The bantam microRNA acts through Numb to exert cell growth control and feedback regulation of Notch in tumor-forming stem cells in the Drosophila brain

Abstract

Notch (N) signaling is central to the self-renewal of neural stem cells (NSCs) and other tissue stem cells. Its deregulation compromises tissue homeostasis and contributes to tumorigenesis and other diseases. How N regulates stem cell behavior in health and disease is not well understood. Here we show that N regulates bantam (ban) microRNA to impact cell growth, a process key to NSC maintenance and particularly relied upon by tumor-forming cancer stem cells. Notch signaling directly regulates ban expression at the transcriptional level, and ban in turn feedback regulates N activity through negative regulation of the Notch inhibitor Numb. This feedback regulatory mechanism helps maintain the robustness of N signaling activity and NSC fate. Moreover, we show that a Numb-Myc axis mediates the effects of ban on nucleolar and cellular growth independently or downstream of N. Our results highlight intricate transcriptional as well as translational control mechanisms and feedback regulation in the N signaling network, with important implications for NSC biology and cancer biology.

Fig 1. The growth regulator ban is preferentially required for CSC-like NB proliferation and maintenance.

(A) Clonal analysis of NBs overexpressing N^act in control, dicer-1^Q1948X, or ban^Δ1 backgrounds. Larval brains of different genotypes at 96 h ALH were immunostained for Dpn and Pros. Type II NB MARCM clones are marked with GFP and outlined with white dashed lines. (B) Quantification of NB number in samples from A. *, p<0.001 in Student’s t-test; n = 4–10 clones. (C) Effect of ban inhibition by ban-sp on N-induced NB overproliferation. Larval brains were stained for F-actin (Green, cell cortex), Dpn (red, NBs and mature IPs), and Pros (GMCs and neurons). The central brain area is outlined with a bold white dashed line, and the Dpn⁺ NBs within this area were quantified. (D) Quantification of data from C. **, p<0.0001; n = 8–10 brains. (E) Effects of ban inhibition by ban-sp on Dpn OE-induced type II lineage NB overproliferation. (F) Quantification of number of NBs in E. **, p<0.00001; n = 8 brains. Scale bars: A, 20 μm; C, E, 50 μm.

Fig 2. Regulation of nucleolar growth by ban and Numb in type II NB lineages.

(A) Genetic interaction between ban and myc in nucleolar size regulation. Green, Fibrillarin; Red, Miranda; Bracket, NB. Arrowheads: nucleoli of immature IPs. (B) Quantification of effect on nucleolar size by ban GOF (ban-D OE) and ban LOF (ban^Δ1) in type II NBs from A. *, p< 0.005; NS, not significant; n = 10–12 type II NB lineages/genotype. (C) Quantification of nucleolar size of immature IPs in type II NB lineages from A. *, p< 0.001; **, P< 0.00001; n = 9–12 type II NB lineages/genotype. (D) Genetic interactions between Numb and ban and between Numb and Myc in nucleolar growth regulation. (E) Quantification of nucleolar size of immature IPs in type II NB lineages from D. *, p<0.001, **, p<0.00001; n = 8 type II NB lineages /genotype. Scale bars: A, D, 5 μm.

Fig 3. Regulation of ban expression and activity by N signaling and feedback regulation of N activity by ban.

(A) Effects of N OE and N RNAi (N-IR) on ban expression as monitored with the ban-lacZ transcriptional reporter in type II NB lineages. N-V5 or N-IR transgene induction was carried out using a 1407^ts system (1407-GAL4: tub-GAL80^ts). Larva were shifted to 29°C at 24 hr ALH, and analyzed at late third instars. Green, F-actin; Red, LacZ; Blue, Dpn; Bracket, NB; Arrowheads, mature IPs. Note that N-IR brain has no type II NB lineages. (B) Effects of N OE and N-IR on ban activity as measured with ban GFP sensor expression. Green, GFP; Red, Dpn; Blue, Pros; Bracket, NBs; Open arrowhead, type I NB; Closed arrowhead and dotted outline, type II NB and its lineage; Yellow arrowhead, mature IP. (C) Quantification of ban-LacZ expression shown in A. n = 10–15 type I or II NBs from posterior brain /genotype (left), and n = 16–22 mature IPs/genotype (right). **, p<0.0005. (D) Quantification of ban GFP sensor expression. n = 5 type I or II NBs from posterior brain/genotype (left), and n = 5 mature IPs /genotype (right). *, p<0.005; **, p<0.0005. (E) Quantitative RT-PCR analysis of ban levels in third instar larval brains. ban levels were normalized to 2S rRNA. *, p<0.005; mean ± SEM, n = 3 repeats. (F) ChIP analysis of Su(H) binding to genomic DNA in ban locus. (Left) Schematic drawing of the ban locus. The position of pre-miRNA sequence in yellow box is between 993–1073 nucleotides (counting from transcription start site at position 1). Sequences containing two putative Su(H)-binding sites (S1 and S2) that matched the consensus sequence RTGRGAA, and a control sequence that does not contain Su(H)-binding site (ban con) in the upstream regulatory region of pre-bantam are indicated. (Right) Histogram showing enrichment of ban genomic sequence surrounding S1, S2, but not the ban con region, in the Su(H) ChIP. E(spl)m8 and rp49 are positive and negative controls, respectively. ***, p<0.0001; *, p<0.005, n = 3 repeats. (G, H) Effects of ban LOF and GOF on E(spl)mγ-GFP reporter expression in type I NBs located in posterior brain region. H, quantification of GFP fluorescence intensity from G. *, p<0.005; Scale bars: A, B, 10 μm; G, 20 μm.

