being required to activate Myc to promote cell growth, Psi also directly binds and activates stg to couple growth and cell division. Thus, Psi knockdown results in reduced cell division in the wing imaginal disc. In addition to activating these pro-proliferative targets, Psi directly represses transcription of the growth inhibitor tolkin (tok, a metallopeptidase implicated in TGFβ signalling). We further demonstrate tok overexpression inhibits proliferation, while tok loss of function increases mitosis alone and suppresses impaired cell division caused by Psi knockdown. Thus, Psi orchestrates growth through concurrent transcriptional activation of the pro-proliferative genes Myc and stg, in combination with repression of the growth inhibitor tok.

Keywords: Cell cycle; Drosophila; FUBP1; Myc; Psi; Transcription.

Fig. 1.

Psi depletion reduces proliferation. (A) Third instar larval wing discs with ser-GAL4 expression of UAS-FUCCI with two alternate Psi RNAi lines, or control, stained using anti-pH3 antibody. (B) Quantification of the proportion of cells undergoing each cell cycle stage (n>10), and the total number of mitotic cells. NS, no significance; ****P<0.0001 when compared with control (t-test). Each data point represents a single wing disc. Data are mean±s.d.

Fig. 2.

Psi binds multiple genomic regions, including Myc. (A) Psi- and RNA Pol II-binding profiles across the Myc gene in larval wing discs (sd-GAL4 driver used for targeted Psi-DamID and RNA Pol-DamID), shown as log₂ of the ratio to Dam-only control. (B) Average Pol and Psi binding across genic regions and heatmap of DamID signal, clustered by k-means into three clusters using Pol signal. Ontology networks containing common genes and functions are highlighted and labelled manually with general terms. (C) Intersection of Pol and Psi genes with significant enrichment (FDR 1%). Proportions of coding and non-coding genes are shown for each subset, and non-coding genes are further classified by type.

Fig. 3.

Significantly altered genes after Psi knockdown in wing discs. (A) Genes with statistically significant altered expression after Psi knockdown at FDR<0.01. Top 50 genes with greatest fold change and smallest P-value are labelled. Myc and stg are highlighted with black squares. (B) Proportion of differential splicing events detected by rMATS at FDR<0.01: alternative 3′ splice site (A3′SS), alternative 5′ splice site (A5′SS), skipped exon (SE), retained intron (RI) and mutually exclusive exons (MXE). (C) Intersection of differentially expressed genes with genes exhibiting altered splicing.

Fig. 4.

Psi binds and regulates developmental genes. (A) Intersection of differentially expressed genes after Psi knockdown with genes bound by Psi. (B) MA plot showing only genes bound by Psi (blue) while statistically significant DGE events at FDR<0.01 are shown in red. (C) Fold change and expression of ncRNA Psi targets. (D) Ontology of mutually inclusive genes from the intersection in A. (E) Genes regulated by Psi with annotated roles in wing morphogenesis; log₂(fold change) values are as indicated.

Fig. 5.

Stg overexpression rescues impaired growth associated with Psi depletion. (A) Psi and RNA Pol II binding profiles across the stg gene in larval wing discs (log₂ of the ratio to Dam-only control). (B) Adult wings with ser-GAL4 driven knockdown of Psi alone or in combination with stg overexpression (OE). (C) Quantification of the posterior compartment of the adult wing defined by the L5 vein, marked with a red outline in B. ****P_adj<0.0001 (corrected for multiple testing using the Benjamini-Hochberg FDR method). (D) Third instar larval wing discs with ser-GAL4 expression of UAS-FUCCI with genotypes as marked, stained using anti-pH3 antibody. Quantification of the proportion of cells undergoing each cell cycle stage (n>6) and the total number of mitotic cells. ****P<0.0001 compared with control, **P<0.01 (t-test). Each data point represents a single adult wing or wing disc. Data are mean±s.d.

Fig. 6.

Repression of tok is required for Psi-dependent cell division. (A) Larval wing discs with ser-GAL4-driven tok overexpression (OE) and/or Psi RNAi, or tok¹/tok³ transheterozygotes with or without ser-GAL4 driven Psi RNAi KD stained using anti-pH3 to detect mitosis. (B) Quantification of mitotic cells in the ser-GAL4 compartment, ****P<0.0001. Each data point represents a single wing disc. Data are mean±s.d. (C) Model for Psi function in regulating proliferative growth during development.