Fig 4. Numb is a target of ban and it negatively regulates Myc.

(A) Translational reporter assay showing targeting of numb 3´UTR by ban miRNA. Luciferase activity derived from firefly luciferase (FL)-ban sensor, renila luciferase (RL)-numb 3´UTR, or RL-numb 3´UTR^mut reporters in response to ban-5P miRNA co-expression was normalized with values from let-7A control miRNA co-expression. *, p<0.05; **, p<0.001. n = 3 repeats. The gel images under the bar graph represent measurements of luciferase mRNA expression by RT-PCR, with actin serving as a control. (B) Western blot analysis of larval brain extracts showing effects of ban LOF or GOF on Numb protein levels. Actin serves as loading control. Bar graph shows quantification of normalized Numb signals from three independent blots. *, p<0.001. n = 3 repeats. (C) Co-IP between Numb and c-Myc in HEK293T cells. GFP-mNumb or GFP vector was transfected into HEK293T cells. Anti-GFP immunoprecipitates were probed for GFP and c-Myc by western blot. (D) Western blot analysis of c-Myc levels in various cellular fractions from HEK293T cells in response to Numb-myc (myc epitope tagged Numb) expression. The anti-c-Myc (9E10) was used to detect both c-Myc and Numb-myc. c-Myc levels were quantified after normalization to Actin. Total: total lysates (#1, #2); Cyto: cytosol fractions (#3, #4); Nu: nuclear fractions (#5, #6). Effect of Numb on endogenous cMyc level was verified with another anti-c-Myc Ab (Y69). (E) Quantification of c-Myc levels from (D) after normalization to Actin. c-Myc levels detected by c-Myc (9E10) were quantified in comparison to nuclear lysates in #5. *, p<0.05, **, p<0.005; n = 3 independent experiments. (F) Effect of the proteasome inhibitor MG132 on endogenous c-Myc level in Numb transfected HEK293 cells. HEK293T cells with or without Numb-myc transfection were incubated in the presence or absence of 10 μM MG132 for 2 hrs. Cell lysates (#1-#12) were analyzed by western blot as indicated. (S): Short exposure; (L): Long exposure. (G) Quantification of c-Myc levels upon MG132 treatment shown in F. *, p<0.05; **, p<0.001; ***, p<0.0001 vs. without MG132 treatment. n = 3 independent experiments. (H, I) Genetic interaction between Numb-TS4D and ban in regulating NB homeostasis. Type II NB lineages co-expressing Numb-TS4D and ban-sp or Numb-TS4D and ban-D are marked with white dashed lines. Asterisks indicate ectopic NBs in the clones. J, quantification of NB number in type II NB clones from I. **, p<0.00001; n = 8–10 NB clones. Scale bars: H, 20 μm.

Fig 5. Numb mediates the effects of ban in the feedback regulation of N activity and CSC-like NB proliferation.

(A) Genetic interaction between ban and numb on the expression of N activity reporter E(spl)m g-GFP. The effect of ban-sp (II) OE driven by 1407-Gal4 on N activity was rescued by numb-RNAi. (B) Quantification of data from A. *, p<0.05 vs. 1407>ban-sp (II); n = 10–14 type I NBs/genotype. (C) The effect of ban-sp in blocking N-induced brain tumor formation is sensitive to the gene dosage of numb but not pros or brat. N-V5 (II) and N-V5 (III) stand for N-V5 transgenes on the 2nd or 3rd chromosomes, respectively. (D) Quantification of data from C. **, p<0.00001 comparing 1407>ban-sp (III) with and without numb-IR in N-V5 background; *, p<0.0005 comparing 1407>ban-sp (III) with or without removing one copy of numb, in N-V5 (II) or N-V5 (III) OE background. n = 5–10 brain samples. (E) Myc OE, but not CycE OE, rescued ban-sp effect in blocking N-induced brain tumor formation. Data is quantified using images from C. *, p<0.00001 comparing 1407>N-V5; ban-sp without Tg expression (w^-) vs. with Tg expression (Myc or CycE). Scale bars: A, 20 μm; C, 50 μm.

Fig 6. A diagram depicting the regulatory network involving N, ban, Numb, and Myc in cell fate determination in NSC/CSC lineages.

In stem cells, positive transcriptional regulation of ban by N and a feedback regulation of N by ban help maintain high N activity and stem cell fate. A key aspect of stem cell maintenance is nucleolar growth promoted by Myc. Myc is a known transcriptional target of N and is independently regulated at the protein stability level by Numb through Huwe1 as shown in this study. As differentiation proceeds, the feedback regulatory loop is weakened, presumably contributed by the asymmetric segregation of Numb, resulting in gradual decline of N and ban activities and corresponding increase of Numb activity in differentiated progenies.